![Datastores
•
Pushdown_sql function
•
Pre-load commands in table loader
•
Post-load commands in table loader
Note:
For any table in Data Services, the maximum size of the owner name is 64 characters. In the case of
Attunity tables, the maximum size of the Attunity data source name and actual owner name is 63 (the
colon accounts for 1 character). Data Services cannot access a table with an owner name larger than
64 characters.
5.2.1.5 Limitations
All Data Services features are available when you use an Attunity Connector datastore except the
following:
•
Bulk loading
•
Imported functions (imports metadata for tables only)
•
Template tables (creating tables)
•
The datetime data type supports up to 2 sub-seconds only
•
Data Services cannot load timestamp data into a timestamp column in a table because Attunity
truncates varchar data to 8 characters, which is not enough to correctly represent a timestamp value.
•
When running a job on UNIX, the job could fail with following error:
[D000] Cannot open file /usr1/attun/navroot/def/sys System error 13: The file access permissions do not
allow the specified action.; (OPEN)
This error occurs because of insufficient file permissions to some of the files in the Attunity installation
directory. To avoid this error, change the file permissions for all files in the Attunity directory to 777
by executing the following command from the Attunity installation directory:
$ chmod -R 777 *
5.2.2 Defining a database datastore
Define at least one database datastore for each database or mainframe file system with which you are
exchanging data.
To define a datastore, get appropriate access privileges to the database or file system that the datastore
describes.
77
2011-06-09](https://image.slidesharecdn.com/sbo401dsdesigneren-131213054333-phpapp02/85/SAP-BODS-Designer-PDF-77-320.jpg)
![Datastores
5.2.7.5 Memory table target options
The Delete data from table before loading option is available for memory table targets. The default
is on (the box is selected). To set this option, open the memory target table editor. If you deselect this
option, new data will append to the existing table data.
5.2.7.6 Create Row ID option
If the Create Row ID is checked in the Create Memory Table window, the software generates an integer
column called DI_Row_ID in which the first row inserted gets a value of 1, the second row inserted
gets a value of 2, etc. This new column allows you to use a LOOKUP_EXT expression as an iterator
in a script.
Note:
The same functionality is available for other datastore types using the SQL function.
Use the DI_Row_ID column to iterate through a table using a lookup_ext function in a script. For
example:
$NumOfRows = total_rows (memory_DS..table1)
$I = 1;
$count=0
while ($count < $NumOfRows)
begin
$data =
lookup_ext([memory_DS..table1, 'NO_CACHE','MAX'],[A],[O],[DI_Row_ID,'=',$I]);
$1 = $I + 1;
if ($data != NULL)
begin
$count = $count + 1;
end
end
In the preceding script, table1 is a memory table. The table's name is preceded by its datastore name
(memory_DS), a dot, a blank space (where a table owner would be for a regular table), then a second
dot. There are no owners for memory datastores, so tables are identified by just the datastore name
and the table name as shown.
Select the LOOKUP_EXT function arguments (line 7) from the function editor when you define a
LOOKUP_EXT function.
The TOTAL_ROWS(DatastoreName.Owner.TableName) function returns the number of rows in a
particular table in a datastore. This function can be used with any type of datastore. If used with a
memory datastore, use the following syntax: TOTAL_ROWS( DatastoreName..TableName )
93
2011-06-09](https://image.slidesharecdn.com/sbo401dsdesigneren-131213054333-phpapp02/85/SAP-BODS-Designer-PDF-93-320.jpg)
![Datastores
2. Click OK in the confirmation window.
The software removes the datastore and all metadata objects contained within that datastore from
the metadata repository.
If these objects exist in established flows, they appear with a deleted icon
.
5.3.2 Browsing metadata through an adapter datastore
The metadata you can browse depends on the specific adapter.
5.3.2.1 To browse application metadata
1. Right-click the datastore you want to browse and select Open.
A window opens showing source metadata.
2. Scroll to view metadata name and description attributes.
3. Click plus signs [+] to expand objects and view subordinate objects.
4. Right-click any object to check importability.
5.3.3 Importing metadata through an adapter datastore
The metadata you can import depends on the specific adapter. After importing metadata, you can edit
it. Your edits propagate to all objects that call these objects.
5.3.3.1 To import application metadata while browsing
1. Right-click the datastore you want to browse, then select Open.
2. Find the metadata object you want to import from the browsable list.
3. Right-click the object and select Import.
4. The object is imported into one of the adapter datastore containers (documents, functions, tables,
outbound messages, or message functions).
102
2011-06-09](https://image.slidesharecdn.com/sbo401dsdesigneren-131213054333-phpapp02/85/SAP-BODS-Designer-PDF-102-320.jpg)
![File formats
•
User names
•
Timestamps
•
Requested action (might include search string)
•
Bytes transferred
•
Referred address
•
Cookie ID
Web logs use a common file format and an extended common file format.
Common Web log format:
151.99.190.27 - - [01/Jan/1997:13:06:51 -0600]
"GET /~bacuslab HTTP/1.0" 301 -4
Extended common Web log format:
saturn5.cun.com - - [25/JUN/1998:11:19:58 -0500]
"GET /wew/js/mouseover.html HTTP/1.0" 200 1936
"http://av.yahoo.com/bin/query?p=mouse+over+javascript+source+code&hc=0"
"Mozilla/4.02 [en] (x11; U; SunOS 5.6 sun4m)"
The software supports both common and extended common Web log formats as sources. The file
format editor also supports the following:
•
•
Dash as NULL indicator
Time zone in date-time, e.g. 01/Jan/1997:13:06:51 –0600
The software includes several functions for processing Web log data:
•
•
•
Word_ext function
Concat_data_time function
WL_GetKeyValue function
Related Topics
• Word_ext function
• Concat_date_time function
• WL_GetKeyValue function
6.9.1 Word_ext function
The word_ext is a string function that extends the word function by returning the word identified by
its position in a delimited string. This function is useful for parsing URLs or file names.
Format
word_ext(string, word_number, separator(s))
A negative word number means count from right to left
148
2011-06-09](https://image.slidesharecdn.com/sbo401dsdesigneren-131213054333-phpapp02/85/SAP-BODS-Designer-PDF-148-320.jpg)
![Work Flows
5. Enter the script statements, each followed by a semicolon.
The following example shows a script that determines the start time from the output of a custom
function.
AW_StartJob ('NORMAL','DELTA', $G_STIME,$GETIME);
$GETIME =to_date(
sql('ODS_DS','SELECT to_char(MAX(LAST_UPDATE) ,
'YYYY-MM-DDD HH24:MI:SS')
FROM EMPLOYEE'),
'YYYY_MMM_DDD_HH24:MI:SS');
Click the function button to include functions in your script.
6. After you complete the script, select Validation > Validate.
The software tests your script for syntax errors and displays any errors encountered.
7. Click the ... button and then save to name and save your script.
The script is saved by default in <LINKDIR>/BusinessObjects Data Services/ DataQuality/Samples.
9.9.2 Debugging scripts using the print function
The software has a debugging feature that allows you to print:
•
The values of variables and parameters during execution
•
The execution path followed within a script
You can use the print function to write the values of parameters and variables in a work flow to the trace
log. For example, this line in a script:
print('The value of parameter $x: [$x]');
produces the following output in the trace log:
The following output is being printed via the Print function in <Session job_name>.
The value of parameter $x: value
Related Topics
• Reference Guide: Functions and Procedures, print
215
2011-06-09](https://image.slidesharecdn.com/sbo401dsdesigneren-131213054333-phpapp02/85/SAP-BODS-Designer-PDF-215-320.jpg)
![Variables and Parameters
13.8.1.2 Using substitution parameters
You can use substitution parameters in expressions, SQL statements, option fields, and constant strings.
For example, many options and expression editors include a drop-down menu that displays a list of all
the available substitution parameters.
The software installs some default substitution parameters that are used by some Data Quality transforms.
For example, the USA Regulatory Address Cleanse transform uses the following built-in substitution
parameters:
•
•
$$RefFilesAddressCleanse defines the location of the address cleanse directories.
$$ReportsAddressCleanse (set to Yes or No) enables data collection for creating reports with address
cleanse statistics. This substitution parameter provides one location where you can enable or disable
that option for all jobs in the repository.
Other examples of where you can use substitution parameters include:
• In a script, for example:
Print('Data read in : [$$FilePath]'); or Print('[$$FilePath]');
•
In a file format, for example with [$$FilePath]/file.txt as the file name
13.8.2 Using the Substitution Parameter Editor
Open the Substitution Parameter Editor from the Designer by selecting Tools > Tools Substitution
Parameter Configurations. Use the Substitution Parameter editor to do the following tasks:
•
•
•
•
•
•
•
293
Add and define a substitution parameter by adding a new row in the editor.
For each substitution parameter, use right-click menus and keyboard shortcuts to Cut, Copy, Paste,
Delete, and Insert parameters.
Change the order of substitution parameters by dragging rows or using the Cut, Copy, Paste, and
Insert commands.
Add a substitution parameter configuration by clicking the Create New Substitution Parameter
Configuration icon in the toolbar.
Duplicate an existing substitution parameter configuration by clicking the Create Duplicate
Substitution Parameter Configuration icon.
Rename a substitution parameter configuration by clicking the Rename Substitution Parameter
Configuration icon.
Delete a substitution parameter configuration by clicking the Delete Substitution Parameter
Configuration icon.
2011-06-09](https://image.slidesharecdn.com/sbo401dsdesigneren-131213054333-phpapp02/85/SAP-BODS-Designer-PDF-293-320.jpg)
![Variables and Parameters
•
•
•
Reorder the display of configurations by clicking the Sort Configuration Names in Ascending
Order and Sort Configuration Names in Descending Order icons.
Move the default configuration so it displays next to the list of substitution parameters by clicking
the Move Default Configuration To Front icon.
Change the default configuration.
Related Topics
• Adding and defining substitution parameters
13.8.2.1 Naming substitution parameters
When you name and define substitution parameters, use the following rules:
•
•
The name prefix is two dollar signs $$ (global variables are prefixed with one dollar sign). When
adding new substitution parameters in the Substitution Parameter Editor, the editor automatically
adds the prefix.
When typing names in the Substitution Parameter Editor, do not use punctuation (including quotes
or brackets) except underscores. The following characters are not allowed:
,: / ' " = < > + | - * % ; t [ ] ( ) r n $ ] +
•
•
•
•
•
You can type names directly into fields, column mappings, transform options, and so on. However,
you must enclose them in square brackets, for example [$$SamplesInstall].
Names can include any alpha or numeric character or underscores but cannot contain spaces.
Names are not case sensitive.
The maximum length of a name is 64 characters.
Names must be unique within the repository.
13.8.2.2 Adding and defining substitution parameters
1. In the Designer, open the Substitution Parameter Editor by selecting Tools > Substitution Parameter
Configurations.
2. The first column lists the substitution parameters available in the repository. To create a new one,
double-click in a blank cell (a pencil icon will appear in the left) and type a name. The software
automatically adds a double dollar-sign prefix ($$) to the name when you navigate away from the
cell.
3. The second column identifies the name of the first configuration, by default Configuration1 (you
can change configuration names by double-clicking in the cell and retyping the name). Double-click
in the blank cell next to the substitution parameter name and type the constant value that the
parameter represents in that configuration. The software applies that value when you run the job.
294
2011-06-09](https://image.slidesharecdn.com/sbo401dsdesigneren-131213054333-phpapp02/85/SAP-BODS-Designer-PDF-294-320.jpg)
![Executing Jobs
14.4.1 Using logs
This section describes how to use logs in the Designer.
•
To open the trace log on job execution, select Tools > Options > Designer > General > Open
monitor on job execution.
•
To copy log content from an open log, select one or multiple lines and use the key commands
[Ctrl+C].
14.4.1.1 To access a log during job execution
If your Designer is running when job execution begins, the execution window opens automatically,
displaying the trace log information.
Use the monitor and error log icons (middle and right icons at the top of the execution window) to view
these logs.
The execution window stays open until you close it.
14.4.1.2 To access a log after the execution window has been closed
1. In the project area, click the Log tab.
2. Click a job name to view all trace, monitor, and error log files in the workspace. Or expand the job
you are interested in to view the list of trace log files and click one.
Log indicators signify the following:
Job Log Indicator
N_
Description
Indicates that the job executed successfully on this explicitly selected
Job Server.
Indicates that the was job executed successfully by a server group. The
Job Server listed executed the job.
307
2011-06-09](https://image.slidesharecdn.com/sbo401dsdesigneren-131213054333-phpapp02/85/SAP-BODS-Designer-PDF-307-320.jpg)
![Data Quality
Option
Description
No_Cache
Captures data at a point in time. The data doesn't change until the job restarts.
Pre-load Cache
Use this option for static data.
3. Depending on how your input data and secondary table are structured, do one of the following:
• Select Auto-generate SQL. Then select the Use break column from database option, if you
have one, and choose a column from the Break key field drop-down list.
Note:
If you choose the Auto-generate SQL option, we recommend that you have a break key column
in your secondary table and select the Use break column from database option. If you don't,
the SQL that is created could be incorrect.
•
Select Create custom SQL, and either click the Launch SQL Editor button or type your SQL
in the SQL edit box.
4. If you want to track your records from the input source, select Use constant source value.
5. Enter a value that represents your source in the Physical source value option, and then choose a
field that holds this value in the Physical source field drop-down list.
6. In the Column mapping table, add as many rows as you want. Each row is a field that will be added
to the collection.
a. Choose a field in the Mapped name column.
b. Choose a column from your secondary table (or from a custom query) in the Column name
option that contains the same type of data as specified in the Mapped name column.
If you have already defined your break keys in the Break Group option group, the fields used to
create the break key are posted here, with the Break Group column set to YES.
Writing custom SQL
Use placeholders
To avoid complicated SQL statements, you should use placeholders (which are replaced with real input
data) in your WHERE clause.
For example, let's say the customer database contains a field called MatchKey, and the record that
goes through the cleansing process gets a field generated called MATCH_KEY. This field has a
placeholder of [MATCHKEY]. The records that are selected from the customer database and appended
to the existing data collection are those that contain the same value in MatchKey as in the transaction's
MATCH_KEY. For this example, let's say the actual value is a 10-digit phone number.
The following is an example of what your SQL would look like with an actual phone number instead of
the [MATCHKEY] placeholder.
SELECT ContactGivenName1, ContactGivenName2, ContactFamilyName, Address1, Address2, City, Region, Postcode,
Country, AddrStreet, AddrStreetNumber, AddrUnitNumber
FROM TblCustomer
412
2011-06-09](https://image.slidesharecdn.com/sbo401dsdesigneren-131213054333-phpapp02/85/SAP-BODS-Designer-PDF-412-320.jpg)
![Data Quality
WHERE MatchKey = '123-555-9876';
Caution:
You must make sure that the SQL statement is optimized for best performance and will generate valid
results. The Candidate Selection operation does not do this for you.
Replace placeholder with actual values
After testing the SQL with actual values, you must replace the actual values with placeholders
([MATCHKEY], for example).
Your SQL should now look similar to the following.
SELECT ContactGivenName1, ContactGivenName2, ContactFamilyName, Address1, Address2, City, Region, Postcode,
Country, AddrStreet, AddrStreetNumber, AddrUnitNumber
FROM TblCustomer
WHERE MatchKey = [MATCHKEY];
Note:
Placeholders cannot be used for list values, for example in an IN clause:
WHERE status IN ([status])
If [status] is a list of values, this SQL statement will fail.
16.4.7.2.3 Compare tables
Compare tables are sets of rules that define which records to compare, sort of an additional way to
create break groups. You use your logical source values to determine which records are compared or
are not compared.
By using compare tables, you can compare records within sources, or you can compare records across
sources, or a combination of both.
To set up a compare table
Be sure to include a field that contains a logical source value before you add a Compare table operation
to the Match transform (in the Match level option group).
Here is an example of how to set up your compare table. Suppose you have two IDs (A and B), and
you only want to compare across sources, not within the sources.
1. If no Compare Table is present in the Matching section, right-click Matching > <Level Name>, and
select Add > Compare.
2. Set the Default action option to No_Match, and type None in the Default logical source value
option.
This tells the Match transform to not compare everything, but follow the comparison rules set by the
table entries.
413
2011-06-09](https://image.slidesharecdn.com/sbo401dsdesigneren-131213054333-phpapp02/85/SAP-BODS-Designer-PDF-413-320.jpg)
![Data Quality
Option
Setting
Best record strategy
NON_BLANK
Strategy priority
Priority option not available for the NON_BLANK strategy.
Strategy field
NORTH_AMERICAN_PHONE1_NORTH_AMERICAN_PHONE_STANDARDIZED.
Posting destination
ALL
Post only once per destination YES
Here is what the Python code looks like.
# Setup local temp variable to store updated compare condition
dct = locals()
# Store source and destination values to temporary variables
# Reset the temporary variable when the destination changes
if (dct.has_key('BEST_RECORD_TEMP') and NEWDST.GetBuffer() == u'F'):
DESTINATION = dct['BEST_RECORD_TEMP']
else:
DESTINATION = DST.GetField(u'NORTH_AMERICAN_PHONE1_NORTH_AMERICAN_PHONE_STANDARDIZED')
SOURCE = SRC.GetField(u'NORTH_AMERICAN_PHONE1_NORTH_AMERICAN_PHONE_STANDARDIZED')
if len(SOURCE.strip()) > 0 and len(DESTINATION.strip()) == 0:
RET.SetBuffer(u'T')
dct['BEST_RECORD_TEMP'] = SOURCE
else:
RET.SetBuffer(u'F')
dct['BEST_RECORD_TEMP'] = DESTINATION
# Delete temporary variables
del SOURCE
del DESTINATION
Example: NEWDST and NEWGRP
Suppose you have two match groups, each with three records.
Match group
Records
Match group 1
Record A
Record B
Record C
Match group 2
Record D
Record E
Record F
Each new destination or match group is flagged with a "T".
443
2011-06-09](https://image.slidesharecdn.com/sbo401dsdesigneren-131213054333-phpapp02/85/SAP-BODS-Designer-PDF-443-320.jpg)
![Data Quality
3. In the Global Address Transform, complete all applicable options in the Global Address > Report
Options > New Zealand subgroup.
4. In the Global Address Cleanse Transform, set Engines > Global Address to Yes.
After you run the job and produce the New Zealand Statement of Accuracy (SOA) report, you need to
rename the New Zealand Statement of Accuracy (SOA) report and New Zealand Statement of Accuracy
(SOA) Production Log before submitting your mailing. For more information on the required naming
format, See New Zealand SOA Report and SOA production log file.
Related Topics
• Management Console Guide: New Zealand Statement of Accuracy (SOA) report
• Reference Guide: Report options for New Zealand
16.5.10.2 New Zealand SOA Report and SOA production log file
New Zealand Statement of Accuracy (SOA) Report
The New Zealand Statement of Accuracy (SOA) report includes statistical information about address
cleansing for New Zealand.
New Zealand Statement of Accuracy (SOA) Production Log
The New Zealand Statement of Accuracy (SOA) production log contains identical information as the
SOA report in a pipe-delimited ASCII text file (with a header record). The software creates the SOA
production log by extracting data from the Sendrightaddraccuracy table within the repository. The
software appends a new record to the Sendrightaddraccuracy table each time a file is processed with
the DISABLE_CERTIFICATION option set to No. If the DISABLE_CERTIFICATION option is set to
Yes, the software does not produce the SOA report and an entry will not be appended to the
Sendrightaddraccuracy table. Mailers must retain the production log file for at least 2 years.
The default location of the SOA production log is <DataServicesInstallLocation>Business
ObjectsBusinessObjects Data ServicesDataQualitycertificationsCertifica
tionLogs.
Mailing requirements
The SOA report and production log are only required when you submit the data processed for a mailing
and want to receive postage discounts. Submit the SOA production log at least once a month. Submit
an SOA report for each file that is processed for mailing discounts.
File naming format
The SOA production log and SOA report must have a file name in the following format:
Production Log [SOA% (9999)]_[SOA Expiry Date (YYYYMMDD)]_[SOA ID].txt
493
2011-06-09](https://image.slidesharecdn.com/sbo401dsdesigneren-131213054333-phpapp02/85/SAP-BODS-Designer-PDF-493-320.jpg)
![Data Quality
SOA Report [SOA% (9999)]_[SOA Expiry Date (YYYYMMDD)]_[SOA ID].PDF
Example:
An SOA with:
SOA % = 94.3%
SOA expiry date = 15 Oct 2008
SOA ID = AGD07_12345678
The file names will be:
Production Log - 0943_20081015_AGD07_12345678.txt
SOA Report - 0943_20081015_AGD07_12345678.pdf
Related Topics
• Management Console Guide: New Zealand Statement of Accuracy (SOA) report
• Management Console Guide: Exporting New Zealand SOA certification logs
16.5.10.3 The New Zealand Certification blueprint
Do the following to edit the blueprint, run the job for New Zealand Certification, and generate the SOA
production log file:
1. Import nz_sendright_certification.atl located in the DataQualitycertifications folder in
the location where you installed the software. The default location is <DataServicesInstallLo
cation>Business ObjectsBusinessObjects Data
ServicesDataQualitycertifications.
The import adds the following objects to the repository:
• The project DataQualityCertifications
• The job Job_DqBatchNewZealand_SOAProductionLog
• The dataflow DF_DqBatchNewZealand_SOAProductionLog
• The datastore DataQualityCertifications
• The file format DqNewZealandSOAProductionLog
2. Edit the datastore DataQualityCertifications. Follow the steps listed in Editing the datastore .
3. Optional: By default, the software places the SOA Production Log in <DataServicesInstallLo
cation>Business ObjectsBusinessObjects Data ServicesDataQualitycerti
ficationsCertificationLogs. If the default location is acceptable, ignore this step. If you
want to output the production log file to a different location, edit the substitution parameter configu
494
2011-06-09](https://image.slidesharecdn.com/sbo401dsdesigneren-131213054333-phpapp02/85/SAP-BODS-Designer-PDF-494-320.jpg)
![Data Quality
Delivery Sequence Invoice Report
The USPS requires that you submit the Delivery Sequence Invoice report if you claim DSF2
walk-sequence discounts for this job.
US Addressing Report
•
•
The US Addressing Report is generated by the USA Regulatory Address Cleanse transform.
The Second Generation Delivery Sequence File Summary and Address Delivery Types sections of
the US Addressing Report shows counts and percentages of addresses in your file that match the
various DSF2 categories (if NCOALink is enabled). The information is listed for pre and post
NCOALink processing.
DSF2 Augment Statistics Log File
The USPS requires that DSF2 licensees save information about their processing in the DSF2 log file.
The USPS dictates the contents of the DSF2 log file and requires that you submit it to them monthly.
Log files are available to users with administrator or operator permissions.
Related Topics
• Management Console Guide: Administrator, Administrator management, Exporting DSF2 certification
log
• Management Console Guide: Data Quality reports, Delivery Sequence Invoice Report
• Management Console Guide: Data Quality reports, US Addressing Report
16.6.4.13.1 DSF2 Augment Statistics Log File
The DSF2 Augment Statistics Log File is stored in the repository. The software generates the log file
to the repository where you can export them by using the Data Services Management Console (for
Administrators or Operators only).
The naming format for the log file is as follows:
[DSF2_licensee_ID][mm][yy].dat
The USPS dictates the contents of the DSF2 log file and requires that you submit it to them monthly.
For details, see the DSF2 Licensee Performance Requirements document, which is available on the
USPS RIBBS website (http://ribbs.usps.gov/dsf2/documents/tech_guides).
You must submit the DSF2 log file to the USPS by the third business day of each month by e-mail.
16.6.5 NCOALink® overview
The USPS Move Update standard helps users and the USPS to reduce the number of records that are
returned because the address is out of date. NCOALink is a part of this effort. Move Updating is the
531
2011-06-09](https://image.slidesharecdn.com/sbo401dsdesigneren-131213054333-phpapp02/85/SAP-BODS-Designer-PDF-531-320.jpg)
![Data Quality
Option
Description
Windows
UNIX
/t
-t
Specify transfer location.
/nos
-nos
Do not stop on error (return failure code as exit status).
/a
-a
Answer all warning messages with Yes.
You can combine p options. For example, if you want to transfer, unpack, and verify all in the same
process, enter /p:tuv or -p:tuv.
After performing the p option specified, the program closes.
Example:
Your command line may look something like this:
Windows
ncoadvdver /p:tuv /d D: /t C:pwdirsncoa
UNIX
ncoadvdver [-c] [-a] [-nos] [-p:(t|u|v)][-d<path>] [-t <filename>]
16.6.5.11 About the NCOALink daily delete file
If you are a service provider, then every day before you perform NCOALink processing, you must
download the daily delete file and install it in the same directory as your NCOALink directories are
located.
The daily delete file contains records that are pending deletion from the NCOALink data. For example,
if Jane Doe filed a change of address with the USPS and then didn’t move, Jane’s record would be in
the daily delete file. Because the change of address is stored in the NCOALink directories, and they
are updated only weekly or monthly, the daily delete file is needed in the interim, until the NCOALink
directories are updated again.
Note:
If you are an end user, you only need the daily delete file for processing Stage I or II files. It is not
required for normal NCOALink processing.
Important points to know about the daily delete file:
•
542
The software will fail verification if an NCOALink certification stage test is being performed and the
daily delete file is not installed.
2011-06-09](https://image.slidesharecdn.com/sbo401dsdesigneren-131213054333-phpapp02/85/SAP-BODS-Designer-PDF-542-320.jpg)
SAP BODS Designer PDF
- 1. Designer Guide ■ SAP BusinessObjects Data Services 4.0 (14.0.1) 2011-06-09
- 2. Copyright © 2011 SAP AG. All rights reserved.SAP, R/3, SAP NetWeaver, Duet, PartnerEdge, ByDesign, SAP BusinessObjects Explorer, StreamWork, and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP AG in Germany and other countries.Business Objects and the Business Objects logo, BusinessObjects, Crystal Reports, Crystal Decisions, Web Intelligence, Xcelsius, and other Business Objects products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of Business Objects Software Ltd. Business Objects is an SAP company.Sybase and Adaptive Server, iAnywhere, Sybase 365, SQL Anywhere, and other Sybase products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of Sybase, Inc. Sybase is an SAP company. All other product and service names mentioned are the trademarks of their respective companies. Data contained in this document serves informational purposes only. National product specifications may vary.These materials are subject to change without notice. These materials are provided by SAP AG and its affiliated companies ("SAP Group") for informational purposes only, without representation or warranty of any kind, and SAP Group shall not be liable for errors or omissions with respect to the materials. The only warranties for SAP Group products and services are those that are set forth in the express warranty statements accompanying such products and services, if any. Nothing herein should be construed as constituting an additional warranty. 2011-06-09
- 3. Contents Chapter 1 1.1 1.1.1 1.1.2 1.1.3 1.1.4 1.2 1.2.1 1.2.2 Welcome to SAP BusinessObjects Data Services.................................................................19 Chapter 2 Logging into the Designer.....................................................................................................27 2.1 2.2 Version restrictions................................................................................................................27 Chapter 3 Designer User Interface........................................................................................................29 3.1 3.1.1 3.1.2 3.1.3 3.2 3.3 3.3.1 3.3.2 3.3.3 3.3.4 3.3.5 3.3.6 3.3.7 3.3.8 3.3.9 3.4 3.5 3 Introduction...........................................................................................................................19 Objects..................................................................................................................................29 Welcome...............................................................................................................................19 Documentation set for SAP BusinessObjects Data Services.................................................19 Accessing documentation......................................................................................................22 SAP BusinessObjects information resources.........................................................................23 Overview of this guide............................................................................................................24 About this guide.....................................................................................................................25 Who should read this guide....................................................................................................25 Resetting users......................................................................................................................28 Reusable objects...................................................................................................................29 Single-use objects..................................................................................................................30 Object hierarchy.....................................................................................................................30 Designer window...................................................................................................................31 Menu bar...............................................................................................................................32 Project menu..........................................................................................................................33 Edit menu...............................................................................................................................33 View menu.............................................................................................................................34 Tools menu............................................................................................................................34 Debug menu..........................................................................................................................36 Validation menu.....................................................................................................................36 Dictionary menu.....................................................................................................................37 Window menu........................................................................................................................38 Help menu..............................................................................................................................38 Toolbar...................................................................................................................................39 Project area ..........................................................................................................................41 2011-06-09
- 4. Contents 3.6 3.7 3.8 3.8.1 3.8.2 3.8.3 3.8.4 3.8.5 3.8.6 3.8.7 3.9 3.9.1 3.9.2 3.9.3 3.10 3.11 3.11.1 3.11.2 3.11.3 3.11.4 3.11.5 3.11.6 3.11.7 3.11.8 3.12 3.12.1 3.12.2 3.12.3 3.12.4 3.12.5 3.12.6 3.12.7 Chapter 4 Projects and Jobs.................................................................................................................67 4.1 4.1.1 4.1.2 4.1.3 4.1.4 4.2 4.2.1 4 Tool palette............................................................................................................................42 Projects.................................................................................................................................67 Designer keyboard accessibility.............................................................................................43 Workspace............................................................................................................................44 Moving objects in the workspace area...................................................................................44 Connecting objects................................................................................................................45 Disconnecting objects............................................................................................................45 Describing objects ................................................................................................................45 Scaling the workspace...........................................................................................................46 Arranging workspace windows...............................................................................................46 Closing workspace windows..................................................................................................46 Local object library.................................................................................................................47 To open the object library.......................................................................................................47 To display the name of each tab as well as its icon.................................................................48 To sort columns in the object library.......................................................................................48 Object editors........................................................................................................................49 Working with objects..............................................................................................................49 Creating new reusable objects...............................................................................................50 Changing object names..........................................................................................................51 Viewing and changing object properties.................................................................................52 Creating descriptions.............................................................................................................53 Creating annotations .............................................................................................................55 Copying objects.....................................................................................................................56 Saving and deleting objects....................................................................................................57 Searching for objects.............................................................................................................59 General and environment options...........................................................................................61 Designer — Environment.......................................................................................................61 Designer — General..............................................................................................................62 Designer — Graphics.............................................................................................................64 Designer — Central Repository Connections.........................................................................65 Data — General.....................................................................................................................65 Job Server — Environment....................................................................................................66 Job Server — General...........................................................................................................66 Objects that make up a project..............................................................................................67 Creating a new project...........................................................................................................68 Opening existing projects.......................................................................................................68 Saving projects......................................................................................................................69 Jobs.......................................................................................................................................69 Creating jobs.........................................................................................................................70 2011-06-09
- 5. Contents 4.2.2 Chapter 5 Datastores.............................................................................................................................73 5.1 5.2 5.2.1 5.2.2 5.2.3 5.2.4 5.2.5 5.2.6 5.2.7 5.2.8 5.2.9 5.3 5.3.1 5.3.2 5.3.3 5.4 5.4.1 5.4.2 5.4.3 5.5 5.5.1 5.5.2 5.5.3 5.5.4 5.5.5 5.5.6 5.5.7 5.5.8 5.5.9 What are datastores?.............................................................................................................73 Chapter 6 File formats.........................................................................................................................123 6.1 6.2 6.3 6.3.1 6.3.2 6.3.3 5 Naming conventions for objects in jobs..................................................................................71 Understanding file formats...................................................................................................123 Database datastores..............................................................................................................74 Mainframe interface...............................................................................................................74 Defining a database datastore................................................................................................77 Configuring ODBC data sources on UNIX..............................................................................80 Changing a datastore definition..............................................................................................80 Browsing metadata through a database datastore..................................................................81 Importing metadata through a database datastore..................................................................84 Memory datastores................................................................................................................90 Persistent cache datastores...................................................................................................94 Linked datastores...................................................................................................................97 Adapter datastores................................................................................................................99 Defining an adapter datastore..............................................................................................100 Browsing metadata through an adapter datastore................................................................102 Importing metadata through an adapter datastore................................................................102 Web service datastores.......................................................................................................103 Defining a web service datastore.........................................................................................103 Browsing WSDL metadata through a web service datastore................................................104 Importing metadata through a web service datastore...........................................................106 Creating and managing multiple datastore configurations.....................................................106 Definitions............................................................................................................................107 Why use multiple datastore configurations?.........................................................................108 Creating a new configuration................................................................................................108 Adding a datastore alias.......................................................................................................110 Functions to identify the configuration..................................................................................110 Portability solutions..............................................................................................................112 Job portability tips................................................................................................................116 Renaming table and function owner......................................................................................117 Defining a system configuration...........................................................................................121 File format editor..................................................................................................................124 Creating file formats.............................................................................................................126 To create a new file format...................................................................................................126 Modeling a file format on a sample file.................................................................................127 Replicating and renaming file formats...................................................................................128 2011-06-09
- 6. Contents 6.3.4 6.3.5 6.4 6.4.1 6.4.2 6.4.3 6.5 6.5.1 6.5.2 6.5.3 6.5.4 6.5.5 6.5.6 6.5.7 6.6 6.6.1 6.6.2 6.6.3 6.6.4 6.7 6.7.1 6.7.2 6.7.3 6.7.4 6.8 6.8.1 6.9 6.9.1 6.9.2 6.9.3 6.10 Chapter 7 Data Flows..........................................................................................................................151 7.1 7.1.1 7.1.2 7.1.3 7.1.4 7.1.5 7.1.6 7.1.7 6 To create a file format from an existing flat table schema.....................................................129 What is a data flow?.............................................................................................................151 To create a specific source or target file...............................................................................129 Editing file formats................................................................................................................130 To edit a file format template................................................................................................130 To edit a source or target file...............................................................................................131 Change multiple column properties......................................................................................131 File format features..............................................................................................................132 Reading multiple files at one time.........................................................................................132 Identifying source file names ...............................................................................................133 Number formats...................................................................................................................133 Ignoring rows with specified markers....................................................................................134 Date formats at the field level...............................................................................................135 Parallel process threads.......................................................................................................135 Error handling for flat-file sources.........................................................................................136 File transfers........................................................................................................................139 Custom transfer system variables for flat files......................................................................139 Custom transfer options for flat files....................................................................................140 Setting custom transfer options...........................................................................................141 Design tips...........................................................................................................................142 Creating COBOL copybook file formats...............................................................................143 To create a new COBOL copybook file format.....................................................................144 To create a new COBOL copybook file format and a data file..............................................144 To create rules to identify which records represent which schemas.....................................145 To identify the field that contains the length of the schema's record.....................................146 Creating Microsoft Excel workbook file formats on UNIX platforms .....................................146 To create a Microsoft Excel workbook file format on UNIX ..................................................147 Creating Web log file formats...............................................................................................147 Word_ext function................................................................................................................148 Concat_date_time function...................................................................................................149 WL_GetKeyValue function...................................................................................................149 Unstructured file formats......................................................................................................149 Naming data flows................................................................................................................151 Data flow example................................................................................................................151 Steps in a data flow..............................................................................................................152 Data flows as steps in work flows........................................................................................152 Intermediate data sets in a data flow....................................................................................153 Operation codes..................................................................................................................153 Passing parameters to data flows.........................................................................................154 2011-06-09
- 7. Contents 7.2 7.2.1 7.2.2 7.2.3 7.3 7.3.1 7.3.2 7.3.3 7.3.4 7.3.5 7.4 7.4.1 7.4.2 7.5 7.5.1 7.5.2 7.5.3 7.6 7.6.1 7.6.2 7.6.3 7.6.4 7.7 Chapter 8 Transforms..........................................................................................................................175 8.1 8.2 8.3 8.3.1 8.3.2 8.4 8.4.1 8.4.2 8.5 8.5.1 8.5.2 8.6 8.6.1 8.6.2 8.6.3 8.6.4 7 Creating and defining data flows..........................................................................................155 To add a transform to a data flow.........................................................................................177 To define a new data flow using the object library.................................................................155 To define a new data flow using the tool palette...................................................................155 To change properties of a data flow.....................................................................................155 Source and target objects....................................................................................................156 Source objects.....................................................................................................................157 Target objects......................................................................................................................157 Adding source or target objects to data flows......................................................................158 Template tables....................................................................................................................160 Converting template tables to regular tables........................................................................161 Adding columns within a data flow .......................................................................................162 To add columns within a data flow........................................................................................163 Propagating columns in a data flow containing a Merge transform........................................163 Lookup tables and the lookup_ext function...........................................................................164 Accessing the lookup_ext editor..........................................................................................165 Example: Defining a simple lookup_ext function....................................................................166 Example: Defining a complex lookup_ext function ................................................................169 Data flow execution.............................................................................................................171 Push down operations to the database server......................................................................171 Distributed data flow execution............................................................................................172 Load balancing.....................................................................................................................173 Caches................................................................................................................................173 Audit Data Flow overview.....................................................................................................174 Transform editors.................................................................................................................178 Transform configurations......................................................................................................179 To create a transform configuration......................................................................................179 To add a user-defined field ..................................................................................................180 The Query transform ...........................................................................................................181 To add a Query transform to a data flow..............................................................................181 Query Editor.........................................................................................................................182 Data Quality transforms ......................................................................................................184 To add a Data Quality transform to a data flow.....................................................................184 Data Quality transform editors.............................................................................................186 Text Data Processing transforms.........................................................................................189 Text Data Processing overview............................................................................................189 Entity Extraction transform overview.....................................................................................190 Using the Entity Extraction transform....................................................................................193 Differences between text data processing and data cleanse transforms...............................194 2011-06-09
- 8. Contents 8.6.5 8.6.6 8.6.7 8.6.8 8.6.9 Chapter 9 Work Flows.........................................................................................................................201 9.1 9.2 9.3 9.4 9.5 9.5.1 9.5.2 9.5.3 9.6 9.6.1 9.7 9.7.1 9.7.2 9.7.3 9.8 9.8.1 9.8.2 9.8.3 9.9 9.9.1 9.9.2 What is a work flow?............................................................................................................201 Chapter 10 Nested Data........................................................................................................................217 10.1 10.2 10.3 10.3.1 10.3.2 10.3.3 10.3.4 10.3.5 10.4 10.4.1 8 Using multiple transforms.....................................................................................................195 What is nested data?...........................................................................................................217 Examples for using the Entity Extraction transform...............................................................195 To add a text data processing transform to a data flow........................................................196 Entity Extraction transform editor.........................................................................................198 Using filtering options..........................................................................................................199 Steps in a work flow.............................................................................................................202 Order of execution in work flows..........................................................................................202 Example of a work flow........................................................................................................203 Creating work flows.............................................................................................................204 To create a new work flow using the object library...............................................................204 To create a new work flow using the tool palette .................................................................204 To specify that a job executes the work flow one time.........................................................204 Conditionals.........................................................................................................................205 To define a conditional.........................................................................................................206 While loops..........................................................................................................................207 Design considerations..........................................................................................................207 Defining a while loop............................................................................................................209 Using a while loop with View Data........................................................................................210 Try/catch blocks..................................................................................................................210 Defining a try/catch block....................................................................................................211 Categories of available exceptions.......................................................................................212 Example: Catching details of an error...................................................................................213 Scripts.................................................................................................................................214 To create a script.................................................................................................................214 Debugging scripts using the print function............................................................................215 Representing hierarchical data.............................................................................................217 Formatting XML documents.................................................................................................220 Importing XML Schemas......................................................................................................220 Specifying source options for XML files ..............................................................................225 Mapping optional schemas...................................................................................................226 Using Document Type Definitions (DTDs) ...........................................................................228 Generating DTDs and XML Schemas from an NRDM schema.............................................230 Operations on nested data...................................................................................................230 Overview of nested data and the Query transform...............................................................231 2011-06-09
- 9. Contents 10.4.2 10.4.3 10.4.4 10.4.5 10.4.6 10.4.7 10.4.8 10.5 10.5.1 Chapter 11 Real-time Jobs....................................................................................................................249 11.1 11.2 11.2.1 11.2.2 11.2.3 11.3 11.3.1 11.3.2 11.3.3 11.4 11.4.1 11.4.2 11.4.3 11.4.4 11.5 11.5.1 11.5.2 11.5.3 11.6 11.6.1 11.6.2 11.6.3 11.7 11.7.1 11.7.2 11.7.3 Request-response message processing...............................................................................249 Chapter 12 Embedded Data Flows........................................................................................................269 12.1 9 FROM clause construction...................................................................................................231 Overview of embedded data flows.......................................................................................269 Nesting columns .................................................................................................................234 Using correlated columns in nested data..............................................................................236 Distinct rows and nested data..............................................................................................237 Grouping values across nested schemas.............................................................................237 Unnesting nested data ........................................................................................................238 Transforming lower levels of nested data.............................................................................241 XML extraction and parsing for columns...............................................................................241 Sample scenarios.................................................................................................................242 What is a real-time job?........................................................................................................250 Real-time versus batch.........................................................................................................250 Messages............................................................................................................................251 Real-time job examples........................................................................................................252 Creating real-time jobs.........................................................................................................254 Real-time job models............................................................................................................254 Using real-time job models...................................................................................................255 To create a real-time job with a single dataflow....................................................................257 Real-time source and target objects.....................................................................................258 To view an XML message source or target schema.............................................................259 Secondary sources and targets............................................................................................259 Transactional loading of tables.............................................................................................259 Design tips for data flows in real-time jobs...........................................................................260 Testing real-time jobs...........................................................................................................261 Executing a real-time job in test mode..................................................................................261 Using View Data..................................................................................................................261 Using an XML file target.......................................................................................................262 Building blocks for real-time jobs..........................................................................................263 Supplementing message data..............................................................................................263 Branching data flow based on a data cache value.................................................................265 Calling application functions.................................................................................................266 Designing real-time applications...........................................................................................267 Reducing queries requiring back-office application access....................................................267 Messages from real-time jobs to adapter instances.............................................................267 Real-time service invoked by an adapter instance.................................................................268 2011-06-09
- 10. Contents 12.2 12.3 12.3.1 12.3.2 12.3.3 12.3.4 12.3.5 Chapter 13 Variables and Parameters...................................................................................................277 13.1 13.2 13.2.1 13.3 13.3.1 13.3.2 13.3.3 13.3.4 13.4 13.4.1 13.4.2 13.4.3 13.5 13.5.1 13.5.2 13.5.3 13.6 13.7 13.7.1 13.8 13.8.1 13.8.2 13.8.3 13.8.4 13.8.5 13.8.6 Overview of variables and parameters..................................................................................277 Chapter 14 Executing Jobs....................................................................................................................301 14.1 14.2 14.2.1 10 Example of when to use embedded data flows.....................................................................270 Overview of job execution....................................................................................................301 Creating embedded data flows.............................................................................................270 Using the Make Embedded Data Flow option.......................................................................271 Creating embedded data flows from existing flows...............................................................272 Using embedded data flows.................................................................................................273 Separately testing an embedded data flow...........................................................................275 Troubleshooting embedded data flows.................................................................................276 The Variables and Parameters window.................................................................................278 To view the variables and parameters in each job, work flow, or data flow............................278 Using local variables and parameters...................................................................................280 Parameters..........................................................................................................................281 Passing values into data flows..............................................................................................281 To define a local variable......................................................................................................282 Defining parameters.............................................................................................................282 Using global variables ..........................................................................................................284 Creating global variables......................................................................................................284 Viewing global variables ......................................................................................................285 Setting global variable values...............................................................................................285 Local and global variable rules..............................................................................................289 Naming................................................................................................................................289 Replicating jobs and work flows...........................................................................................289 Importing and exporting........................................................................................................289 Environment variables..........................................................................................................290 Setting file names at run-time using variables.......................................................................290 To use a variable in a flat file name.......................................................................................290 Substitution parameters.......................................................................................................291 Overview of substitution parameters....................................................................................291 Using the Substitution Parameter Editor...............................................................................293 Associating a substitution parameter configuration with a system configuration...................295 Overriding a substitution parameter in the Administrator......................................................297 Executing a job with substitution parameters .......................................................................297 Exporting and importing substitution parameters..................................................................298 Preparing for job execution...................................................................................................301 Validating jobs and job components.....................................................................................302 2011-06-09
- 11. Contents 14.2.2 14.2.3 14.3 14.3.1 14.3.2 14.3.3 14.4 14.4.1 14.4.2 14.5 14.5.1 14.5.2 Chapter 15 Data Assessment................................................................................................................315 15.1 15.1.1 15.1.2 15.1.3 15.1.4 15.1.5 15.1.6 15.2 15.2.1 15.2.2 15.2.3 15.3 15.3.1 15.3.2 15.4 15.4.1 15.4.2 15.4.3 15.4.4 15.4.5 15.4.6 15.4.7 Using the Data Profiler.........................................................................................................316 Chapter 16 Data Quality........................................................................................................................353 16.1 16.2 11 Ensuring that the Job Server is running................................................................................303 Overview of data quality.......................................................................................................353 Setting job execution options...............................................................................................303 Executing jobs as immediate tasks.......................................................................................304 To execute a job as an immediate task.................................................................................304 Monitor tab .........................................................................................................................305 Log tab ...............................................................................................................................306 Debugging execution errors.................................................................................................306 Using logs............................................................................................................................307 Examining target data...........................................................................................................309 Changing Job Server options...............................................................................................309 To change option values for an individual Job Server...........................................................312 To use mapped drive names in a path..................................................................................314 Data sources that you can profile.........................................................................................316 Connecting to the profiler server..........................................................................................317 Profiler statistics..................................................................................................................318 Executing a profiler task.......................................................................................................321 Monitoring profiler tasks using the Designer........................................................................326 Viewing the profiler results...................................................................................................327 Using View Data to determine data quality...........................................................................333 Data tab...............................................................................................................................334 Profile tab............................................................................................................................335 Relationship Profile or Column Profile tab.............................................................................335 Using the Validation transform.............................................................................................335 Analyzing the column profile.................................................................................................336 Defining a validation rule based on a column profile..............................................................337 Using Auditing .....................................................................................................................338 Auditing objects in a data flow..............................................................................................339 Accessing the Audit window................................................................................................343 Defining audit points, rules, and action on failure..................................................................344 Guidelines to choose audit points .......................................................................................346 Auditing embedded data flows.............................................................................................347 Resolving invalid audit labels................................................................................................350 Viewing audit results ...........................................................................................................350 Data Cleanse.......................................................................................................................353 2011-06-09
- 12. Contents 16.2.1 16.2.2 16.2.3 16.2.4 16.2.5 16.2.6 16.2.7 16.2.8 16.2.9 16.2.10 16.3 16.3.1 16.3.2 16.3.3 16.4 16.4.1 16.4.2 16.4.3 16.4.4 16.4.5 16.4.6 16.4.7 16.4.8 16.4.9 16.4.10 16.4.11 16.4.12 16.4.13 16.5 16.5.1 16.5.2 16.5.3 16.5.4 16.5.5 16.5.6 16.5.7 16.5.8 16.5.9 16.5.10 16.5.11 16.5.12 12 About cleansing data............................................................................................................353 Cleansing package lifecycle: develop, deploy and maintain ..................................................354 Configuring the Data Cleanse transform..............................................................................356 Ranking and prioritizing parsing engines...............................................................................357 About parsing data...............................................................................................................358 About standardizing data......................................................................................................364 About assigning gender descriptions and prenames.............................................................364 Prepare records for matching...............................................................................................365 Region-specific data.............................................................................................................367 Japanese data......................................................................................................................368 Geocoding...........................................................................................................................369 POI and address geocoding ................................................................................................370 POI and address reverse geocoding ....................................................................................376 Understanding your output...................................................................................................387 Match..................................................................................................................................389 Matching strategies..............................................................................................................389 Match components..............................................................................................................389 Match Wizard.......................................................................................................................391 Transforms for match data flows..........................................................................................398 Working in the Match and Associate editors........................................................................399 Physical and logical sources.................................................................................................400 Match preparation................................................................................................................404 Match criteria.......................................................................................................................424 Post-match processing.........................................................................................................440 Association matching...........................................................................................................458 Unicode matching................................................................................................................458 Phonetic matching................................................................................................................461 Set up for match reports .....................................................................................................463 Address Cleanse..................................................................................................................464 How address cleanse works.................................................................................................464 Prepare your input data........................................................................................................467 Determine which transform(s) to use...................................................................................469 Identify the country of destination.........................................................................................472 Set up the reference files.....................................................................................................473 Define the standardization options.......................................................................................474 Process Japanese addresses .............................................................................................475 Process Chinese addresses.................................................................................................485 Supported countries (Global Address Cleanse)....................................................................490 New Zealand Certification....................................................................................................492 Global Address Cleanse Suggestion List.............................................................................496 Global Suggestion List.........................................................................................................496 2011-06-09
- 13. Contents 16.6 16.6.1 16.6.2 16.6.3 16.6.4 16.6.5 16.6.6 16.6.7 16.6.8 16.6.9 16.6.10 16.6.11 16.6.12 16.7 16.7.1 16.8 Chapter 17 Design and Debug..............................................................................................................585 17.1 17.1.1 17.1.2 17.1.3 17.2 17.2.1 17.2.2 17.2.3 17.2.4 17.2.5 17.3 17.3.1 17.3.2 17.3.3 17.3.4 17.3.5 17.3.6 17.3.7 17.4 17.4.1 17.4.2 17.4.3 17.4.4 13 Beyond the basic address cleansing.....................................................................................497 Using View Where Used......................................................................................................585 USPS DPV®.........................................................................................................................497 LACSLink®...........................................................................................................................508 SuiteLink™............................................................................................................................518 USPS DSF2®.......................................................................................................................521 NCOALink® overview...........................................................................................................531 USPS eLOT® .......................................................................................................................550 Early Warning System (EWS)...............................................................................................551 USPS RDI®..........................................................................................................................552 GeoCensus (USA Regulatory Address Cleanse).................................................................556 Z4Change (USA Regulatory Address Cleanse)....................................................................560 Suggestion lists overview.....................................................................................................562 Multiple data source statistics reporting...............................................................................565 Data Quality support for native data types............................................................................583 Data Quality data type definitions.........................................................................................583 Data Quality support for NULL values..................................................................................584 Accessing View Where Used from the object library............................................................586 Accessing View Where Used from the workspace...............................................................588 Limitations...........................................................................................................................588 Using View Data..................................................................................................................589 Accessing View Data...........................................................................................................589 Viewing data in the workspace.............................................................................................590 View Data Properties...........................................................................................................592 View Data tool bar options...................................................................................................596 View Data tabs....................................................................................................................597 Using the interactive debugger.............................................................................................600 Before starting the interactive debugger...............................................................................601 Starting and stopping the interactive debugger.....................................................................604 Panes...................................................................................................................................606 Debug menu options and tool bar.........................................................................................610 Viewing data passed by transforms......................................................................................612 Push-down optimizer............................................................................................................613 Limitations...........................................................................................................................613 Comparing Objects..............................................................................................................614 To compare two different objects.........................................................................................614 To compare two versions of the same object.......................................................................615 Overview of the Difference Viewer window..........................................................................615 Navigating through differences.............................................................................................619 2011-06-09
- 14. Contents 17.5 17.5.1 17.5.2 Chapter 18 Exchanging Metadata..........................................................................................................623 18.1 18.1.1 18.1.2 18.2 18.2.1 18.2.2 18.2.3 18.2.4 Metadata exchange..............................................................................................................623 Chapter 19 Recovery Mechanisms........................................................................................................629 19.1 19.2 19.2.1 19.2.2 19.2.3 19.2.4 19.2.5 19.2.6 19.2.7 19.3 19.4 19.4.1 19.4.2 19.4.3 Recovering from unsuccessful job execution........................................................................629 Chapter 20 Techniques for Capturing Changed Data............................................................................643 20.1 20.1.1 20.1.2 20.1.3 20.2 20.2.1 20.2.2 20.2.3 20.2.4 14 Calculating column mappings...............................................................................................619 Understanding changed-data capture...................................................................................643 To automatically calculate column mappings ........................................................................620 To manually calculate column mappings ..............................................................................620 Importing metadata files into the software............................................................................624 Exporting metadata files from the software...........................................................................624 Creating SAP universes.......................................................................................................625 To create universes using the Tools menu ...........................................................................625 To create universes using the object library..........................................................................626 Mappings between repository and universe metadata..........................................................626 Attributes that support metadata exchange..........................................................................627 Automatically recovering jobs...............................................................................................630 Enabling automated recovery...............................................................................................630 Marking recovery units.........................................................................................................631 Running in recovery mode....................................................................................................632 Ensuring proper execution path............................................................................................632 Using try/catch blocks with automatic recovery...................................................................633 Ensuring that data is not duplicated in targets.......................................................................635 Using preload SQL to allow re-executable data flows ..........................................................636 Manually recovering jobs using status tables........................................................................637 Processing data with problems.............................................................................................638 Using overflow files..............................................................................................................639 Filtering missing or bad values .............................................................................................639 Handling facts with missing dimensions................................................................................640 Full refresh...........................................................................................................................643 Capturing only changes........................................................................................................643 Source-based and target-based CDC..................................................................................644 Using CDC with Oracle sources..........................................................................................646 Overview of CDC for Oracle databases...............................................................................646 Setting up Oracle CDC........................................................................................................650 To create a CDC datastore for Oracle.................................................................................651 Importing CDC data from Oracle..........................................................................................651 2011-06-09
- 15. Contents 20.2.5 20.2.6 20.2.7 20.2.8 20.2.9 20.3 20.3.1 20.3.2 20.3.3 20.3.4 20.3.5 20.3.6 20.4 20.4.1 20.4.2 20.4.3 20.4.4 20.4.5 20.4.6 20.5 20.5.1 20.5.2 20.5.3 20.5.4 20.5.5 20.6 Chapter 21 Monitoring Jobs..................................................................................................................697 21.1 21.2 21.2.1 21.2.2 21.2.3 21.2.4 21.2.5 21.2.6 Administrator.......................................................................................................................697 Chapter 22 Multi-user Development......................................................................................................713 22.1 22.2 15 Viewing an imported CDC table...........................................................................................654 Central versus local repository.............................................................................................713 To configure an Oracle CDC source table............................................................................656 To create a data flow with an Oracle CDC source................................................................659 Maintaining CDC tables and subscriptions...........................................................................659 Limitations...........................................................................................................................660 Using CDC with Attunity mainframe sources.......................................................................661 Setting up Attunity CDC......................................................................................................662 Setting up the software for CDC on mainframe sources......................................................663 Importing mainframe CDC data............................................................................................664 Configuring a mainframe CDC source..................................................................................666 Using mainframe check-points.............................................................................................668 Limitations...........................................................................................................................669 Using CDC with Microsoft SQL Server databases ..............................................................669 Overview of CDC for SQL Server databases.......................................................................669 Setting up Microsoft SQL Server for CDC...........................................................................671 Setting up the software for CDC on SQL Server.................................................................673 Importing SQL Server CDC data..........................................................................................674 Configuring a SQL Server CDC source...............................................................................675 Limitations...........................................................................................................................678 Using CDC with timestamp-based sources..........................................................................679 Processing timestamps........................................................................................................680 Overlaps..............................................................................................................................682 Types of timestamps............................................................................................................688 Timestamp-based CDC examples........................................................................................689 Additional job design tips.....................................................................................................695 Using CDC for targets.........................................................................................................696 SNMP support.....................................................................................................................697 About the SNMP agent........................................................................................................697 Job Server, SNMP agent, and NMS application architecture...............................................698 About SNMP Agent's Management Information Base (MIB)................................................699 About an NMS application...................................................................................................701 Configuring the software to support an NMS application......................................................702 Troubleshooting...................................................................................................................711 Multiple users......................................................................................................................714 2011-06-09
- 16. Contents 22.3 Chapter 23 Multi-user Environment Setup............................................................................................717 23.1 23.2 23.3 23.3.1 23.3.2 23.3.3 23.3.4 Create a nonsecure central repository.................................................................................717 Chapter 24 Implementing Central Repository Security.........................................................................721 24.1 24.1.1 24.1.2 24.1.3 24.2 24.2.1 24.2.2 24.3 24.4 24.5 24.6 24.6.1 Overview..............................................................................................................................721 Chapter 25 Working in a Multi-user Environment.................................................................................727 25.1 25.2 25.2.1 25.2.2 25.3 25.3.1 25.3.2 25.3.3 25.4 25.4.1 25.4.2 25.5 25.5.1 25.5.2 16 Security and the central repository.......................................................................................716 Filtering................................................................................................................................727 Define a connection to a nonsecure central repository.........................................................718 Activating a central repository..............................................................................................718 To activate a central repository............................................................................................719 To open the central object library.........................................................................................719 To change the active central repository................................................................................719 To change central repository connections............................................................................720 Group-based permissions....................................................................................................721 Permission levels.................................................................................................................722 Process summary................................................................................................................722 Creating a secure central repository.....................................................................................723 To create a secure central repository...................................................................................723 To upgrade a central repository from nonsecure to secure...................................................723 Adding a multi-user administrator (optional)..........................................................................724 Setting up groups and users................................................................................................724 Defining a connection to a secure central repository............................................................724 Working with objects in a secure central repository..............................................................725 Viewing and modifying permissions......................................................................................725 Adding objects to the central repository...............................................................................728 To add a single object to the central repository....................................................................728 To add an object and its dependent objects to the central repository...................................729 Checking out objects...........................................................................................................729 Check out single objects or objects with dependents...........................................................730 Check out single objects or objects with dependents without replacement..........................731 Check out objects with filtering............................................................................................732 Undoing check out...............................................................................................................732 To undo single object check out...........................................................................................733 To undo check out of an object and its dependents..............................................................733 Checking in objects..............................................................................................................733 Checking in single objects, objects with dependents............................................................734 Checking in an object with filtering.......................................................................................735 2011-06-09
- 17. Contents 25.6 25.6.1 25.7 25.7.1 25.7.2 25.7.3 25.8 25.9 25.9.1 25.9.2 25.9.3 25.10 Labeling objects...................................................................................................................735 Chapter 26 Migrating Multi-user Jobs...................................................................................................743 26.1 26.2 26.2.1 26.3 Application phase management............................................................................................743 Index 17 To label an object and its dependents..................................................................................737 Getting objects....................................................................................................................737 To get a single object...........................................................................................................737 To get an object and its dependent objects..........................................................................738 To get an object and its dependent objects with filtering......................................................738 Comparing objects...............................................................................................................738 Viewing object history..........................................................................................................739 To examine the history of an object......................................................................................739 To get a previous version of an object..................................................................................740 To get an object with a particular label..................................................................................740 Deleting objects...................................................................................................................740 Copying contents between central repositories....................................................................744 To copy the contents of one central repository to another central repository.......................744 Central repository migration.................................................................................................745 747 2011-06-09
- 19. Introduction Introduction 1.1 Welcome to SAP BusinessObjects Data Services 1.1.1 Welcome SAP BusinessObjects Data Services delivers a single enterprise-class solution for data integration, data quality, data profiling, and text data processing that allows you to integrate, transform, improve, and deliver trusted data to critical business processes. It provides one development UI, metadata repository, data connectivity layer, run-time environment, and management console—enabling IT organizations to lower total cost of ownership and accelerate time to value. With SAP BusinessObjects Data Services, IT organizations can maximize operational efficiency with a single solution to improve data quality and gain access to heterogeneous sources and applications. 1.1.2 Documentation set for SAP BusinessObjects Data Services You should become familiar with all the pieces of documentation that relate to your SAP BusinessObjects Data Services product. Document What this document provides Administrator's Guide Information about administrative tasks such as monitoring, lifecycle management, security, and so on. Customer Issues Fixed Information about customer issues fixed in this release. Designer Guide Information about how to use SAP BusinessObjects Data Services Designer. Documentation Map Information about available SAP BusinessObjects Data Services books, languages, and locations. 19 2011-06-09
- 20. Introduction Document What this document provides Installation Guide for Windows Information about and procedures for installing SAP BusinessObjects Data Services in a Windows environment. Installation Guide for UNIX Information about and procedures for installing SAP BusinessObjects Data Services in a UNIX environment. Integrator's Guide Information for third-party developers to access SAP BusinessObjects Data Services functionality using web services and APIs. Management Console Guide Information about how to use SAP BusinessObjects Data Services Administrator and SAP BusinessObjects Data Services Metadata Reports. Performance Optimization Guide Information about how to improve the performance of SAP BusinessObjects Data Services. Reference Guide Detailed reference material for SAP BusinessObjects Data Services Designer. Release Notes Important information you need before installing and deploying this version of SAP BusinessObjects Data Services. Technical Manuals A compiled “master” PDF of core SAP BusinessObjects Data Services books containing a searchable master table of contents and index: • Administrator's Guide • Designer Guide • Reference Guide • Management Console Guide • Performance Optimization Guide • Supplement for J.D. Edwards • Supplement for Oracle Applications • Supplement for PeopleSoft • Supplement for Salesforce.com • Supplement for Siebel • Supplement for SAP Text Data Processing Extraction Customization Guide Information about building dictionaries and extraction rules to create your own extraction patterns to use with Text Data Processing transforms. Text Data Processing Language Reference Guide Information about the linguistic analysis and extraction processing features that the Text Data Processing component provides, as well as a reference section for each language supported. 20 2011-06-09
- 21. Introduction Document What this document provides Tutorial A step-by-step introduction to using SAP BusinessObjects Data Services. Upgrade Guide Release-specific product behavior changes from earlier versions of SAP BusinessObjects Data Services to the latest release. This manual also contains information about how to migrate from SAP BusinessObjects Data Quality Management to SAP BusinessObjects Data Services. What's New Highlights of new key features in this SAP BusinessObjects Data Services release. This document is not updated for support package or patch releases. In addition, you may need to refer to several Adapter Guides and Supplemental Guides. Document What this document provides Supplement for J.D. Edwards Information about interfaces between SAP BusinessObjects Data Services and J.D. Edwards World and J.D. Edwards OneWorld. Supplement for Oracle Applications Information about the interface between SAP BusinessObjects Data Services and Oracle Applications. Supplement for PeopleSoft Information about interfaces between SAP BusinessObjects Data Services and PeopleSoft. Supplement for Salesforce.com Information about how to install, configure, and use the SAP BusinessObjects Data Services Salesforce.com Adapter Interface. Supplement for SAP Information about interfaces between SAP BusinessObjects Data Services, SAP Applications, and SAP NetWeaver BW. Supplement for Siebel Information about the interface between SAP BusinessObjects Data Services and Siebel. We also include these manuals for information about SAP BusinessObjects Information platform services. Document What this document provides Information Platform Services Administrator's Guide Information for administrators who are responsible for configuring, managing, and maintaining an Information platform services installation. Information Platform Services Installation Guide for UNIX Installation procedures for SAP BusinessObjects Information platform services on a UNIX environment. 21 2011-06-09
- 22. Introduction Document What this document provides Information Platform Services Installation Guide for Windows Installation procedures for SAP BusinessObjects Information platform services on a Windows environment. 1.1.3 Accessing documentation You can access the complete documentation set for SAP BusinessObjects Data Services in several places. 1.1.3.1 Accessing documentation on Windows After you install SAP BusinessObjects Data Services, you can access the documentation from the Start menu. 1. Choose Start > Programs > SAP BusinessObjects Data Services 4.0 > Data Services Documentation. Note: Only a subset of the documentation is available from the Start menu. The documentation set for this release is available in <LINK_DIR>DocBooksen. 2. Click the appropriate shortcut for the document that you want to view. 1.1.3.2 Accessing documentation on UNIX After you install SAP BusinessObjects Data Services, you can access the online documentation by going to the directory where the printable PDF files were installed. 1. Go to <LINK_DIR>/doc/book/en/. 2. Using Adobe Reader, open the PDF file of the document that you want to view. 1.1.3.3 Accessing documentation from the Web 22 2011-06-09
- 23. Introduction You can access the complete documentation set for SAP BusinessObjects Data Services from the SAP BusinessObjects Business Users Support site. 1. Go to http://help.sap.com. 2. Click SAP BusinessObjects at the top of the page. 3. Click All Products in the navigation pane on the left. You can view the PDFs online or save them to your computer. 1.1.4 SAP BusinessObjects information resources A global network of SAP BusinessObjects technology experts provides customer support, education, and consulting to ensure maximum information management benefit to your business. Useful addresses at a glance: 23 2011-06-09
- 24. Introduction Address Content Customer Support, Consulting, and Education services Information about SAP Business User Support programs, as well as links to technical articles, downloads, and online forums. Consulting services can provide you with information about how SAP BusinessObjects can help maximize your information management investment. Education services can provide information about training options and modules. From traditional classroom learning to targeted e-learning seminars, SAP BusinessObjects can offer a training package to suit your learning needs and preferred learning style. http://service.sap.com/ SAP BusinessObjects Data Services Community Get online and timely information about SAP BusinessObjects Data Services, including tips and tricks, http://www.sdn.sap.com/irj/sdn/ds additional downloads, samples, and much more. All content is to and from the community, so feel free to join in and contact us if you have a submission. Forums on SCN (SAP Community Network ) http://forums.sdn.sap.com/forum.jspa?foru mID=305 Blueprints http://www.sdn.sap.com/irj/boc/blueprints Product documentation Search the SAP BusinessObjects forums on the SAP Community Network to learn from other SAP BusinessObjects Data Services users and start posting questions or share your knowledge with the community. Blueprints for you to download and modify to fit your needs. Each blueprint contains the necessary SAP BusinessObjects Data Services project, jobs, data flows, file formats, sample data, template tables, and custom functions to run the data flows in your environment with only a few modifications. SAP BusinessObjects product documentation. http://help.sap.com/businessobjects/ Supported Platforms (Product Availability Matrix) https://service.sap.com/PAM Get information about supported platforms for SAP BusinessObjects Data Services. Use the search function to search for Data Services. Click the link for the version of Data Services you are searching for. 1.2 Overview of this guide 24 2011-06-09
- 25. Introduction Welcome to the Designer Guide. The Data Services Designer provides a graphical user interface (GUI) development environment in which you define data application logic to extract, transform, and load data from databases and applications into a data warehouse used for analytic and on-demand queries. You can also use the Designer to define logical paths for processing message-based queries and transactions from Web-based, front-office, and back-office applications. 1.2.1 About this guide The guide contains two kinds of information: • Conceptual information that helps you understand the Data Services Designer and how it works • Procedural information that explains in a step-by-step manner how to accomplish a task You will find this guide most useful: • While you are learning about the product • While you are performing tasks in the design and early testing phase of your data-movement projects • As a general source of information during any phase of your projects 1.2.2 Who should read this guide This and other Data Services product documentation assumes the following: • You are an application developer, consultant, or database administrator working on data extraction, data warehousing, data integration, or data quality. • You understand your source data systems, RDBMS, business intelligence, and messaging concepts. • You understand your organization's data needs. • You are familiar with SQL (Structured Query Language). • If you are interested in using this product to design real-time processing, you should be familiar with: • • • 25 DTD and XML Schema formats for XML files Publishing Web Services (WSDL, HTTP, and SOAP protocols, etc.) You are familiar Data Services installation environments—Microsoft Windows or UNIX. 2011-06-09
- 27. Logging into the Designer Logging into the Designer You must have access to a local repository to log into the software. Typically, you create a repository during installation. However, you can create a repository at any time using the Repository Manager, and configure access rights within the Central Management Server. Additionally, each repository must be associated with at least one Job Server before you can run repository jobs from within the Designer. Typically, you define a Job Server and associate it with a repository during installation. However, you can define or edit Job Servers or the links between repositories and Job Servers at any time using the Server Manager. When you log in to the Designer, you must log in as a user defined in the Central Management Server (CMS). 1. Enter your user credentials for the CMS. • System Specify the server name and optionally the port for the CMS. • User name Specify the user name to use to log into CMS. • Password Specify the password to use to log into the CMS. • Authentication Specify the authentication type used by the CMS. 2. Click Log on. The software attempts to connect to the CMS using the specified information. When you log in successfully, the list of local repositories that are available to you is displayed. 3. Select the repository you want to use. 4. If you want the software to remember connection information for future use, click Remember. If you choose this option, your CMS connection information and repository selection are encrypted and stored locally, and will be filled in automatically the next time you log into the Designer. 5. Click OK to log in using the selected repository. 2.1 Version restrictions 27 2011-06-09
- 28. Logging into the Designer Your repository version must be associated with the same major release as the Designer and must be less than or equal to the version of the Designer. During login, the software alerts you if there is a mismatch between your Designer version and your repository version. After you log in, you can view the software and repository versions by selecting Help > About Data Services. Some features in the current release of the Designer might not be supported if you are not logged in to the latest version of the repository. 2.2 Resetting users Occasionally, more than one person may attempt to log in to a single repository. If this happens, the Reset Users window appears, listing the users and the time they logged in to the repository. From this window, you have several options. You can: • Reset Users to clear the users in the repository and set yourself as the currently logged in user. • Continue to log in to the system regardless of who else might be connected. • Exit to terminate the login attempt and close the session. Note: Only use Reset Users or Continue if you know that you are the only user connected to the repository. Subsequent changes could corrupt the repository. 28 2011-06-09
- 29. Designer User Interface Designer User Interface This section provides basic information about the Designer's graphical user interface. 3.1 Objects All "entities" you define, edit, or work with in Designer are called objects. The local object library shows objects such as source and target metadata, system functions, projects, and jobs. Objects are hierarchical and consist of: • Options, which control the operation of objects. For example, in a datastore, the name of the database to which you connect is an option for the datastore object. • Properties, which document the object. For example, the name of the object and the date it was created are properties. Properties describe an object, but do not affect its operation. The software has two types of objects: Reusable and single-use. The object type affects how you define and retrieve the object. 3.1.1 Reusable objects You can reuse and replicate most objects defined in the software. After you define and save a reusable object, the software stores the definition in the local repository. You can then reuse the definition as often as necessary by creating calls to the definition. Access reusable objects through the local object library. A reusable object has a single definition; all calls to the object refer to that definition. If you change the definition of the object in one place, you are changing the object in all other places in which it appears. A data flow, for example, is a reusable object. Multiple jobs, like a weekly load job and a daily load job, can call the same data flow. If the data flow changes, both jobs use the new version of the data flow. The object library contains object definitions. When you drag and drop an object from the object library, you are really creating a new reference (or call) to the existing object definition. 29 2011-06-09
- 30. Designer User Interface 3.1.2 Single-use objects Some objects are defined only within the context of a single job or data flow, for example scripts and specific transform definitions. 3.1.3 Object hierarchy Object relationships are hierarchical. The following figure shows the relationships between major object types: 30 2011-06-09
- 31. Designer User Interface 3.2 Designer window The Designer user interface consists of a single application window and several embedded supporting windows. 31 2011-06-09
- 32. Designer User Interface In addition to the Menu bar and Toolbar, there are other key areas of the application window: Area Description Project area Contains the current project (and the job(s) and other objects within it) available to you at a given time. In the software, all entities you create, modify, or work with are objects. Workspace The area of the application window in which you define, display, and modify objects. Local object library Provides access to local repository objects including built-in system objects, such as transforms, and the objects you build and save, such as jobs and data flows. Tool palette Buttons on the tool palette enable you to add new objects to the workspace. 3.3 Menu bar This section contains a brief description of the Designer's menus. 32 2011-06-09
- 33. Designer User Interface 3.3.1 Project menu The project menu contains standard Windows as well as software-specific options. Option Description New Define a new project, batch job, real-time job, work flow, data flow, transform, datastore, file format, DTD, XML Schema, or custom function. Open Open an existing project. Close Close the currently open project. Delete Delete the selected object. Save Save the object open in the workspace. Save All Save all changes to objects in the current Designer session. Print Print the active workspace. Print Setup Set up default printer information. Compact Reposi- Remove redundant and obsolete objects from the repository tables. tory Exit Exit Designer. 3.3.2 Edit menu The Edit menu provides standard Windows commands with a few restrictions. Option Undo Undo the last operation. Cut Cut the selected objects or text and place it on the clipboard. Copy Copy the selected objects or text to the clipboard. Paste Paste the contents of the clipboard into the active workspace or text box. Delete 33 Description Delete the selected objects. 2011-06-09
- 34. Designer User Interface Option Description Recover Last Deleted Recover deleted objects to the workspace from which they were deleted. Only the most recently deleted objects are recovered. Select All Select all objects in the active workspace. Clear All Clear all objects in the active workspace (no undo). 3.3.3 View menu A check mark indicates that the tool is active. Option Description Toolbar Display or remove the toolbar in the Designer window. Status Bar Display or remove the status bar in the Designer window. Palette Display or remove the floating tool palette. Enabled Descriptions View descriptions for objects with enabled descriptions. Refresh Redraw the display. Use this command to ensure the content of the workspace represents the most up-to-date information from the repository. 3.3.4 Tools menu An icon with a different color background indicates that the tool is active. Option Object Library Open or close the object library window. Project Area Display or remove the project area from the Designer window. Variables Open or close the Variables and Parameters window. Output Open or close the Output window. The Output window shows errors that occur such as during job validation or object export. Profiler Monitor 34 Description Display the status of Profiler tasks. 2011-06-09
- 35. Designer User Interface Option Description Run Match Wizard Display the Match Wizard to create a match data flow. Select a transform in a data flow to activate this menu item. The transform(s) that the Match Wizard generates will be placed downstream from the transform you selected. Match Editor Display the Match Editor to edit Match transform options. Associate Editor Display the Associate Editor to edit Associate transform options. User-Defined Editor Display the User-Defined Editor to edit User-Defined transform options. Custom Functions Display the Custom Functions window. System Configurations Display the System Configurations editor. Substitution Parameter Configurations Display the Substitution Parameter Editor to create and edit substitution paramters and configurations. Profiler Server Login Connect to the Profiler Server. Export Export individual repository objects to another repository or file. This command opens the Export editor in the workspace. You can drag objects from the object library into the editor for export. To export your whole repository, in the object library right-click and select Repository > Export to file. Import From File Import objects into the current repository from a file. The default file types are ATL, XML, DMT, and FMT. For more information on DMT and FMT files, see the Upgrade Guide. Metadata Exchange Import and export metadata to third-party systems via a file. BusinessObjects Universes Export (create or update) metadata in BusinessObjects Universes. Central Repositories Create or edit connections to a central repository for managing object versions among multiple users. Options Display the Options window. Data Services Management Display the Management Console. Console Related Topics • Multi-user Environment Setup • Administrator's Guide: Export/Import, Importing from a file • Administrator's Guide: Export/Import, Exporting/importing objects • Reference Guide: Functions and Procedures, Custom functions • Local object library • Project area • Variables and Parameters • Using the Data Profiler • Creating and managing multiple datastore configurations 35 2011-06-09
- 36. Designer User Interface • Connecting to the profiler server • Metadata exchange • Creating SAP universes • General and environment options 3.3.5 Debug menu The only options available on this menu at all times are Show Filters/Breakpoints and Filters/Breakpoints. The Execute and Start Debug options are only active when a job is selected. All other options are available as appropriate when a job is running in the Debug mode. Option Description Execute Opens the Execution Properties window which allows you to execute the selected job. Start Debug Opens the Debug Properties window which allows you to run a job in the debug mode. Show Filters/Breakpoints Shows and hides filters and breakpoints in workspace diagrams. Filters/Breakpoints Opens a window you can use to manage filters and breakpoints. Related Topics • Using the interactive debugger • Filters and Breakpoints window 3.3.6 Validation menu The Designer displays options on this menu as appropriate when an object is open in the workspace. 36 2011-06-09
- 37. Designer User Interface Option Description Validate Validate the objects in the current workspace view or all objects in the job before executing the application. Show ATL View a read-only version of the language associated with the job. Display Optimized SQL Display the SQL that Data Services generated for a selected data flow. Related Topics • Performance Optimization Guide: Maximizing Push-Down Operations, To view SQL 3.3.7 Dictionary menu The Dictionary menu contains options for interacting with the dictionaries used by cleansing packages and the Data Cleanse transform. Option Description Search Search for existing dictionary entries. Add New Dictionary En- Create a new primary dictionary entry. try Bulk Load Import a group of dictionary changes from an external file. View Bulk Load Conflict Display conflict logs generated by the Bulk Load feature. Logs Export Dictionary Changes Export changes from a dictionary to an XML file. Universal Data Cleanse Dictionary-related options specific to the Universal Data Cleanse feature. Add New Classification Add a new dictionary classification. Edit Classification Add Custom Output 37 Edit an existing dictionary classification. Add custom output categories and fields to a dictionary. 2011-06-09
- 38. Designer User Interface Option Description Create Dictionary Create a new dictionary in the repository. Delete Dictionary Delete a dictionary from the repository. Manage Connection Update the connection information for the dictionary repository connection. 3.3.8 Window menu The Window menu provides standard Windows options. Option Description Back Move back in the list of active workspace windows. Forward Move forward in the list of active workspace windows. Cascade Display window panels overlapping with titles showing. Tile Horizontally Display window panels side by side. Tile Vertically Display window panels one above the other. Close All Windows Close all open windows. A list of objects open in the workspace also appears on the Windows menu. The name of the currently-selected object is indicated by a check mark. Navigate to another open object by selecting its name in the list. 3.3.9 Help menu The Help menu provides standard help options. 38 2011-06-09
- 39. Designer User Interface Option Description Release Notes Displays the Release Notes for this release. What's New Displays a summary of new features for this release. Technical Manuals Displays the Technical Manuals CHM file, a compilation of many of the Data Services technical documents. You can also access the same documentation from the <LINKDIR>DocBooks directory. Tutorial Displays the Data Services Tutorial, a step-by-step introduction to using SAP BusinessObjects Data Services. Data Services Community Get online and timely information about SAP BusinessObjects Data Services, including tips and tricks, additional downloads, samples, and much more. All content is to and from the community, so feel free to join in and contact us if you have a submission. Forums on SCN (SAP Com- Search the SAP BusinessObjects forums on the SAP Community Network to learn from other SAP BusinessObjects Data Services users munity Network) and start posting questions or share your knowledge with the community. Blueprints Blueprints for you to download and modify to fit your needs. Each blueprint contains the necessary SAP BusinessObjects Data Services project, jobs, data flows, file formats, sample data, template tables, and custom functions to run the data flows in your environment with only a few modifications. Show Start Page Displays the home page of the Data ServicesDesigner. About Data Services Display information about the software including versions of the Design er, Job Server and engine, and copyright information. 3.4 Toolbar In addition to many of the standard Windows tools, the software provides application-specific tools, including: Icon Description Close all windows Closes all open windows in the workspace. Local Object Library 39 Tool Opens and closes the local object library window. 2011-06-09
- 40. Designer User Interface Icon Description Central Object Library Opens and closes the central object library window. Variables Opens and closes the variables and parameters creation window. Project Area Opens and closes the project area. Output Opens and closes the output window. View Enabled Descriptions Enables the system level setting for viewing object descriptions in the workspace. Validate Current View Validates the object definition open in the workspace. Other objects included in the definition are also validated. Validate All Objects in View Validates the object definition open in the workspace. Objects included in the definition are also validated. Audit Objects in Data Flow Opens the Audit window to define audit labels and rules for the data flow. View Where Used Opens the Output window, which lists parent objects (such as jobs) of the object currently open in the workspace (such as a data flow). Use this command to find other jobs that use the same data flow, before you decide to make design changes. To see if an object in a data flow is reused elsewhere, rightclick one and select View Where Used. Go Back Move back in the list of active workspace windows. Go Forward Move forward in the list of active workspace windows. Management Console Opens and closes the Management Console window. Contents 40 Tool Opens the Technical Manuals PDF for information about using the software. 2011-06-09
- 41. Designer User Interface Use the tools to the right of the About tool with the interactive debugger. Related Topics • Debug menu options and tool bar 3.5 Project area The project area provides a hierarchical view of the objects used in each project. Tabs on the bottom of the project area support different tasks. Tabs include: Create, view and manage projects. Provides a hierarchical view of all objects used in each project. View the status of currently executing jobs. Selecting a specific job execution displays its status, including which steps are complete and which steps are executing. These tasks can also be done using the Administrator. View the history of complete jobs. Logs can also be viewed with the Administrator. To control project area location, right-click its gray border and select/deselect Allow Docking, or select Hide from the menu. • When you select Allow Docking, you can click and drag the project area to dock at and undock from any edge within the Designer window. When you drag the project area away from a Designer window edge, it stays undocked. To quickly switch between your last docked and undocked locations, just double-click the gray border. When you deselect Allow Docking, you can click and drag the project area to any location on your screen and it will not dock inside the Designer window. • When you select Hide, the project area disappears from the Designer window. To unhide the project area, click its toolbar icon. Here's an example of the Project window's Designer tab, which shows the project hierarchy: 41 2011-06-09
- 42. Designer User Interface As you drill down into objects in the Designer workspace, the window highlights your location within the project hierarchy. 3.6 Tool palette The tool palette is a separate window that appears by default on the right edge of the Designer workspace. You can move the tool palette anywhere on your screen or dock it on any edge of the De signer window. The icons in the tool palette allow you to create new objects in the workspace. The icons are disabled when they are not allowed to be added to the diagram open in the workspace. To show the name of each icon, hold the cursor over the icon until the tool tip for the icon appears, as shown. When you create an object from the tool palette, you are creating a new definition of an object. If a new object is reusable, it will be automatically available in the object library after you create it. For example, if you select the data flow icon from the tool palette and define a new data flow, later you can drag that existing data flow from the object library, adding a call to the existing definition. The tool palette contains the following icons: Icon Description (class) Available Pointer Returns the tool pointer to a selection pointer for selecting and moving objects in a diagram. Everywhere Work flow Creates a new work flow. (reusable) Jobs and work flows Data flow 42 Tool Creates a new data flow. (reusable) Jobs and work flows 2011-06-09
- 43. Designer User Interface Icon Tool Description (class) Available ABAP data flow Used only with the SAP application. Query transform Creates a template for a query. Use it to define column mappings and row selections. (single-use) Data flows Template table Creates a table for a target. (singleuse) Data flows Template XML Creates an XML template. (single-use) Data flows Data transport Used only with the SAP application. Script Creates a new script object. (singleuse) Jobs and work flows Conditional Creates a new conditional object. (single-use) Jobs and work flows Try Creates a new try object. (single-use) Jobs and work flows Catch Creates a new catch object. (singleuse) Jobs and work flows Annotation Creates an annotation. (single-use) Jobs, work flows, and data flows 3.7 Designer keyboard accessibility The following keys are available for navigation in Designer. All dialogs and views support these keys. To Enter edit mode. F2 Close a menu or dialog box or cancel an operation in progress. ESC Close the current window. CTRL+F4 Cycle through windows one window at a time. CTRL+TAB Display a system menu for the application window. 43 Press ALT+SPACEBAR 2011-06-09
- 44. Designer User Interface To Press Move to the next page of a property sheet. CTRL+PAGE DOWN Move to the previous page of a property sheet. CTRL+PAGE UP Move to the next control on a view or dialog. TAB Move to the previous control on a view or dialog. SHIFT+TAB Press a button when focused. ENTER or SPACE Enable the context menu (right-click mouse operations). SHIFT+F10 or Menu Key Expand or collapse a tree (+). Right Arrow or Left Arrow Move up and down a tree. Up Arrow or Down Arrow Show focus. ALT Hot Key operations. ALT+<LETTER> 3.8 Workspace When you open or select a job or any flow within a job hierarchy, the workspace becomes "active" with your selection. The workspace provides a place to manipulate system objects and graphically assemble data movement processes. These processes are represented by icons that you drag and drop into a workspace to create a workspace diagram. This diagram is a visual representation of an entire data movement application or some part of a data movement application. 3.8.1 Moving objects in the workspace area Use standard mouse commands to move objects in the workspace. 44 2011-06-09
- 45. Designer User Interface To move an object to a different place in the workspace area: 1. Click to select the object. 2. Drag the object to where you want to place it in the workspace. 3.8.2 Connecting objects You specify the flow of data through jobs and work flows by connecting objects in the workspace from left to right in the order you want the data to be moved. To connect objects: 1. Place the objects you want to connect in the workspace. 2. Click and drag from the triangle on the right edge of an object to the triangle on the left edge of the next object in the flow. 3.8.3 Disconnecting objects To disconnect objects 1. Click the connecting line. 2. Press the Delete key. 3.8.4 Describing objects You can use descriptions to add comments about objects. You can use annotations to explain a job, work flow, or data flow. You can view object descriptions and annotations in the workspace. Together, descriptions and annotations allow you to document an SAP BusinessObjects Data Services application. For example, you can describe the incremental behavior of individual jobs with numerous annotations and label each object with a basic description. This job loads current categories and expenses and produces tables for analysis. Related Topics • Creating descriptions • Creating annotations 45 2011-06-09
- 46. Designer User Interface 3.8.5 Scaling the workspace You can control the scale of the workspace. By scaling the workspace, you can change the focus of a job, work flow, or data flow. For example, you might want to increase the scale to examine a particular part of a work flow, or you might want to reduce the scale so that you can examine the entire work flow without scrolling. To change the scale of the workspace 1. In the drop-down list on the tool bar, select a predefined scale or enter a custom value (for example, 100%). 2. Alternatively, right-click in the workspace and select a desired scale. Note: You can also select Scale to Fit and Scale to Whole: • Select Scale to Fit and the Designer calculates the scale that fits the entire project in the current view area. • Select Scale to Whole to show the entire workspace area in the current view area. 3.8.6 Arranging workspace windows The Window menu allows you to arrange multiple open workspace windows in the following ways: cascade, tile horizontally, or tile vertically. 3.8.7 Closing workspace windows When you drill into an object in the project area or workspace, a view of the object's definition opens in the workspace area. The view is marked by a tab at the bottom of the workspace area, and as you open more objects in the workspace, more tabs appear. (You can show/hide these tabs from the Tools > Options menu. Go to Designer > General options and select/deselect Show tabs in workspace.) Note: These views use system resources. If you have a large number of open views, you might notice a decline in performance. Close the views individually by clicking the close box in the top right corner of the workspace. Close all open views by selecting Window > Close All Windows or clicking the Close All Windows icon on the toolbar. 46 2011-06-09
- 47. Designer User Interface Related Topics • General and environment options 3.9 Local object library The local object library provides access to reusable objects. These objects include built-in system objects, such as transforms, and the objects you build and save, such as datastores, jobs, data flows, and work flows. The local object library is a window into your local repository and eliminates the need to access the repository directly. Updates to the repository occur through normal software operation. Saving the objects you create adds them to the repository. Access saved objects through the local object library. To control object library location, right-click its gray border and select/deselect Allow Docking, or select Hide from the menu. • When you select Allow Docking, you can click and drag the object library to dock at and undock from any edge within the Designer window. When you drag the object library away from a Designer window edge, it stays undocked. To quickly switch between your last docked and undocked locations, just double-click the gray border. When you deselect Allow Docking, you can click and drag the object library to any location on your screen and it will not dock inside the Designer window. • When you select Hide, the object library disappears from the Designer window. To unhide the object library, click its toolbar icon. Related Topics • Central versus local repository 3.9.1 To open the object library • Choose Tools > Object Library, or click the object library icon in the icon bar. The object library gives you access to the object types listed in the following table. The table shows the tab on which the object type appears in the object library and describes the context in which you can use each type of object. 47 2011-06-09
- 48. Designer User Interface Tab Description Projects are sets of jobs available at a given time. Jobs are executable work flows. There are two job types: batch jobs and real-time jobs. Work flows order data flows and the operations that support data flows, defining the interdependencies between them. Data flows describe how to process a task. Transforms operate on data, producing output data sets from the sources you specify. The object library lists both built-in and custom transforms. Datastores represent connections to databases and applications used in your project. Under each datastore is a list of the tables, documents, and functions imported into the software. Formats describe the structure of a flat file, XML file, or XML message. Custom Functions are functions written in the software's Scripting Language. You can use them in your jobs. 3.9.2 To display the name of each tab as well as its icon 1. Make the object library window wider until the names appear. or 2. Hold the cursor over the tab until the tool tip for the tab appears. 3.9.3 To sort columns in the object library 48 2011-06-09
- 49. Designer User Interface • Click the column heading. For example, you can sort data flows by clicking the Data Flow column heading once. Names are listed in ascending order. To list names in descending order, click the Data Flow column heading again. 3.10 Object editors To work with the options for an object, in the workspace click the name of the object to open its editor. The editor displays the input and output schemas for the object and a panel below them listing options set for the object. If there are many options, they are grouped in tabs in the editor. A schema is a data structure that can contain columns, other nested schemas, and functions (the contents are called schema elements). A table is a schema containing only columns. In an editor, you can: • Undo or redo previous actions performed in the window (right-click and choose Undo or Redo) • Find a string in the editor (right-click and choose Find) • Drag-and-drop column names from the input schema into relevant option boxes • Use colors to identify strings and comments in text boxes where you can edit expressions (keywords appear blue; strings are enclosed in quotes and appear pink; comments begin with a pound sign and appear green) Note: You cannot add comments to a mapping clause in a Query transform. For example, the following syntax is not supported on the Mapping tab: table.column # comment The job will not run and you cannot successfully export it. Use the object description or workspace annotation feature instead. Related Topics • Query Editor • Data Quality transform editors 3.11 Working with objects 49 2011-06-09
- 50. Designer User Interface This section discusses common tasks you complete when working with objects in the Designer. With these tasks, you use various parts of the Designer—the toolbar, tool palette, workspace, and local object library. 3.11.1 Creating new reusable objects You can create reusable objects from the object library or by using the tool palette. After you create an object, you can work with the object, editing its definition and adding calls to other objects. 3.11.1.1 To create a reusable object (in the object library) 1. Open the object library by choosing Tools > Object Library. 2. Click the tab corresponding to the object type. 3. Right-click anywhere except on existing objects and choose New. 4. Right-click the new object and select Properties. Enter options such as name and description to define the object. 3.11.1.2 To create a reusable object (using the tool palette) 1. In the tool palette, left-click the icon for the object you want to create. 2. Move the cursor to the workspace and left-click again. The object icon appears in the workspace where you have clicked. 3.11.1.3 To open an object's definition You can open an object's definition in one of two ways: 1. From the workspace, click the object name. The software opens a blank workspace in which you define the object. 2. From the project area, click the object. You define an object using other objects. For example, if you click the name of a batch data flow, a new workspace opens for you to assemble sources, targets, and transforms that make up the actual flow. 50 2011-06-09
- 51. Designer User Interface 3.11.1.4 To add an existing object (create a new call to an existing object) 1. Open the object library by choosing Tools > Object Library. 2. Click the tab corresponding to any object type. 3. Select an object. 4. Drag the object to the workspace. Note: Objects dragged into the workspace must obey the hierarchy logic. For example, you can drag a data flow into a job, but you cannot drag a work flow into a data flow. Related Topics • Object hierarchy 3.11.2 Changing object names You can change the name of an object from the workspace or the object library. You can also create a copy of an existing object. Note: You cannot change the names of built-in objects. 1. To change the name of an object in the workspace a. Click to select the object in the workspace. b. Right-click and choose Edit Name. c. Edit the text in the name text box. d. Click outside the text box or press Enter to save the new name. 2. To change the name of an object in the object library a. Select the object in the object library. b. Right-click and choose Properties. c. Edit the text in the first text box. d. Click OK. 3. To copy an object a. Select the object in the object library. b. Right-click and choose Replicate. c. The software makes a copy of the top-level object (but not of objects that it calls) and gives it a new name, which you can edit. 51 2011-06-09
- 52. Designer User Interface 3.11.3 Viewing and changing object properties You can view (and, in some cases, change) an object's properties through its property page. 3.11.3.1 To view, change, and add object properties 1. Select the object in the object library. 2. Right-click and choose Properties. The General tab of the Properties window opens. 3. Complete the property sheets. The property sheets vary by object type, but General, Attributes and Class Attributes are the most common and are described in the following sections. 4. When finished, click OK to save changes you made to the object properties and to close the window. Alternatively, click Apply to save changes without closing the window. 3.11.3.2 General tab The General tab contains two main object properties: name and description. From the General tab, you can change the object name as well as enter or edit the object description. You can add object descriptions to single-use objects as well as to reusable objects. Note that you can toggle object descriptions on and off by right-clicking any object in the workspace and selecting/clearing View Enabled Descriptions. Depending on the object, other properties may appear on the General tab. Examples include: • • • • • Execute only once Recover as a unit Degree of parallelism Use database links Cache type Related Topics • Performance Optimization Guide: Using Caches • Linked datastores • Performance Optimization Guide: Using Parallel Execution • Recovery Mechanisms • Creating and defining data flows 52 2011-06-09
- 53. Designer User Interface 3.11.3.3 Attributes tab The Attributes tab allows you to assign values to the attributes of the current object. To assign a value to an attribute, select the attribute and enter the value in the Value box at the bottom of the window. Some attribute values are set by the software and cannot be edited. When you select an attribute with a system-defined value, the Value field is unavailable. 3.11.3.4 Class Attributes tab The Class Attributes tab shows the attributes available for the type of object selected. For example, all data flow objects have the same class attributes. To create a new attribute for a class of objects, right-click in the attribute list and select Add. The new attribute is now available for all of the objects of this class. To delete an attribute, select it then right-click and choose Delete. You cannot delete the class attributes predefined by Data Services. 3.11.4 Creating descriptions Use descriptions to document objects. You can see descriptions on workspace diagrams. Therefore, descriptions are a convenient way to add comments to workspace objects. A description is associated with a particular object. When you import or export that repository object (for example, when migrating between development, test, and production environments), you also import or export its description. The Designer determines when to show object descriptions based on a system-level setting and an object-level setting. Both settings must be activated to view the description for a particular object. The system-level setting is unique to your setup. The system-level setting is disabled by default. To activate that system-level setting, select ViewEnabled Descriptions, or click the View Enabled Descriptions button on the toolbar. The object-level setting is saved with the object in the repository. The object-level setting is also disabled by default unless you add or edit a description from the workspace. To activate the object-level setting, right-click the object and select Enable object description. 53 2011-06-09
- 54. Designer User Interface An ellipses after the text in a description indicates that there is more text. To see all the text, resize the description by clicking and dragging it. When you move an object, its description moves as well. To see which object is associated with which selected description, view the object's name in the status bar. 3.11.4.1 To add a description to an object 1. In the project area or object library, right-click an object and select Properties. 2. Enter your comments in the Description text box. 3. Click OK. The description for the object displays in the object library. 3.11.4.2 To display a description in the workspace 1. In the project area, select an existing object (such as a job) that contains an object to which you have added a description (such as a work flow). 2. From the View menu, select Enabled Descriptions. Alternately, you can select the View Enabled Descriptions button on the toolbar. 3. Right-click the work flow and select Enable Object Description. The description displays in the workspace under the object. 3.11.4.3 To add a description to an object from the workspace 1. From the View menu, select Enabled Descriptions. 2. In the workspace, right-click an object and select Properties. 3. In the Properties window, enter text in the Description box. 4. Click OK. The description displays automatically in the workspace (and the object's Enable Object Description option is selected). 54 2011-06-09
- 55. Designer User Interface 3.11.4.4 To hide a particular object's description 1. In the workspace diagram, right-click an object. Alternately, you can select multiple objects by: • Pressing and holding the Control key while selecting objects in the workspace diagram, then right-clicking one of the selected objects. • Dragging a selection box around all the objects you want to select, then right-clicking one of the selected objects. 2. In the pop-up menu, deselect Enable Object Description. The description for the object selected is hidden, even if the View Enabled Descriptions option is checked, because the object-level switch overrides the system-level switch. 3.11.4.5 To edit object descriptions 1. In the workspace, double-click an object description. 2. Enter, cut, copy, or paste text into the description. 3. In the Project menu, select Save. Alternately, you can right-click any object and select Properties to open the object's Properties window and add or edit its description. Note: If you attempt to edit the description of a reusable object, the software alerts you that the description will be updated for every occurrence of the object, across all jobs. You can select the Do not show me this again check box to avoid this alert. However, after deactivating the alert, you can only reactivate the alert by calling Technical Support. 3.11.5 Creating annotations Annotations describe a flow, part of a flow, or a diagram in a workspace. An annotation is associated with the job, work flow, or data flow where it appears. When you import or export that job, work flow, or data flow, you import or export associated annotations. 55 2011-06-09
- 56. Designer User Interface 3.11.5.1 To annotate a workspace diagram 1. Open the workspace diagram you want to annotate. You can use annotations to describe any workspace such as a job, work flow, data flow, catch, conditional, or while loop. 2. In the tool palette, click the annotation icon. 3. Click a location in the workspace to place the annotation. An annotation appears on the diagram. You can add, edit, and delete text directly on the annotation. In addition, you can resize and move the annotation by clicking and dragging. You can add any number of annotations to a diagram. 3.11.5.2 To delete an annotation 1. Right-click an annotation. 2. Select Delete. Alternately, you can select an annotation and press the Delete key. 3.11.6 Copying objects Objects can be cut or copied and then pasted on the workspace where valid. Multiple objects can be copied and pasted either within the same or other data flows, work flows, or jobs. Additionally, calls to data flows and works flows can be cut or copied and then pasted to valid objects in the workspace. References to global variables, local variables, parameters, and substitution parameters are copied; however, you must be define each within its new context. Note: The paste operation duplicates the selected objects in a flow, but still calls the original objects. In other words, the paste operation uses the original object in another location. The replicate operation creates a new object in the object library. To cut or copy and then paste objects: 1. In the workspace, select the objects you want to cut or copy. 56 2011-06-09
- 57. Designer User Interface You can select multiple objects using Ctrl-click, Shift-click, or Ctrl+A. 2. Right-click and then select either Cut or Copy. 3. Click within the same flow or select a different flow. Right-click and select Paste. Where necessary to avoid a naming conflict, a new name is automatically generated. Note: The objects are pasted in the selected location if you right-click and select Paste. The objects are pasted in the upper left-hand corner of the workspace if you paste using any of the following methods: • cIick the Paste icon. • click Edit > Paste. • use the Ctrl+V keyboard short-cut. If you use a method that pastes the objects to the upper left-hand corner, subsequent pasted objects are layered on top of each other. 3.11.7 Saving and deleting objects "Saving" an object in the software means storing the language that describes the object to the repository. You can save reusable objects; single-use objects are saved only as part of the definition of the reusable object that calls them. You can choose to save changes to the reusable object currently open in the workspace. When you save the object, the object properties, the definitions of any single-use objects it calls, and any calls to other reusable objects are recorded in the repository. The content of the included reusable objects is not saved; only the call is saved. The software stores the description even if the object is not complete or contains an error (does not validate). 3.11.7.1 To save changes to a single reusable object 1. Open the project in which your object is included. 2. Choose Project > Save. This command saves all objects open in the workspace. Repeat these steps for other individual objects you want to save. 57 2011-06-09
- 58. Designer User Interface 3.11.7.2 To save all changed objects in the repository 1. Choose Project > Save All. The software lists the reusable objects that were changed since the last save operation. 2. (optional) Deselect any listed object to avoid saving it. 3. Click OK. Note: The software also prompts you to save all objects that have changes when you execute a job and when you exit the Designer. Saving a reusable object saves any single-use object included in it. 3.11.7.3 To delete an object definition from the repository 1. In the object library, select the object. 2. Right-click and choose Delete. • • If you attempt to delete an object that is being used, the software provides a warning message and the option of using the View Where Used feature. If you select Yes, the software marks all calls to the object with a red "deleted" icon to indicate that the calls are invalid. You must remove or replace these calls to produce an executable job. Note: Built-in objects such as transforms cannot be deleted from the object library. Related Topics • Using View Where Used 3.11.7.4 To delete an object call 1. Open the object that contains the call you want to delete. 2. Right-click the object call and choose Delete. 58 2011-06-09
- 59. Designer User Interface If you delete a reusable object from the workspace or from the project area, only the object call is deleted. The object definition remains in the object library. 3.11.8 Searching for objects From within the object library, you can search for objects defined in the repository or objects available through a datastore. 3.11.8.1 To search for an object 1. Right-click in the object library and choose Search. The software displays the Search window. 2. Enter the appropriate values for the search. Options available in the Search window are described in detail following this procedure. 3. Click Search. The objects matching your entries are listed in the window. From the search results window you can use the context menu to: • Open an item • View the attributes (Properties) • Import external tables as repository metadata You can also drag objects from the search results window and drop them in the desired location. The Search window provides you with the following options: Option Description Where to search. Look in Choose from the repository or a specific datastore. When you designate a datastore, you can also choose to search the imported data (Internal Data) or the entire datastore (External Data). 59 2011-06-09
- 60. Designer User Interface Option Description The type of object to find. Object type When searching the repository, choose from Tables, Files, Data flows, Work flows, Jobs, Hierarchies, IDOCs, and Domains. When searching a datastore or application, choose from object types available through that datastore. The object name to find. If you are searching in the repository, the name is not case sensitive. If you are searching in a datastore and the name is case sensitive in that datastore, enter the name as it appears in the database or application and use double quotation marks (") around the name to preserve the case. Name You can designate whether the information to be located Contains the specified name or Equals the specified name using the drop-down box next to the Name field. The object description to find. Description Objects imported into the repository have a description from their source. By default, objects you create in the Designer have no description unless you add a one. The search returns objects whose description attribute contains the value entered. The Search window also includes an Advanced button where, you can choose to search for objects based on their attribute values. You can search by attribute values only when searching in the repository. The Advanced button provides the following options: Option Description Attribute The object attribute in which to search. Value The attribute value to find. The type of search performed. Match 60 Select Contains to search for any attribute that contains the value specified. Select Equals to search for any attribute that contains only the value specified. 2011-06-09
- 61. Designer User Interface 3.12 General and environment options To open the Options window, select Tools > Options. The window displays option groups for Designer, Data, and Job Server options. Expand the options by clicking the plus icon. As you select each option group or option, a description appears on the right. 3.12.1 Designer — Environment Table 3-9: Default Administrator for Metadata Reporting Option Description Administrator Select the Administrator that the metadata reporting tool uses. An Administrator is defined by host name and port. Table 3-10: Default Job Server Option Description Current Displays the current value of the default Job Server. New Allows you to specify a new value for the default Job Server from a drop-down list of Job Servers associated with this repository. Changes are effective immediately. If a repository is associated with several Job Servers, one Job Server must be defined as the default Job Server to use at login. Note: Job-specific options and path names specified in Designer refer to the current default Job Server. If you change the default Job Server, modify these options and path names. 61 2011-06-09
- 62. Designer User Interface Table 3-11: Designer Communication Ports Option Description Allow Designer to set the port for Job Server communication If checked, Designer automatically sets an available port to receive messages from the current Job Server. The default is checked. Uncheck to specify a listening port or port range. Enter port numbers in the port text boxes. To specify a specific listening port, enter the same port number in both the From port and To port text boxes. Changes will not take effect until you restart the software. From Only activated when you deselect the previous control. Allows you to specify a range of ports from which the Designer can choose a listening port. To You may choose to constrain the port used for communication between Designer and Job Server when the two components are separated by a firewall. Interactive Debugger Allows you to set a communication port for the Design er to communicate with a Job Server while running in Debug mode. Server group for local repository If the local repository that you logged in to when you opened the Designer is associated with a server group, the name of the server group appears. Related Topics • Changing the interactive debugger port 3.12.2 Designer — General 62 2011-06-09
- 63. Designer User Interface Option Description View data sampling size Controls the sample size used to display the data in sources and targets (rows) in open data flows in the workspace. View data by clicking the magnifying glass icon on source and target objects. Number of characters in Controls the length of the object names displayed in the workspace. Object workspace icon name names are allowed to exceed this number, but the Designer only displays the number entered here. The default is 17 characters. Maximum schema tree The number of elements displayed in the schema tree. Element names elements to auto expand are not allowed to exceed this number. Enter a number for the Input schema and the Output schema. The default is 100. Default parameters to variables of the same name When you declare a variable at the work-flow level, the software automatically passes the value as a parameter with the same name to a data flow called by a work flow. Automatically import doSelect this check box to automatically import domains when importing a mains table that references a domain. Perform complete validaIf checked, the software performs a complete job validation before running tion before job execution a job. The default is unchecked. If you keep this default setting, you should validate your design manually before job execution. Open monitor on job exeAffects the behavior of the Designer when you execute a job. With this cution option enabled, the Designer switches the workspace to the monitor view during job execution; otherwise, the workspace remains as is. The default is on. Automatically calculate column mappings 63 Calculates information about target tables and columns and the sources used to populate them. The software uses this information for metadata reports such as impact and lineage, auto documentation, or custom reports. Column mapping information is stored in the AL_COLMAP table (ALVW_MAPPING view) after you save a data flow or import objects to or export objects from a repository. If the option is selected, be sure to validate your entire job before saving it because column mapping calculation is sensitive to errors and will skip data flows that have validation problems. 2011-06-09
- 64. Designer User Interface Option Description Show dialog when job is Allows you to choose if you want to see an alert or just read the trace completed: messages. Show tabs in workspace Allows you to decide if you want to use the tabs at the bottom of the workspace to navigate. Exclude non-executable Excludes elements not processed during job execution from exported elements from exported XML documents. For example, Designer workspace display coordinates XML would not be exported. Related Topics • Using View Data • Management Console Guide: Refresh Usage Data tab 3.12.3 Designer — Graphics Choose and preview stylistic elements to customize your workspaces. Using these options, you can easily distinguish your job/work flow design workspace from your data flow design workspace. 64 2011-06-09
- 65. Designer User Interface Option Workspace flow type Line Type Line Thickness Background style Color scheme Use navigation watermark Description Switch between the two workspace flow types (Job/Work Flow and Data Flow) to view default settings. Modify settings for each type using the remaining options. Choose a style for object connector lines. Set the connector line thickness. Choose a plain or tiled background pattern for the selected flow type. Set the background color to blue, gray, or white. Add a watermark graphic to the background of the flow type selected. Note that this option is only available with a plain background style. 3.12.4 Designer — Central Repository Connections Option Description Central Repository ConDisplays the central repository connections and the active central reposinections tory. To activate a central repository, right-click one of the central repository connections listed and select Activate. Reactivate automatically Select if you want the active central repository to be reactivated whenever you log in to the software using the current local repository. 3.12.5 Data — General 65 2011-06-09
- 66. Designer User Interface Option Description Century Change Year Indicates how the software interprets the century for two-digit years. Twodigit years greater than or equal to this value are interpreted as 19##. Two-digit years less than this value are interpreted as 20##. The default value is 15. For example, if the Century Change Year is set to 15: Two-digit year 99 1999 16 1916 15 1915 14 Convert blanks to nulls for Oracle bulk loader Interpreted as 2014 Converts blanks to NULL values when loading data using the Oracle bulk loader utility and: • the column is not part of the primary key • the column is nullable 3.12.6 Job Server — Environment Option Description Maximum number of engine processes Sets a limit on the number of engine processes that this Job Server can have running concurrently. 3.12.7 Job Server — General Use this window to reset Job Server options or with guidance from SAP Technical customer Support. Related Topics • Changing Job Server options 66 2011-06-09
- 67. Projects and Jobs Projects and Jobs Project and job objects represent the top two levels of organization for the application flows you create using the Designer. 4.1 Projects A project is a reusable object that allows you to group jobs. A project is the highest level of organization offered by the software. Opening a project makes one group of objects easily accessible in the user interface. You can use a project to group jobs that have schedules that depend on one another or that you want to monitor together. Projects have common characteristics: • Projects are listed in the object library. • Only one project can be open at a time. • Projects cannot be shared among multiple users. 4.1.1 Objects that make up a project The objects in a project appear hierarchically in the project area. If a plus sign (+) appears next to an object, expand it to view the lower-level objects contained in the object. The software shows you the contents as both names in the project area hierarchy and icons in the workspace. In the following example, the Job_KeyGen job contains two data flows, and the DF_EmpMap data flow contains multiple objects. 67 2011-06-09
- 68. Projects and Jobs Each item selected in the project area also displays in the workspace: 4.1.2 Creating a new project 1. Choose Project > New > Project. 2. Enter the name of your new project. The name can include alphanumeric characters and underscores (_). It cannot contain blank spaces. 3. Click Create. The new project appears in the project area. As you add jobs and other lower-level objects to the project, they also appear in the project area. 4.1.3 Opening existing projects 4.1.3.1 To open an existing project 68 2011-06-09
- 69. Projects and Jobs 1. Choose Project > Open. 2. Select the name of an existing project from the list. 3. Click Open. Note: If another project was already open, the software closes that project and opens the new one. 4.1.4 Saving projects 4.1.4.1 To save all changes to a project 1. Choose Project > Save All. The software lists the jobs, work flows, and data flows that you edited since the last save. 2. (optional) Deselect any listed object to avoid saving it. 3. Click OK. Note: The software also prompts you to save all objects that have changes when you execute a job and when you exit the Designer. Saving a reusable object saves any single-use object included in it. 4.2 Jobs A job is the only object you can execute. You can manually execute and test jobs in development. In production, you can schedule batch jobs and set up real-time jobs as services that execute a process when the software receives a message request. A job is made up of steps you want executed together. Each step is represented by an object icon that you place in the workspace to create a job diagram. A job diagram is made up of two or more objects connected together. You can include any of the following objects in a job definition: • Data flows • • Targets • 69 Sources Transforms 2011-06-09
- 70. Projects and Jobs • Work flows • Scripts • Conditionals • While Loops • Try/catch blocks If a job becomes complex, organize its content into individual work flows, then create a single job that calls those work flows. Real-time jobs use the same components as batch jobs. You can add work flows and data flows to both batch and real-time jobs. When you drag a work flow or data flow icon into a job, you are telling the software to validate these objects according the requirements of the job type (either batch or real-time). There are some restrictions regarding the use of some software features with real-time jobs. Related Topics • Work Flows • Real-time Jobs 4.2.1 Creating jobs 4.2.1.1 To create a job in the project area 1. In the project area, select the project name. 2. Right-click and choose New BatchJob or Real Time Job. 3. Edit the name. The name can include alphanumeric characters and underscores (_). It cannot contain blank spaces. The software opens a new workspace for you to define the job. 4.2.1.2 To create a job in the object library 1. Go to the Jobs tab. 70 2011-06-09
- 71. Projects and Jobs 2. Right-click Batch Jobs or Real Time Jobs and choose New. 3. A new job with a default name appears. 4. Right-click and select Properties to change the object's name and add a description. The name can include alphanumeric characters and underscores (_). It cannot contain blank spaces. 5. To add the job to the open project, drag it into the project area. 4.2.2 Naming conventions for objects in jobs We recommend that you follow consistent naming conventions to facilitate object identification across all systems in your enterprise. This allows you to more easily work with metadata across all applications such as: • Data-modeling applications • ETL applications • Reporting applications • Adapter software development kits Examples of conventions recommended for use with jobs and other objects are shown in the following table. Prefix Object Example DF_ n/a Data flow DF_Currency EDF_ _Input Embedded data flow EDF_Example_Input EDF_ _Output Embedded data flow EDF_Example_Output RTJob_ n/a Real-time job RTJob_OrderStatus WF_ n/a Work flow WF_SalesOrg JOB_ n/a Job JOB_SalesOrg n/a _DS Datastore ORA_DS DC_ n/a Datastore configuration DC_DB2_production SC_ n/a System configuration SC_ORA_test n/a _Memory_DS Memory datastore Catalog_Memory_DS PROC_ 71 Suffix n/a Stored procedure PROC_SalesStatus 2011-06-09
- 72. Projects and Jobs Although the Designer is a graphical user interface with icons representing objects in its windows, other interfaces might require you to identify object types by the text alone. By using a prefix or suffix, you can more easily identify your object's type. In addition to prefixes and suffixes, you might want to provide standardized names for objects that identify a specific action across all object types. For example: DF_OrderStatus, RTJob_OrderStatus. In addition to prefixes and suffixes, naming conventions can also include path name identifiers. For example, the stored procedure naming convention can look like either of the following: <datastore>.<owner>.<PROC_Name> <datastore>.<owner>.<package>.<PROC_Name> 72 2011-06-09
- 73. Datastores Datastores This section describes different types of datastores, provides details about the Attunity Connector datastore, and instructions for configuring datastores. 5.1 What are datastores? Datastores represent connection configurations between the software and databases or applications. These configurations can be direct or through adapters. Datastore configurations allow the software to access metadata from a database or application and read from or write to that database or application while the software executes a job. SAP BusinessObjects Data Services datastores can connect to: • Databases and mainframe file systems. • Applications that have pre-packaged or user-written adapters. • J.D. Edwards One World and J.D. Edwards World, Oracle Applications, PeopleSoft, SAP applications and SAP NetWeaver BW, and Siebel Applications. See the appropriate supplement guide. Note: The software reads and writes data stored in flat files through flat file formats. The software reads and writes data stored in XML documents through DTDs and XML Schemas. The specific information that a datastore object can access depends on the connection configuration. When your database or application changes, make corresponding changes in the datastore information in the software. The software does not automatically detect the new information. Note: Objects deleted from a datastore connection are identified in the project area and workspace by a red "deleted" icon. changes. This visual flag allows you to find and update data flows affected by datastore You can create multiple configurations for a datastore. This allows you to plan ahead for the different environments your datastore may be used in and limits the work involved with migrating jobs. For example, you can add a set of configurations (DEV, TEST, and PROD) to the same datastore name. These connection settings stay with the datastore during export or import. You can group any set of datastore configurations into a system configuration. When running or scheduling a job, select a system configuration, and thus, the set of datastore configurations for your current environment. 73 2011-06-09
- 74. Datastores Related Topics • Database datastores • Adapter datastores • File formats • Formatting XML documents • Creating and managing multiple datastore configurations 5.2 Database datastores Database datastores can represent single or multiple connections with: • Legacy systems using Attunity Connect • IBM DB2, HP Neoview, Informix, Microsoft SQL Server, Oracle, Sybase ASE, Sybase IQ, MySQL, Netezza, SAP BusinessObjects Data Federator, and Teradata databases (using native connections) • Other databases (through ODBC) • A repository, using a memory datastore or persistent cache datastore 5.2.1 Mainframe interface The software provides the Attunity Connector datastore that accesses mainframe data sources through Attunity Connect. The data sources that Attunity Connect accesses are in the following list. For a complete list of sources, refer to the Attunity documentation. • Adabas • DB2 UDB for OS/390 and DB2 UDB for OS/400 • IMS/DB • VSAM • Flat files on OS/390 and flat files on OS/400 5.2.1.1 Prerequisites for an Attunity datastore 74 2011-06-09
- 75. Datastores Attunity Connector accesses mainframe data using software that you must manually install on the mainframe server and the local client (Job Server) computer. The software connects to Attunity Connector using its ODBC interface. It is not necessary to purchase a separate ODBC driver manager for UNIX and Windows platforms. Servers Install and configure the Attunity Connect product on the server (for example, an zSeries computer). Clients To access mainframe data using Attunity Connector, install the Attunity Connect product. The ODBC driver is required. Attunity also offers an optional tool called Attunity Studio, which you can use for configuration and administration. Configure ODBC data sources on the client (SAP BusinessObjectsData Services Job Server). When you install a Job Server on UNIX, the installer will prompt you to provide an installation directory path for Attunity connector software. In addition, you do not need to install a driver manager, because the software loads ODBC drivers directly on UNIX platforms. For more information about how to install and configure these products, refer to their documentation. 5.2.1.2 Configuring an Attunity datastore To use the Attunity Connector datastore option, upgrade your repository to SAP BusinessObjectsData Services version 6.5.1 or later. To create an Attunity Connector datastore: 1. In the Datastores tab of the object library, right-click and select New. 2. Enter a name for the datastore. 3. In the Datastore type box, select Database. 4. In the Database type box, select Attunity Connector. 5. Type the Attunity data source name, location of the Attunity daemon (Host location), the Attunity daemon port number, and a unique Attunity server workspace name. 6. To change any of the default options (such as Rows per Commit or Language), click the Advanced button. 7. Click OK. You can now use the new datastore connection to import metadata tables into the current repository. 75 2011-06-09
- 76. Datastores 5.2.1.3 Specifying multiple data sources in one Attunity datastore You can use the Attunity Connector datastore to access multiple Attunity data sources on the same Attunity Daemon location. If you have several types of data on the same computer, for example a DB2 database and VSAM, you might want to access both types of data using a single connection. For example, you can use a single connection to join tables (and push the join operation down to a remote server), which reduces the amount of data transmitted through your network. To specify multiple sources in the Datastore Editor: 1. Separate data source names with semicolons in the Attunity data source box using the following format: AttunityDataSourceName;AttunityDataSourceName For example, if you have a DB2 data source named DSN4 and a VSAM data source named Navdemo, enter the following values into the Data source box: DSN4;Navdemo 2. If you list multiple data source names for one Attunity Connector datastore, ensure that you meet the following requirements: • All Attunity data sources must be accessible by the same user name and password. • All Attunity data sources must use the same workspace. When you setup access to the data sources in Attunity Studio, use the same workspace name for each data source. 5.2.1.4 Data Services naming convention for Attunity tables Data Services' format for accessing Attunity tables is unique to Data Services. Because a single datastore can access multiple software systems that do not share the same namespace, the name of the Attunity data source must be specified when referring to a table. With an Attunity Connector, precede the table name with the data source and owner names separated by a colon. The format is as follows: AttunityDataSource:OwnerName.TableName When using the Designer to create your jobs with imported Attunity tables, Data Services automatically generates the correct SQL for this format. However, when you author SQL, be sure to use this format. You can author SQL in the following constructs: • • 76 SQL function SQL transform 2011-06-09
- 77. Datastores • Pushdown_sql function • Pre-load commands in table loader • Post-load commands in table loader Note: For any table in Data Services, the maximum size of the owner name is 64 characters. In the case of Attunity tables, the maximum size of the Attunity data source name and actual owner name is 63 (the colon accounts for 1 character). Data Services cannot access a table with an owner name larger than 64 characters. 5.2.1.5 Limitations All Data Services features are available when you use an Attunity Connector datastore except the following: • Bulk loading • Imported functions (imports metadata for tables only) • Template tables (creating tables) • The datetime data type supports up to 2 sub-seconds only • Data Services cannot load timestamp data into a timestamp column in a table because Attunity truncates varchar data to 8 characters, which is not enough to correctly represent a timestamp value. • When running a job on UNIX, the job could fail with following error: [D000] Cannot open file /usr1/attun/navroot/def/sys System error 13: The file access permissions do not allow the specified action.; (OPEN) This error occurs because of insufficient file permissions to some of the files in the Attunity installation directory. To avoid this error, change the file permissions for all files in the Attunity directory to 777 by executing the following command from the Attunity installation directory: $ chmod -R 777 * 5.2.2 Defining a database datastore Define at least one database datastore for each database or mainframe file system with which you are exchanging data. To define a datastore, get appropriate access privileges to the database or file system that the datastore describes. 77 2011-06-09
- 78. Datastores For example, to allow the software to use parameterized SQL when reading or writing to DB2 databases, authorize the user (of the datastore/database) to create, execute and drop stored procedures. If a user is not authorized to create, execute and drop stored procedures, jobs will still run. However, they will produce a warning message and will run less efficiently. 5.2.2.1 To define a Database datastore 1. In the Datastores tab of the object library, right-click and select New. 2. Enter the name of the new datastore in the Datastore Name field. The name can contain any alphabetical or numeric characters or underscores (_). It cannot contain spaces. 3. Select the Datastore type. Choose Database. When you select a Datastore Type, the software displays other options relevant to that type. 4. Select the Database type. Note: If you select Data Federator, you must also specify the catalog name and the schema name in the URL. If you do not, you may see all of the tables from each catalog. a. Select ODBC Admin and then the System DSN tab. b. Highlight Data Federator, and then click Configure. c. In the URL option, enter the catalog name and the schema name, for example, jdbc:lese lect://localhost/catalogname;schema=schemaname. 5. Enter the appropriate information for the selected database type. 6. The Enable automatic data transfer check box is selected by default when you create a new datastore and you chose Database for Datastore type. This check box displays for all databases except Attunity Connector, Data Federator, Memory, and Persistent Cache. Keep Enable automatic data transfer selected to enable transfer tables in this datastore that the Data_Transfer transform can use to push down subsequent database operations. 7. At this point, you can save the datastore or add more information to it: • To save the datastore and close the Datastore Editor, click OK. • To add more information, select Advanced. To enter values for each configuration option, click the cells under each configuration name. For the datastore as a whole, the following buttons are available: 78 2011-06-09
- 79. Datastores Buttons Description Import unsupported data types as VARCHAR of size The data types that the software supports are documented in the Reference Guide. If you want the software to convert a data type in your source that it would not normally support, select this option and enter the number of characters that you will allow. Edit Opens the Configurations for Datastore dialog. Use the tool bar on this window to add, configure, and manage multiple configurations for a datastore. Show ATL Opens a text window that displays how the software will code the selections you make for this datastore in its scripting language. OK Saves selections and closes the Datastore Editor (Create New Datastore) window. Cancel Cancels selections and closes the Datastore Editor window. Apply Saves selections. 8. Click OK. Note: On versions of Data Integrator prior to version 11.7.0, the correct database type to use when creating a datastore on Netezza was ODBC. SAP BusinessObjectsData Services 11.7.1 provides a specific Netezza option as the Database type instead of ODBC. When using Netezza as the database with the software, we recommend that you choose the software's Netezza option as the Database type rather than ODBC. Related Topics • Performance Optimization Guide: Data Transfer transform for push-down operations • Reference Guide: Datastore • Creating and managing multiple datastore configurations • Ways of importing metadata 79 2011-06-09
- 80. Datastores 5.2.3 Configuring ODBC data sources on UNIX To use ODBC data sources on UNIX platforms, you may need to perform additional configuration. Data Services provides the dsdb_setup.sh utility to simplify configuration of natively-supported ODBC data sources such as MySQL and Teradata. Other ODBC data sources may require manual configuration. Related Topics • Administrator's Guide: Configuring ODBC data sources on UNIX 5.2.4 Changing a datastore definition Like all objects, datastores are defined by both options and properties: • Options control the operation of objects. For example, the name of the database to connect to is a datastore option. • Properties document the object. For example, the name of the datastore and the date on which it was created are datastore properties. Properties are merely descriptive of the object and do not affect its operation. 5.2.4.1 To change datastore options 1. Go to the Datastores tab in the object library. 2. Right-click the datastore name and choose Edit. The Datastore Editor appears (the title bar for this dialog displays Edit Datastore). You can do the following tasks: • Change the connection information for the current datastore configuration. • Click Advanced and change properties for the current configuration, • Click Edit to add, edit, or delete additional configurations. The Configurations for Datastore dialog opens when you select Edit in the Datastore Editor. Once you add a new configuration to an existing datastore, you can use the fields in the grid to change connection values and properties for the new configuration. 3. Click OK. 80 2011-06-09
- 81. Datastores The options take effect immediately. Related Topics • Reference Guide: Database datastores 5.2.4.2 To change datastore properties 1. Go to the datastore tab in the object library. 2. Right-click the datastore name and select Properties. The Properties window opens. 3. Change the datastore properties. 4. Click OK. Related Topics • Reference Guide: Datastore 5.2.5 Browsing metadata through a database datastore The software stores metadata information for all imported objects in a datastore. You can use the software to view metadata for imported or non-imported objects and to check whether the metadata has changed for objects already imported. 5.2.5.1 To view imported objects 1. Go to the Datastores tab in the object library. 2. Click the plus sign (+) next to the datastore name to view the object types in the datastore. For example, database datastores have functions, tables, and template tables. 3. Click the plus sign (+) next to an object type to view the objects of that type imported from the datastore. For example, click the plus sign (+) next to tables to view the imported tables. 81 2011-06-09
- 82. Datastores 5.2.5.2 To sort the list of objects Click the column heading to sort the objects in each grouping and the groupings in each datastore alphabetically. Click again to sort in reverse-alphabetical order. 5.2.5.3 To view datastore metadata 1. Select the Datastores tab in the object library. 2. Choose a datastore, right-click, and select Open. (Alternatively, you can double-click the datastore icon.) The software opens the datastore explorer in the workspace. The datastore explorer lists the tables in the datastore. You can view tables in the external database or tables in the internal repository. You can also search through them. 3. Select External metadata to view tables in the external database. If you select one or more tables, you can right-click for further options. Command Description Open (Only available if you select one table.) Opens the editor for the table metadata. Import Imports (or re-imports) metadata from the database into the repository. Reconcile Checks for differences between metadata in the database and metadata in the repository. 4. Select Repository metadata to view imported tables. If you select one or more tables, you can right-click for further options. Command Open (Only available if you select one table) 82 Description Opens the editor for the table metadata. 2011-06-09
- 83. Datastores Command Description Reconcile Checks for differences between metadata in the repository and metadata in the database. Reimport Reimports metadata from the database into the repository. Delete Deletes the table or tables from the repository. Properties (Only available if you select one table) Shows the properties of the selected table. View Data Opens the View Data window which allows you to see the data currently in the table. Related Topics • To import by searching 5.2.5.4 To determine if a schema has changed since it was imported 1. In the browser window showing the list of repository tables, select External Metadata. 2. Choose the table or tables you want to check for changes. 3. Right-click and choose Reconcile. The Changed column displays YES to indicate that the database tables differ from the metadata imported into the software. To use the most recent metadata from the software, reimport the table. The Imported column displays YES to indicate that the table has been imported into the repository. 5.2.5.5 To browse the metadata for an external table 1. In the browser window showing the list of external tables, select the table you want to view. 2. Right-click and choose Open. 83 2011-06-09
- 84. Datastores A table editor appears in the workspace and displays the schema and attributes of the table. 5.2.5.6 To view the metadata for an imported table 1. Select the table name in the list of imported tables. 2. Right-click and select Open. A table editor appears in the workspace and displays the schema and attributes of the table. 5.2.5.7 To view secondary index information for tables Secondary index information can help you understand the schema of an imported table. 1. From the datastores tab in the Designer, right-click a table to open the shortcut menu. 2. From the shortcut menu, click Properties to open the Properties window. 3. In the Properties window, click the Indexes tab. The left portion of the window displays the Index list. 4. Click an index to see the contents. 5.2.6 Importing metadata through a database datastore For database datastores, you can import metadata for tables and functions. 5.2.6.1 Imported table information The software determines and stores a specific set of metadata information for tables. After importing metadata, you can edit column names, descriptions, and data types. The edits are propagated to all objects that call these objects. 84 2011-06-09
- 85. Datastores Metadata Description The name of the table as it appears in the database. Table name Note: The maximum table name length supported by the software is 64 characters. If the table name exceeds 64 characters, you may not be able to import the table. Table description The description of the table. Column name The name of the column. Column description The description of the column. The data type for the column. Column data type Column content type If a column is defined as an unsupported data type, the software converts the data type to one that is supported. In some cases, if the software cannot convert the data type, it ignores the column entirely. The content type identifies the type of data in the field. The column(s) that comprise the primary key for the table. Primary key column Table attribute After a table has been added to a data flow diagram, these columns are indicated in the column list by a key icon next to the column name. Information the software records about the table such as the date created and date modified if these values are available. Name of the table owner. Owner name 85 Note: The owner name for MySQL and Netezza data sources corresponds to the name of the database or schema where the table appears. 2011-06-09
- 86. Datastores Varchar and Column Information from SAP BusinessObjects Data Federator tables Any decimal column imported to Data Serves from an SAP BusinessObjects Data Federator data source is converted to the decimal precision and scale(28,6). Any varchar column imported to the software from an SAP BusinessObjects Data Federator data source is varchar(1024). You may change the decimal precision or scale and varchar size within the software after importing from the SAP BusinessObjects Data Federator data source. 5.2.6.2 Imported stored function and procedure information The software can import stored procedures from DB2, MS SQL Server, Oracle, Sybase ASE, Sybase IQ, and Teredata databases. You can also import stored functions and packages from Oracle. You can use these functions and procedures in the extraction specifications you give Data Services. Information that is imported for functions includes: • Function parameters • Return type • Name, owner Imported functions and procedures appear on the Datastores tab of the object library. Functions and procedures appear in the Function branch of each datastore tree. You can configure imported functions and procedures through the function wizard and the smart editor in a category identified by the datastore name. Related Topics • Reference Guide: About procedures 5.2.6.3 Ways of importing metadata This section discusses methods you can use to import metadata. 5.2.6.3.1 To import by browsing Note: Functions cannot be imported by browsing. 1. Open the object library. 86 2011-06-09
- 87. Datastores 2. Go to the Datastores tab. 3. Select the datastore you want to use. 4. Right-click and choose Open. The items available to import through the datastore appear in the workspace. In some environments, the tables are organized and displayed as a tree structure. If this is true, there is a plus sign (+) to the left of the name. Click the plus sign to navigate the structure. The workspace contains columns that indicate whether the table has already been imported into the software (Imported) and if the table schema has changed since it was imported (Changed). To verify whether the repository contains the most recent metadata for an object, right-click the object and choose Reconcile. 5. Select the items for which you want to import metadata. For example, to import a table, you must select a table rather than a folder that contains tables. 6. Right-click and choose Import. 7. In the object library, go to the Datastores tab to display the list of imported objects. 5.2.6.3.2 To import by name 1. Open the object library. 2. Click the Datastores tab. 3. Select the datastore you want to use. 4. Right-click and choose Import By Name. 5. In the Import By Name window, choose the type of item you want to import from the Type list. If you are importing a stored procedure, select Function. 6. To import tables: a. Enter a table name in the Name box to specify a particular table, or select the All check box, if available, to specify all tables. If the name is case-sensitive in the database (and not all uppercase), enter the name as it appears in the database and use double quotation marks (") around the name to preserve the case. b. Enter an owner name in the Owner box to limit the specified tables to a particular owner. If you leave the owner name blank, you specify matching tables regardless of owner (that is, any table with the specified table name). 7. To import functions and procedures: • In the Name box, enter the name of the function or stored procedure. If the name is case-sensitive in the database (and not all uppercase), enter the name as it appears in the database and use double quotation marks (") around the name to preserve the case. Otherwise, the software will convert names into all upper-case characters. You can also enter the name of a package. An Oracle package is an encapsulated collection of related program objects (e.g., procedures, functions, variables, constants, cursors, and exceptions) 87 2011-06-09
- 88. Datastores stored together in the database. The software allows you to import procedures or functions created within packages and use them as top-level procedures or functions. If you enter a package name, the software imports all stored procedures and stored functions defined within the Oracle package. You cannot import an individual function or procedure defined within a package. • Enter an owner name in the Owner box to limit the specified functions to a particular owner. If you leave the owner name blank, you specify matching functions regardless of owner (that is, any function with the specified name). • If you are importing an Oracle function or stored procedure and any of the following conditions apply, clear the Callable from SQL expression check box. A stored procedure cannot be pushed down to a database inside another SQL statement when the stored procedure contains a DDL statement, ends the current transaction with COMMIT or ROLLBACK, or issues any ALTER SESSION or ALTER SYSTEM commands. 8. Click OK. 5.2.6.3.3 To import by searching Note: Functions cannot be imported by searching. 1. Open the object library. 2. Click the Datastores tab. 3. Select the name of the datastore you want to use. 4. Right-click and select Search. The Search window appears. 5. Enter the entire item name or some part of it in the Name text box. If the name is case-sensitive in the database (and not all uppercase), enter the name as it appears in the database and use double quotation marks (") around the name to preserve the case. 6. Select Contains or Equals from the drop-down list to the right depending on whether you provide a complete or partial search value. Equals qualifies only the full search string. That is, you need to search for owner.table_name rather than simply table_name. 7. (Optional) Enter a description in the Description text box. 8. Select the object type in the Type box. 9. Select the datastore in which you want to search from the Look In box. 10. Select External from the drop-down box to the right of the Look In box. External indicates that the software searches for the item in the entire database defined by the datastore. Internal indicates that the software searches only the items that have been imported. 11. Go to the Advanced tab to search using the software's attribute values. 88 2011-06-09
- 89. Datastores The advanced options only apply to searches of imported items. 12. Click Search. The software lists the tables matching your search criteria. 13. To import a table from the returned list, select the table, right-click, and choose Import. 5.2.6.4 Reimporting objects If you have already imported an object such as a datastore, function, or table, you can reimport it, which updates the object's metadata from your database (reimporting overwrites any changes you might have made to the object in the software). To reimport objects in previous versions of the software, you opened the datastore, viewed the repository metadata, and selected the objects to reimport. In this version of the software, you can reimport objects using the object library at various levels: • Individual objects — Reimports the metadata for an individual object such as a table or function • Category node level — Reimports the definitions of all objects of that type in that datastore, for example all tables in the datastore • Datastore level — Reimports the entire datastore and all its dependent objects including tables, functions, IDOCs, and hierarchies 5.2.6.4.1 To reimport objects from the object library 1. In the object library, click the Datastores tab. 2. Right-click an individual object and click Reimport, or right-click a category node or datastore name and click Reimport All. You can also select multiple individual objects using Ctrl-click or Shift-click. 3. Click Yes to reimport the metadata. 4. If you selected multiple objects to reimport (for example with Reimport All), the software requests confirmation for each object unless you check the box Don't ask me again for the remaining objects. You can skip objects to reimport by clicking No for that object. If you are unsure whether to reimport (and thereby overwrite) the object, click View Where Used to display where the object is currently being used in your jobs. 89 2011-06-09
- 90. Datastores 5.2.7 Memory datastores The software also allows you to create a database datastore using Memory as the Database type. Memory datastores are designed to enhance processing performance of data flows executing in real-time jobs. Data (typically small amounts in a real-time job) is stored in memory to provide immediate access instead of going to the original source data. A memory datastore is a container for memory tables. A datastore normally provides a connection to a database, application, or adapter. By contrast, a memory datastore contains memory table schemas saved in the repository. Memory tables are schemas that allow you to cache intermediate data. Memory tables can cache data from relational database tables and hierarchical data files such as XML messages and SAP IDocs (both of which contain nested schemas). Memory tables can be used to: • Move data between data flows in real-time jobs. By caching intermediate data, the performance of real-time jobs with multiple data flows is far better than it would be if files or regular tables were used to store intermediate data. For best performance, only use memory tables when processing small quantities of data. • Store table data in memory for the duration of a job. By storing table data in memory, the LOOKUP_EXT function and other transforms and functions that do not require database operations can access data without having to read it from a remote database. The lifetime of memory table data is the duration of the job. The data in memory tables cannot be shared between different real-time jobs. Support for the use of memory tables in batch jobs is not available. 5.2.7.1 Creating memory datastores You can create memory datastores using the Datastore Editor window. 5.2.7.1.1 To define a memory datastore 1. From the Project menu, select NewDatastore. 2. In the Name box, enter the name of the new datastore. Be sure to use the naming convention "Memory_DS". Datastore names are appended to table names when table icons appear in the workspace. Memory tables are represented in the workspace with regular table icons. Therefore, label a memory datastore to distinguish its memory tables from regular database tables in the workspace. 3. In the Datastore type box keep the default Database. 90 2011-06-09
- 91. Datastores 4. In the Database Type box select Memory. No additional attributes are required for the memory datastore. 5. Click OK. 5.2.7.2 Creating memory tables When you create a memory table, you do not have to specify the table's schema or import the table's metadata. Instead, the software creates the schema for each memory table automatically based on the preceding schema, which can be either a schema from a relational database table or hierarchical data files such as XML messages. The first time you save the job, the software defines the memory table's schema and saves the table. Subsequently, the table appears with a table icon in the workspace and in the object library under the memory datastore. 5.2.7.2.1 To create a memory table 1. From the tool palette, click the template table icon. 2. Click inside a data flow to place the template table. The Create Table window opens. 3. From the Create Table window, select the memory datastore. 4. Enter a table name. 5. If you want a system-generated row ID column in the table, click the Create Row ID check box. 6. Click OK. The memory table appears in the workspace as a template table icon. 7. Connect the memory table to the data flow as a target. 8. From the Project menu select Save. In the workspace, the memory table's icon changes to a target table icon and the table appears in the object library under the memory datastore's list of tables. Related Topics • Create Row ID option 5.2.7.3 Using memory tables as sources and targets 91 2011-06-09
- 92. Datastores After you create a memory table as a target in one data flow, you can use a memory table as a source or target in any data flow. Related Topics • Real-time Jobs 5.2.7.3.1 To use a memory table as a source or target 1. In the object library, click the Datastores tab. 2. Expand the memory datastore that contains the memory table you want to use. 3. Expand Tables. A list of tables appears. 4. Select the memory table you want to use as a source or target, and drag it into an open data flow. 5. Connect the memory table as a source or target in the data flow. If you are using a memory table as a target, open the memory table's target table editor to set table options. 6. Save the job. Related Topics • Memory table target options 5.2.7.4 Update Schema option You might want to quickly update a memory target table's schema if the preceding schema changes. To do this, use the Update Schema option. Otherwise, you would have to add a new memory table to update a schema. 5.2.7.4.1 To update the schema of a memory target table 1. Right-click the memory target table's icon in the work space. 2. Select Update Schema. The schema of the preceding object is used to update the memory target table's schema. The current memory table is updated in your repository. All occurrences of the current memory table are updated with the new schema. 92 2011-06-09
- 93. Datastores 5.2.7.5 Memory table target options The Delete data from table before loading option is available for memory table targets. The default is on (the box is selected). To set this option, open the memory target table editor. If you deselect this option, new data will append to the existing table data. 5.2.7.6 Create Row ID option If the Create Row ID is checked in the Create Memory Table window, the software generates an integer column called DI_Row_ID in which the first row inserted gets a value of 1, the second row inserted gets a value of 2, etc. This new column allows you to use a LOOKUP_EXT expression as an iterator in a script. Note: The same functionality is available for other datastore types using the SQL function. Use the DI_Row_ID column to iterate through a table using a lookup_ext function in a script. For example: $NumOfRows = total_rows (memory_DS..table1) $I = 1; $count=0 while ($count < $NumOfRows) begin $data = lookup_ext([memory_DS..table1, 'NO_CACHE','MAX'],[A],[O],[DI_Row_ID,'=',$I]); $1 = $I + 1; if ($data != NULL) begin $count = $count + 1; end end In the preceding script, table1 is a memory table. The table's name is preceded by its datastore name (memory_DS), a dot, a blank space (where a table owner would be for a regular table), then a second dot. There are no owners for memory datastores, so tables are identified by just the datastore name and the table name as shown. Select the LOOKUP_EXT function arguments (line 7) from the function editor when you define a LOOKUP_EXT function. The TOTAL_ROWS(DatastoreName.Owner.TableName) function returns the number of rows in a particular table in a datastore. This function can be used with any type of datastore. If used with a memory datastore, use the following syntax: TOTAL_ROWS( DatastoreName..TableName ) 93 2011-06-09
- 94. Datastores The software also provides a built-in function that you can use to explicitly expunge data from a memory table. This provides finer control than the active job has over your data and memory usage. The TRUNCATE_TABLE( DatastoreName..TableName ) function can only be used with memory tables. Related Topics • Reference Guide: Functions and Procedures, Descriptions of built-in functions 5.2.7.7 Troubleshooting memory tables • One possible error, particularly when using memory tables, is that the software runs out of virtual memory space. The software exits if it runs out of memory while executing any operation. • A validation and run time error occurs if the schema of a memory table does not match the schema of the preceding object in the data flow. To correct this error, use the Update Schema option or create a new memory table to match the schema of the preceding object in the data flow. • Two log files contain information specific to memory tables: trace_memory_reader log and trace_memory_loader log. 5.2.8 Persistent cache datastores The software also allows you to create a database datastore using Persistent cache as the Database type. Persistent cache datastores provide the following benefits for data flows that process large volumes of data. • You can store a large amount of data in persistent cache which the software quickly loads into memory to provide immediate access during a job. For example, you can access a lookup table or comparison table locally (instead of reading from a remote database). • You can create cache tables that multiple data flows can share (unlike a memory table which cannot be shared between different real-time jobs). For example, if a large lookup table used in a lookup_ext function rarely changes, you can create a cache once and subsequent jobs can use this cache instead of creating it each time. A persistent cache datastore is a container for cache tables. A datastore normally provides a connection to a database, application, or adapter. By contrast, a persistent cache datastore contains cache table schemas saved in the repository. Persistent cache tables allow you to cache large amounts of data. Persistent cache tables can cache data from relational database tables and files. 94 2011-06-09
- 95. Datastores Note: You cannot cache data from hierarchical data files such as XML messages and SAP IDocs (both of which contain nested schemas). You cannot perform incremental inserts, deletes, or updates on a persistent cache table. You create a persistent cache table by loading data into the persistent cache target table using one data flow. You can then subsequently read from the cache table in another data flow. When you load data into a persistent cache table, the software always truncates and recreates the table. 5.2.8.1 Creating persistent cache datastores You can create persistent cache datastores using the Datastore Editor window. 5.2.8.1.1 To define a persistent cache datastore 1. From the Project menu, select NewDatastore. 2. In the Name box, enter the name of the new datastore. Be sure to use a naming convention such as "Persist_DS". Datastore names are appended to table names when table icons appear in the workspace. Persistent cache tables are represented in the workspace with regular table icons. Therefore, label a persistent cache datastore to distinguish its persistent cache tables from regular database tables in the workspace. 3. In the Datastore type box, keep the default Database. 4. In the Database Type box, select Persistent cache. 5. In the Cache directory box, you can either type or browse to a directory where you want to store the persistent cache. 6. Click OK. 5.2.8.2 Creating persistent cache tables When you create a persistent cache table, you do not have to specify the table's schema or import the table's metadata. Instead, the software creates the schema for each persistent cache table automatically based on the preceding schema. The first time you save the job, the software defines the persistent cache table's schema and saves the table. Subsequently, the table appears with a table icon in the workspace and in the object library under the persistent cache datastore. You create a persistent cache table in one of the following ways: • • 95 As a target template table in a data flow As part of the Data_Transfer transform during the job execution 2011-06-09
- 96. Datastores Related Topics • Reference Guide: Data_Transfer 5.2.8.2.1 To create a persistent cache table as a target in a data flow 1. Use one of the following methods to open the Create Template window: • From the tool palette: a. Click the template table icon. b. Click inside a data flow to place the template table in the workspace. c. On the Create Template window, select the persistent cache datastore. • From the object library: a. Expand a persistent cache datastore. b. Click the template table icon and drag it to the workspace. 2. On the Create Template window, enter a table name. 3. Click OK. The persistent cache table appears in the workspace as a template table icon. 4. Connect the persistent cache table to the data flow as a target (usually a Query transform). 5. In the Query transform, map the Schema In columns that you want to include in the persistent cache table. 6. Open the persistent cache table's target table editor to set table options. 7. On the Options tab of the persistent cache target table editor, you can change the following options for the persistent cache table. • Column comparison — Specifies how the input columns are mapped to persistent cache table columns. There are two options: • Compare_by_position — The software disregards the column names and maps source columns to target columns by position. • Compare_by_name — The software maps source columns to target columns by name. This option is the default. • Include duplicate keys — Select this check box to cache duplicate keys. This option is selected by default. 8. On the Keys tab, specify the key column or columns to use as the key in the persistent cache table. 9. From the Project menu select Save. In the workspace, the template table's icon changes to a target table icon and the table appears in the object library under the persistent cache datastore's list of tables. 96 2011-06-09
- 97. Datastores Related Topics • Reference Guide:Target persistent cache tables 5.2.8.3 Using persistent cache tables as sources After you create a persistent cache table as a target in one data flow, you can use the persistent cache table as a source in any data flow. You can also use it as a lookup table or comparison table. Related Topics • Reference Guide: Persistent cache source 5.2.9 Linked datastores Various database vendors support one-way communication paths from one database server to another. Oracle calls these paths database links. In DB2, the one-way communication path from a database server to another database server is provided by an information server that allows a set of servers to get data from remote data sources. In Microsoft SQL Server, linked servers provide the one-way communication path from one database server to another. These solutions allow local users to access data on a remote database, which can be on the local or a remote computer and of the same or different database type. For example, a local Oracle database server, called Orders, can store a database link to access information in a remote Oracle database, Customers. Users connected to Customers however, cannot use the same link to access data in Orders. Users logged into database Customers must define a separate link, stored in the data dictionary of database Customers, to access data on Orders. The software refers to communication paths between databases as database links. The datastores in a database link relationship are called linked datastores. The software uses linked datastores to enhance its performance by pushing down operations to a target database using a target datastore. Related Topics • Performance Optimization Guide: Database link support for push-down operations across datastores 5.2.9.1 Relationship between database links and datastores 97 2011-06-09
- 98. Datastores A database link stores information about how to connect to a remote data source, such as its host name, database name, user name, password, and database type. The same information is stored in an SAP BusinessObjects Data Services database datastore.You can associate the datastore to another datastore and then import an external database link as an option of a datastore. The datastores must connect to the databases defined in the database link. Additional requirements are as follows: • • • • • A local server for database links must be a target server in the software A remote server for database links must be a source server in the software An external (exists first in a database) database link establishes the relationship between any target datastore and a source datastore A Local datastore can be related to zero or multiple datastores using a database link for each remote database Two datastores can be related to each other using one link only The following diagram shows the possible relationships between database links and linked datastores: Four database links, DBLink 1 through 4, are on database DB1 and the software reads them through datastore Ds1. • • • • Dblink1 relates datastore Ds1 to datastore Ds2. This relationship is called linked datastore Dblink1 (the linked datastore has the same name as the external database link). Dblink2 is not mapped to any datastore in the software because it relates Ds1 with Ds2, which are also related by Dblink1. Although it is not a regular case, you can create multiple external database links that connect to the same remote source. However, the software allows only one database link between a target datastore and a source datastore pair. For example, if you select DBLink1 to link target datastore DS1 with source datastore DS2, you cannot import DBLink2 to do the same. Dblink3 is not mapped to any datastore in the software because there is no datastore defined for the remote data source to which the external database link refers. Dblink4 relates Ds1 with Ds3. Related Topics • Reference Guide: Datastore editor 98 2011-06-09
- 99. Datastores 5.3 Adapter datastores Depending on the adapter implementation, adapters allow you to: • Browse application metadata • Import application metadata into a repository • Move batch and real-time data between the software and applications SAP offers an Adapter Software Development Kit (SDK) to develop your own custom adapters. Also, you can buy the software pre-packaged adapters to access application metadata and data in any application. For more information on these products, contact your SAP sales representative. Adapters are represented in Designer by adapter datastores. Jobs provide batch and real-time data movement between the software and applications through an adapter datastore's subordinate objects: Subordinate Objects Use as Tables Source or target Documents For Source or target Batch data movement Functions Function call in query Message functions Function call in query Outbound messages Target only Adapters can provide access to an application's data and metadata or just metadata. For example, if the data source is SQL-compatible, the adapter might be designed to access metadata, while the software extracts data from or loads data directly to the application. Related Topics • Management Console Guide: Adapters • Source and target objects • Real-time source and target objects 99 2011-06-09
- 100. Datastores 5.3.1 Defining an adapter datastore You need to define at least one datastore for each adapter through which you are extracting or loading data. To define a datastore, you must have appropriate access privileges to the application that the adapter serves. 5.3.1.1 To define an adapter datastore 1. In the Object Library, click to select the Datastores tab. 2. Right-click and select New. The Datastore Editor dialog opens (the title bar reads, Create new Datastore). 3. Enter a unique identifying name for the datastore. The datastore name appears in the Designer only. It can be the same as the adapter name. 4. In the Datastore type list, select Adapter. 5. Select a Job server from the list. To create an adapter datastore, you must first install the adapter on the Job Server computer, configure the Job Server to support local adapters using the System Manager utility, and ensure that the Job Server's service is running. Adapters residing on the Job Server computer and registered with the selected Job Server appear in the Job server list. 6. Select an adapter instance from the Adapter instance name list. 7. Enter all adapter information required to complete the datastore connection. Note: If the developer included a description for each option, the software displays it below the grid. Also the adapter documentation should list all information required for a datastore connection. For the datastore as a whole, the following buttons are available: Buttons Edit 100 Description Opens the Configurations for Datastore dialog. Use the tool bar on this window to add, configure, and manage multiple configurations for a datastore. 2011-06-09
- 101. Datastores Buttons Description Show ATL Opens a text window that displays how the software will code the selections you make for this datastore in its scripting language. OK Saves selections and closes the Datastore Editor (Create New Datastore) window. Cancel Cancels selections and closes the Datastore Editor window. Apply Saves selections. 8. Click OK. The datastore configuration is saved in your metadata repository and the new datastore appears in the object library. After you complete your datastore connection, you can browse and/or import metadata from the data source through the adapter. 5.3.1.2 To change an adapter datastore's configuration 1. Right-click the datastore you want to browse and select Edit to open the Datastore Editor window. 2. Edit configuration information. When editing an adapter datastore, enter or select a value. The software looks for the Job Server and adapter instance name you specify. If the Job Server and adapter instance both exist, and the Designer can communicate to get the adapter's properties, then it displays them accordingly. If the Designer cannot get the adapter's properties, then it retains the previous properties. 3. Click OK. The edited datastore configuration is saved in your metadata repository. 5.3.1.3 To delete an adapter datastore and associated metadata objects 1. Right-click the datastore you want to delete and select Delete. 101 2011-06-09
- 102. Datastores 2. Click OK in the confirmation window. The software removes the datastore and all metadata objects contained within that datastore from the metadata repository. If these objects exist in established flows, they appear with a deleted icon . 5.3.2 Browsing metadata through an adapter datastore The metadata you can browse depends on the specific adapter. 5.3.2.1 To browse application metadata 1. Right-click the datastore you want to browse and select Open. A window opens showing source metadata. 2. Scroll to view metadata name and description attributes. 3. Click plus signs [+] to expand objects and view subordinate objects. 4. Right-click any object to check importability. 5.3.3 Importing metadata through an adapter datastore The metadata you can import depends on the specific adapter. After importing metadata, you can edit it. Your edits propagate to all objects that call these objects. 5.3.3.1 To import application metadata while browsing 1. Right-click the datastore you want to browse, then select Open. 2. Find the metadata object you want to import from the browsable list. 3. Right-click the object and select Import. 4. The object is imported into one of the adapter datastore containers (documents, functions, tables, outbound messages, or message functions). 102 2011-06-09
- 103. Datastores 5.3.3.2 To import application metadata by name 1. Right-click the datastore from which you want metadata, then select Import by name. The Import by name window appears containing import parameters with corresponding text boxes. 2. Click each import parameter text box and enter specific information related to the object you want to import. 3. Click OK. Any object(s) matching your parameter constraints are imported to one of the corresponding categories specified under the datastore. 5.4 Web service datastores Web service datastores represent a connection from Data Services to an external web service-based data source. 5.4.1 Defining a web service datastore You need to define at least one datastore for each web service with which you are exchanging data. To define a datastore, you must have the appropriate access priveliges to the web services that the datastore describes. 5.4.1.1 To define a web services datastore 1. In the Datastores tab of the object library, right-click and select New. 2. Enter the name of the new datastore in the Datastore name field. The name can contain any alphabetical or numeric characters or underscores (_). It cannot contain spaces. 3. Select the Datastore type. Choose Web Service. When you select a Datastore Type, Data Services displays other options relevant to that type. 103 2011-06-09
- 104. Datastores 4. Specify the Web Service URL. The URL must accept connections and return the WSDL. 5. Click OK. The datastore configuration is saved in your metadata repository and the new datastore appears in the object library. After you complete your datastore connection, you can browse and/or import metadata from the web service through the datastore. 5.4.1.2 To change a web service datastore's configuration 1. Right-click the datastore you want to browse and select Edit to open the Datastore Editor window. 2. Edit configuration information. 3. Click OK. The edited datastore configuration is saved in your metadata repository. 5.4.1.3 To delete a web service datastore and associated metadata objects 1. Right-click the datastore you want to delete and select Delete. 2. Click OK in the confirmation window. Data Services removes the datastore and all metadata objects contained within that datastore from the metadata repository. If these objects exist in established data flows, they appear with a deleted icon. 5.4.2 Browsing WSDL metadata through a web service datastore Data Services stores metadata information for all imported objects in a datastore. You can use Data Services to view metadata for imported or non-imported objects and to check whether the metadata has changed for objects already imported. 5.4.2.1 To view imported objects 104 2011-06-09
- 105. Datastores 1. Go to the Datastores tab in the object library. 2. Click the plus sign (+) next to the datastore name to view the object types in the datastore. Web service datastores have functions. 3. Click the plus sign (+) next to an object type to view the objects of that type imported from the datastore. 5.4.2.2 To sort the list of objects Click the column heading to sort the objects in each grouping and the groupings in each datastore alphabetically. Click again to sort in reverse-alphabetical order. 5.4.2.3 To view WSDL metadata 1. Select the Datastores tab in the object library. 2. Choose a datastore, right-click, and select Open. (Alternatively, you can double-click the datastore icon.) Data Services opens the datastore explorer in the workspace. The datastore explorer lists the web service ports and operations in the datastore. You can view ports and operations in the external web service or in the internal repository. You can also search through them. 3. Select External metadata to view web service ports and operations from the external WSDL. If you select one or more operations, you can right-click for further options. Command Description Import Imports (or re-imports) operations from the database into the repository. 4. Select Repository metadata to view imported web service operations. If you select one or more operations, you can right-click for further options. 105 2011-06-09
- 106. Datastores Command Description Delete Deletes the operation or operations from the repository. Properties Shows the properties of the selected web service operation. 5.4.3 Importing metadata through a web service datastore For web service datastores, you can import metadata for web service operations. 5.4.3.1 To import web service operations 1. Right-click the datastore you want to browse, then select Open. 2. Find the web service operation you want to import from the browsable list. 3. Right-click the operation and select Import. The operation is imported into the web service datastore's function container. 5.5 Creating and managing multiple datastore configurations Creating multiple configurations for a single datastore allows you to consolidate separate datastore connections for similar sources or targets into one source or target datastore with multiple configurations. Then, you can select a set of configurations that includes the sources and targets you want by selecting a system configuration when you execute or schedule the job. The ability to create multiple datastore configurations provides greater ease-of-use for job portability scenarios, such as: • OEM (different databases for design and distribution) • Migration (different connections for DEV, TEST, and PROD) • Multi-instance (databases with different versions or locales) • Multi-user (databases for central and local repositories) For more information about how to use multiple datastores to support these scenarios, see . Related Topics • Portability solutions 106 2011-06-09
- 107. Datastores 5.5.1 Definitions Refer to the following terms when creating and managing multiple datastore configurations: Term Definition “Datastore configuration” Allows you to provide multiple metadata sources or targets for datastores. Each configuration is a property of a datastore that refers to a set of configurable options (such as database connection name, database type, user name, password, and locale) and their values. “Default datastore configura- The datastore configuration that the software uses for browsing and tion ” importing database objects (tables and functions) and executing jobs if no system configuration is specified. If a datastore has more than one configuration, select a default configuration, as needed. If a datastore has only one configuration, the software uses it as the default configuration. “Current datastore configura- The datastore configuration that the software uses to execute a job. If tion ” you define a system configuration, the software will execute the job using the system configuration. Specify a current configuration for each system configuration. If you do not create a system configuration, or the system configuration does not specify a configuration for a datastore, the software uses the default datastore configuration as the current configuration at job execution time. “Database objects” 107 The tables and functions that are imported from a datastore. Database objects usually have owners. Some database objects do not have owners. For example, database objects in an ODBC datastore connecting to an Access database do not have owners. 2011-06-09
- 108. Datastores Term Definition “Owner name” Owner name of a database object (for example, a table) in an underlying database. Also known as database owner name or physical owner name. “Alias” A logical owner name. Create an alias for objects that are in different database environments if you have different owner names in those environments. You can create an alias from the datastore editor for any datastore configuration. “Dependent objects” Dependent objects are the jobs, work flows, data flows, and custom functions in which a database object is used. Dependent object information is generated by the where-used utility. 5.5.2 Why use multiple datastore configurations? By creating multiple datastore configurations, you can decrease end-to-end development time in a multi-source, 24x7, enterprise data warehouse environment because you can easily port jobs among different database types, versions, and instances. For example, porting can be as simple as: 1. Creating a new configuration within an existing source or target datastore. 2. Adding a datastore alias then map configurations with different object owner names to it. 3. Defining a system configuration then adding datastore configurations required for a particular environment. Select a system configuration when you execute a job. 5.5.3 Creating a new configuration You can create multiple configurations for all datastore types except memory datastores. Use the Datastore Editor to create and edit datastore configurations. Related Topics • Reference Guide: Descriptions of objects, Datastore 108 2011-06-09
- 109. Datastores 5.5.3.1 To create a new datastore configuration 1. From the Datastores tab of the object library, right-click any existing datastore and select Edit. 2. Click Advanced to view existing configuration information. Each datastore must have at least one configuration. If only one configuration exists, it is the default configuration. 3. Click Edit to open the Configurations for Datastore window. 4. Click the Create New Configuration icon on the toolbar. The Create New Configuration window opens. 5. In the Create New Configuration window: a. Enter a unique, logical configuration Name. b. Select a Database type from the drop-down menu. c. Select a Database version from the drop-down menu. d. In the Values for table targets and SQL transforms section, the software pre-selects the Use values from value based on the existing database type and version. The Designer automatically uses the existing SQL transform and target values for the same database type and version. Further, if the database you want to associate with a new configuration is a later version than that associated with other existing configurations, the Designer automatically populates the Use values from with the earlier version. However, if database type and version are not already specified in an existing configuration, or if the database version is older than your existing configuration, you can choose to use the values from another existing configuration or the default for the database type and version. e. Select or clear the Restore values if they already exist option. When you delete datastore configurations, the software saves all associated target values and SQL transforms. If you create a new datastore configuration with the same database type and version as the one previously deleted, the Restore values if they already exist option allows you to access and take advantage of the saved value settings.) • If you keep this option (selected as default) the software uses customized target and SQL transform values from previously deleted datastore configurations. • If you deselect Restore values if they already exist, the software does not attempt to restore target and SQL transform values, allowing you to provide new values. f. Click OK to save the new configuration. If your datastore contains pre-existing data flows with SQL transforms or target objects, the software must add any new database type and version values to these transform and target objects. Under these circumstances, when you add a new datastore configuration, the software displays the Added New Values - Modified Objects window which provides detailed information 109 2011-06-09
- 110. Datastores about affected data flows and modified objects. These same results also display in the Output window of the Designer. See For each datastore, the software requires that one configuration be designated as the default configuration. The software uses the default configuration to import metadata and also preserves the default configuration during export and multi-user operations. Your first datastore configuration is automatically designated as the default; however after adding one or more additional datastore configurations, you can use the datastore editor to flag a different configuration as the default. When you export a repository, the software preserves all configurations in all datastores including related SQL transform text and target table editor settings. If the datastore you are exporting already exists in the target repository, the software overrides configurations in the target with source configurations. The software exports system configurations separate from other job related objects. 5.5.4 Adding a datastore alias From the datastore editor, you can also create multiple aliases for a datastore then map datastore configurations to each alias. 5.5.4.1 To create an alias 1. From within the datastore editor, click Advanced, then click Aliases (Click here to create). The Create New Alias window opens. 2. Under Alias Name in Designer, use only alphanumeric characters and the underscore symbol (_) to enter an alias name. 3. Click OK. The Create New Alias window closes and your new alias appears underneath the Aliases category When you define a datastore alias, the software substitutes your specified datastore configuration alias for the real owner name when you import metadata for database objects. You can also rename tables and functions after you import them. For more information, see Renaming table and function owner. 5.5.5 Functions to identify the configuration 110 2011-06-09
- 111. Datastores The software provides six functions that are useful when working with multiple source and target datastore configurations. Function Category Description db_type Miscellaneous Returns the database type of the current datastore configuration. db_version Miscellaneous Returns the database version of the current datastore configuration. db_database_name Miscellaneous Returns the database name of the current datastore configuration if the database type is MS SQL Server or Sybase ASE. db_owner Miscellaneous Returns the real owner name that corresponds to the given alias name under the current datastore configuration. current_configuration Miscellaneous Returns the name of the datastore configuration that is in use at runtime. current_system_configura tion Miscellaneous Returns the name of the current system configuration. If no system configuration is defined, returns a NULL value. The software links any SQL transform and target table editor settings used in a data flow to datastore configurations. You can also use variable interpolation in SQL text with these functions to enable a SQL transform to perform successfully regardless of which configuration the Job Server uses at job execution time. Use the Administrator to select a system configuration as well as view the underlying datastore configuration associated with it when you: • Execute batch jobs • Schedule batch jobs • View batch job history • Create services for real-time jobs To use multiple configurations successfully, design your jobs so that you do not need to change schemas, data types, functions, variables, and so on when you switch between datastore configurations. For example, if you have a datastore with a configuration for Oracle sources and SQL sources, make sure 111 2011-06-09
- 112. Datastores that the table metadata schemas match exactly. Use the same table names, alias names, number and order of columns, as well as the same column names, data types, and content types. Related Topics • Reference Guide: Descriptions of built-in functions • Reference Guide: SQL • Job portability tips 5.5.6 Portability solutions Set multiple source or target configurations for a single datastore if you want to quickly change connections to a different source or target database. The software provides several different solutions for porting jobs. Related Topics • Multi-user Development • Multi-user Environment Setup 5.5.6.1 Migration between environments When you must move repository metadata to another environment (for example from development to test or from test to production) which uses different source and target databases, the process typically includes the following characteristics: • The environments use the same database type but may have unique database versions or locales. • Database objects (tables and functions) can belong to different owners. • Each environment has a unique database connection name, user name, password, other connection properties, and owner mapping. • You use a typical repository migration procedure. Either you export jobs to an ATL file then import the ATL file to another repository, or you export jobs directly from one repository to another repository. Because the software overwrites datastore configurations during export, you should add configurations for the target environment (for example, add configurations for the test environment when migrating from development to test) to the source repository (for example, add to the development repository before migrating to the test environment). The Export utility saves additional configurations in the target environment, which means that you do not have to edit datastores before running ported jobs in the target environment. 112 2011-06-09
- 113. Datastores This solution offers the following advantages: • Minimal production down time: You can start jobs as soon as you export them. • Minimal security issues: Testers and operators in production do not need permission to modify repository objects. Related Topics • Administrator's Guide: Export/Import 5.5.6.2 Loading Multiple instances If you must load multiple instances of a data source to a target data warehouse, the task is the same as in a migration scenario except that you are using only one repository. 5.5.6.2.1 To load multiple instances of a data source to a target data warehouse 1. Create a datastore that connects to a particular instance. 2. Define the first datastore configuration. This datastore configuration contains all configurable properties such as database type, database connection name, user name, password, database version, and locale information. When you define a configuration for an Adapter datastore, make sure that the relevant Job Server is running so the Designer can find all available adapter instances for the datastore. 3. Define a set of alias-to-owner mappings within the datastore configuration. When you use an alias for a configuration, the software imports all objects using the metadata alias rather than using real owner names. This allows you to use database objects for jobs that are transparent to other database instances. 4. Use the database object owner renaming tool to rename owners of any existing database objects. 5. Import database objects and develop jobs using those objects, then run the jobs. 6. To support executing jobs under different instances, add datastore configurations for each additional instance. 7. Map owner names from the new database instance configurations to the aliases that you defined in an earlier step. 8. Run the jobs in all database instances. Related Topics • Renaming table and function owner 113 2011-06-09
- 114. Datastores 5.5.6.3 OEM deployment If you design jobs for one database type and deploy those jobs to other database types as an OEM partner, the deployment typically has the following characteristics: • The instances require various source database types and versions. • Since a datastore can only access one instance at a time, you may need to trigger functions at run-time to match different instances. If this is the case, the software requires different SQL text for functions (such as lookup_ext and sql) and transforms (such as the SQL transform). The software also requires different settings for the target table (configurable in the target table editor). • The instances may use different locales. • Database tables across different databases belong to different owners. • Each instance has a unique database connection name, user name, password, other connection properties, and owner mappings. • You export jobs to ATL files for deployment. 5.5.6.3.1 To deploy jobs to other database types as an OEM partner 1. Develop jobs for a particular database type following the steps described in the Loading Multiple instances scenario. To support a new instance under a new database type, the software copies target table and SQL transform database properties from the previous configuration to each additional configuration when you save it. If you selected a bulk loader method for one or more target tables within your job's data flows, and new configurations apply to different database types, open your targets and manually set the bulk loader option (assuming you still want to use the bulk loader method with the new database type). The software does not copy bulk loader options for targets from one database type to another. When the software saves a new configuration it also generates a report that provides a list of targets automatically set for bulk loading. Reference this report to make manual changes as needed. 2. If the SQL text in any SQL transform is not applicable for the new database type, modify the SQL text for the new database type. If the SQL text contains any hard-coded owner names or database names, consider replacing these names with variables to supply owner names or database names for multiple database types. This way, you will not have to modify the SQL text for each environment. 3. Because the software does not support unique SQL text for each database type or version of the sql(), lookup_ext(), and pushdown_sql() functions, use the db_type() and similar functions to get the database type and version of the current datastore configuration and provide the correct SQL text for that database type and version using the variable substitution (interpolation) technique. 114 2011-06-09
- 115. Datastores Related Topics • Reference Guide: SQL 5.5.6.4 Multi-user development If you are using a central repository management system, allowing multiple developers, each with their own local repository, to check in and check out jobs, the development environment typically has the following characteristics: • It has a central repository and a number of local repositories. • Multiple development environments get merged (via central repository operations such as check in and check out) at times. When this occurs, real owner names (used initially to import objects) must be later mapped to a set of aliases shared among all users. • The software preserves object history (versions and labels). • The instances share the same database type but may have different versions and locales. • Database objects may belong to different owners. • Each instance has a unique database connection name, user name, password, other connection properties, and owner mapping. In the multi-user development scenario you must define aliases so that the software can properly preserve the history for all objects in the shared environment. 5.5.6.4.1 Porting jobs in a multi-user environment When porting jobs in a multi-user environment, consider these points: • 115 Rename table owners and function owners to consolidate object database object owner names into aliases. • Renaming occurs in local repositories. To rename the database objects stored in the central repository, check out the datastore to a local repository and apply the renaming tool in the local repository. • If the objects to be renamed have dependent objects, the software will ask you to check out the dependent objects. • If all the dependent objects can be checked out, renaming will create a new object that has the alias and delete the original object that has the original owner name. • If all the dependent objects cannot be checked out (data flows are checked out by another user), the software displays a message, which gives you the option to proceed or cancel the operation. If you cannot check out some of the dependent objects, the renaming tool only affects the flows that you can check out. After renaming, the original object will co-exist with the new object. The number of flows affected by the renaming process will affect the Usage and Where-Used information in the Designer for both the original object and the new object. 2011-06-09
- 116. Datastores • You are responsible for checking in all the dependent objects that were checked out during the owner renaming process. Checking in the new objects does not automatically check in the dependent objects that were checked out. • The software does not delete original objects from the central repository when you check in the new objects. • Use caution because checking in datastores and checking them out as multi-user operations can override datastore configurations. • Maintain the datastore configurations of all users by not overriding the configurations they created. Instead, add a configuration and make it your default configuration while working in your own environment. • When your group completes the development phase, It is recommended that the last developer delete the configurations that apply to the development environments and add the configurations that apply to the test or production environments. 5.5.7 Job portability tips • The software assumes that the metadata of a table or function is the same across different database types and versions specified in different configurations in the same datastore. For instance, if you import a table when the default configuration of the datastore is Oracle, then later use the table in a job to extract from DB2, your job will run. • Import metadata for a database object using the default configuration and use that same metadata with all configurations defined in the same datastore. • The software supports options in some database types or versions that it does not support in others For example, the software supports parallel reading on Oracle hash-partitioned tables, not on DB2 or other database hash-partitioned tables. If you import an Oracle hash-partitioned table and set your data flow to run in parallel, the software will read from each partition in parallel. However, when you run your job using sources from a DB2 environment, parallel reading will not occur. • The following features support job portability: • Enhanced SQL transform With the enhanced SQL transform, you can enter different SQL text for different database types/versions and use variable substitution in the SQL text to allow the software to read the correct text for its associated datastore configuration. • Enhanced target table editor Using enhanced target table editor options, you can configure database table targets for different database types/versions to match their datastore configurations. • Enhanced datastore editor Using the enhanced datastore editor, when you create a new datastore configuration you can choose to copy the database properties (including the datastore and table target options as well as the SQL transform text) from an existing configuration or use the current values. 116 2011-06-09
- 117. Datastores • When you design a job that will be run from different database types or versions, name database tables, functions, and stored procedures the same for all sources. If you create configurations for both case-insensitive databases and case-sensitive databases in the same datastore, It is recommended that you name the tables, functions, and stored procedures using all upper-case characters. • Table schemas should match across the databases in a datastore. This means the number of columns, the column names, and column positions should be exactly the same. The column data types should be the same or compatible. For example, if you have a VARCHAR column in an Oracle source, use a VARCHAR column in the Microsoft SQL Server source too. If you have a DATE column in an Oracle source, use a DATETIME column in the Microsoft SQL Server source. Define primary and foreign keys the same way. • Stored procedure schemas should match. When you import a stored procedure from one datastore configuration and try to use it for another datastore configuration, the software assumes that the signature of the stored procedure is exactly the same for the two databases. For example, if a stored procedure is a stored function (only Oracle supports stored functions), then you have to use it as a function with all other configurations in a datastore (in other words, all databases must be Oracle). If your stored procedure has three parameters in one database, it should have exactly three parameters in the other databases. Further, the names, positions, data types, and in/out types of the parameters must match exactly. Related Topics • Multi-user Development • Multi-user Environment Setup 5.5.8 Renaming table and function owner The software allows you to rename the owner of imported tables, template tables, or functions. This process is called owner renaming. Use owner renaming to assign a single metadata alias instead of the real owner name for database objects in the datastore. Consolidating metadata under a single alias name allows you to access accurate and consistent dependency information at any time while also allowing you to more easily switch between configurations when you move jobs to different environments. When using objects stored in a central repository, a shared alias makes it easy to track objects checked in by multiple users. If all users of local repositories use the same alias, the software can track dependencies for objects that your team checks in and out of the central repository. When you rename an owner, the instances of a table or function in a data flow are affected, not the datastore from which they were imported. 117 2011-06-09
- 118. Datastores 5.5.8.1 To rename the owner of a table or function 1. From the Datastore tab of the local object library, expand a table, template table, or function category. 2. Right-click the table or function and select Rename Owner. 3. Enter a New Owner Name then click Rename. When you enter a New Owner Name, the software uses it as a metadata alias for the table or function. Note: If the object you are renaming already exists in the datastore, the software determines if that the two objects have the same schema. If they are the same, then the software proceeds. If they are different, then the software displays a message to that effect. You may need to choose a different object name. The software supports both case-sensitive and case-insensitive owner renaming. • If the objects you want to rename are from a case-sensitive database, the owner renaming mechanism preserves case sensitivity. • If the objects you want to rename are from a datastore that contains both case-sensitive and case-insensitive databases, the software will base the case-sensitivity of new owner names on the case sensitivity of the default configuration. To ensure that all objects are portable across all configurations in this scenario, enter all owner names and object names using uppercase characters. During the owner renaming process: • The software updates the dependent objects (jobs, work flows, and data flows that use the renamed object) to use the new owner name. • The object library shows the entry of the object with the new owner name. Displayed Usage and Where-Used information reflect the number of updated dependent objects. • If the software successfully updates all the dependent objects, it deletes the metadata for the object with the original owner name from the object library and the repository. 5.5.8.2 Using the Rename window in a multi-user scenario This section provides a detailed description of Rename Owner window behavior in a multi-user scenario. Using an alias for all objects stored in a central repository allows the software to track all objects checked in by multiple users. If all local repository users use the same alias, the software can track dependencies for objects that your team checks in and out of the central repository. 118 2011-06-09
- 119. Datastores When you are checking objects in and out of a central repository, depending upon the check-out state of a renamed object and whether that object is associated with any dependent objects, there are several behaviors possible when you select the Rename button. Case 1 Object is not checked out, and object has no dependent objects in the local or central repository. Behavior: When you click Rename, the software renames the object owner. Case 2 Object is checked out, and object has no dependent objects in the local or central repository. Behavior: Same as Case 1. Case 3 Object is not checked out, and object has one or more dependent objects (in the local repository). Behavior: When you click Rename, the software displays a second window listing the dependent objects (that use or refer to the renamed object). If you click Continue, the software renames the objects and modifies the dependent objects to refer to the renamed object using the new owner name. If you click Cancel, the Designer returns to the Rename Owner window. Note: An object might still have one or more dependent objects in the central repository. However, if the object to be renamed is not checked out, the Rename Owner mechanism (by design) does not affect the dependent objects in the central repository. Case 4 Object is checked out and has one or more dependent objects. Behavior: This case contains some complexity. • If you are not connected to the central repository, the status message reads: This object is checked out from central repository X. Please select Tools | Central Repository… to activate that repository before renaming. • If you are connected to the central repository, the Rename Owner window opens. When you click Rename, a second window opens to display the dependent objects and a status indicating their check-out state and location. If a dependent object is located in the local repository only, the status message reads: Used only in local repository. No check out necessary. • If the dependent object is in the central repository, and it is not checked out, the status message reads: Not checked out • 119 If you have the dependent object checked out or it is checked out by another user, the status message shows the name of the checked out repository. For example: Oracle.production.user1 2011-06-09
- 120. Datastores As in Case 2, the purpose of this second window is to show the dependent objects. In addition, this window allows you to check out the necessary dependent objects from the central repository, without having to go to the Central Object Library window. Click the Refresh List button to update the check out status in the list. This is useful when the software identifies a dependent object in the central repository but another user has it checked out. When that user checks in the dependent object, click Refresh List to update the status and verify that the dependent object is no longer checked out. To use the Rename Owner feature to its best advantage, check out associated dependent objects from the central repository. This helps avoid having dependent objects that refer to objects with owner names that do not exist. From the central repository, select one or more objects, then right-click and select Check Out. After you check out the dependent object, the Designer updates the status. If the check out was successful, the status shows the name of the local repository. Case 4a You click Continue, but one or more dependent objects are not checked out from the central repository. In this situation, the software displays another dialog box that warns you about objects not yet checked out and to confirm your desire to continue. Click No to return to the previous dialog box showing the dependent objects. Click Yes to proceed with renaming the selected object and to edit its dependent objects. The software modifies objects that are not checked out in the local repository to refer to the new owner name. It is your responsibility to maintain consistency with the objects in the central repository. Case 4b You click Continue, and all dependent objects are checked out from the central repository. The software renames the owner of the selected object, and modifies all dependent objects to refer to the new owner name. Although to you, it looks as if the original object has a new owner name, in reality the software has not modified the original object; it created a new object identical to the original, but uses the new owner name. The original object with the old owner name still exists. The software then performs an "undo checkout" on the original object. It becomes your responsibility to check in the renamed object. When the rename operation is successful, in the Datastore tab of the local object library, the software updates the table or function with the new owner name and the Output window displays the following message: Object <Object_Name>: owner name <Old_Owner> successfully renamed to <New_Owner>, including references from dependent objects. If the software does not successfully rename the owner, the Output window displays the following message: Object <Object_Name>: Owner name <Old_Owner> could not be renamed to <New_Owner >. 120 2011-06-09
- 121. Datastores 5.5.9 Defining a system configuration What is the difference between datastore configurations and system configurations? • Datastore configurations — Each datastore configuration defines a connection to a particular database from a single datastore. • System configurations — Each system configuration defines a set of datastore configurations that you want to use together when running a job. You can define a system configuration if your repository contains at least one datastore with multiple configurations. You can also associate substitution parameter configurations to system configurations. When designing jobs, determine and create datastore configurations and system configurations depending on your business environment and rules. Create datastore configurations for the datastores in your repository before you create system configurations to organize and associate them. Select a system configuration to use at run-time. In many enterprises, a job designer defines the required datastore and system configurations and then a system administrator determines which system configuration to use when scheduling or starting a job. The software maintains system configurations separate from jobs. You cannot check in or check out system configurations in a multi-user environment. However, you can export system configurations to a separate flat file which you can later import. Related Topics • Creating a new configuration 5.5.9.1 To create a system configuration 1. From the Designer menu bar, select Tools > System Configurations. The "Edit System Configurations" window displays. 2. To add a new system configuration, do one of the following: • Click the Create New Configuration icon to add a configuration that references the default configuration of the substitution parameters and each datastore connection. • Select an existing configuration and click the Duplicate Configuration icon to create a copy of the selected configuration. You can use the copy as a template and edit the substitution parameter or datastore configuration selections to suit your needs. 3. If desired, rename the new system configuration. 121 2011-06-09
- 122. Datastores a. Select the system configuration you want to rename. b. Click the Rename Configuration icon to enable the edit mode for the configuration name field. c. Type a new, unique name and click outside the name field to accept your choice. It is recommended that you follow a consistent naming convention and use the prefix SC_ in each system configuration name so that you can easily identify this file as a system configuration. This practice is particularly helpful when you export the system configuration. 4. From the list, select a substitution parameter configuration to associate with the system configuration. 5. For each datastore, select the datastore configuration you want to use when you run a job using the system configuration. If you do not map a datastore configuration to a system configuration, the Job Server uses the default datastore configuration at run-time. 6. Click OK to save your system configuration settings. Related Topics • Associating a substitution parameter configuration with a system configuration 5.5.9.2 To export a system configuration 1. In the object library, select the Datastores tab and right-click a datastore. 2. Select Repository > Export System Configurations. It is recommended that you add the SC_ prefix to each exported system configuration .atl file to easily identify that file as a system configuration. 3. Click OK. 122 2011-06-09
- 123. File formats File formats This section discussed file formats, how to use the file format editor, and how to create a file format in the software. Related Topics • Reference Guide: File format 6.1 Understanding file formats A file format is a set of properties describing the structure of a flat file (ASCII). File formats describe the metadata structure. A file format describes a specific file. A file format template is a generic description that can be used for multiple data files. The software can use data stored in files for data sources and targets. A file format defines a connection to a file. Therefore, you use a file format to connect to source or target data when the data is stored in a file rather than a database table. The object library stores file format templates that you use to define specific file formats as sources and targets in data flows. To work with file formats, perform the following tasks: • • Create a file format template that defines the structure for a file. Create a specific source or target file format in a data flow. The source or target file format is based on a template and specifies connection information such as the file name. File format objects can describe files of the following types: • • • • • Delimited: Characters such as commas or tabs separate each field. Fixed width: You specify the column width. SAP transport: Use to define data transport objects in SAP application data flows. Unstructured text: Use to read one or more files of unstructured text from a directory. Unstructured binary: Use to read one or more binary documents from a directory. Related Topics • File formats 123 2011-06-09
- 124. File formats 6.2 File format editor Use the file format editor to set properties for file format templates and source and target file formats. Available properties vary by the mode of the file format editor: • New mode — Create a new file format template • Edit mode — Edit an existing file format template • Source mode — Edit the file format of a particular source file • Target mode — Edit the file format of a particular target file The file format editor has three work areas: • Properties-Values — Edit the values for file format properties. Expand and collapse the property groups by clicking the leading plus or minus. • Column Attributes — Edit and define the columns or fields in the file. Field-specific formats override the default format set in the Properties-Values area. • Data Preview — View how the settings affect sample data. The file format editor contains "splitter" bars to allow resizing of the window and all the work areas. You can expand the file format editor to the full screen size. The properties and appearance of the work areas vary with the format of the file. 124 2011-06-09
- 125. File formats You can navigate within the file format editor as follows: • Switch between work areas using the Tab key. • Navigate through fields in the Data Preview area with the Page Up, Page Down, and arrow keys. • Open a drop-down menu in the Properties-Values area by pressing the ALT-down arrow key combination. • When the file format type is fixed-width, you can also edit the column metadata structure in the Data Preview area. Note: The Show ATL button displays a view-only copy of the Transformation Language file generated for your file format. You might be directed to use this by SAP Business User Suppport. Related Topics • Reference Guide: File format 125 2011-06-09
- 126. File formats 6.3 Creating file formats To specify a source or target file, you create a file format template that defines the structure for a file. When you drag and drop the file format into a data flow; the format represents a file that is based on the template and specifies connection information such as the file name. 6.3.1 To create a new file format 1. In the local object library, go to the Formats tab, right-click Flat Files, and select New. 2. For Type, select: • • • • Delimited: For a file that uses a character sequence to separate columns. Fixed width: For a file that uses specified widths for each column. SAP transport: For data transport objects in SAP application data flows. Unstructured text: For one or more files of unstructured text from a directory. The schema is fixed for this type. • Unstructured binary: For one or more unstructured text and binary documents from a directory. The schema is fixed for this type. The options change in the editor based on the type selected. 3. For Name, enter a name that describes this file format template. After you save this file format template, you cannot change the name. 4. For Delimited and Fixed width files, you can read and load files using a third-party file-transfer program by selecting Yes for Custom transfer program. 5. Complete the other properties to describe files that this template represents. Look for properties available when the file format editor is in source mode or target mode. 6. For source files, some file formats let you specify the structure of the columns in the Column Attributes work area (the upper-right pane): a. Enter field name. b. Set data types. c. Enter field sizes for data types. d. Enter scale and precision information for decimal and numeric and data types. e. Enter the Content Type. If you have added a column while creating a new format, the content type might be provided for you based on the field name. If an appropriate content type is not available, it defaults to blank. f. Enter information in the Format field for appropriate data types if desired. This information overrides the default format set in the Properties-Values area for that data type. You can model a file format on a sample file. 126 2011-06-09
- 127. File formats Note: • • You do not need to specify columns for files used as targets. If you do specify columns and they do not match the output schema from the preceding transform, the software writes to the target file using the transform's output schema. For a decimal or real data type, if you only specify a source column format and the column names and data types in the target schema do not match those in the source schema, the software cannot use the source column format specified. Instead, it defaults to the format used by the code page on the computer where the Job Server is installed. 7. Click Save & Close to save the file format template and close the file format editor. Related Topics • Reference Guide: Locales and Multi-byte Functionality • File transfers • Reference Guide: File format 6.3.2 Modeling a file format on a sample file 1. From the Formats tab in the local object library, create a new flat file format template or edit an existing flat file format template. 2. Under Data File(s): • If the sample file is on your Designer computer, set Location to Local. Browse to set the Root directory and File(s) to specify the sample file. Note: During design, you can specify a file located on the computer where the Designer runs or on the computer where the Job Server runs. Indicate the file location in the Location property. During execution, you must specify a file located on the Job Server computer that will execute the job. • If the sample file is on the current Job Server computer, set Location to Job Server. Enter the Root directory and File(s) to specify the sample file. When you select Job Server, the Browse icon is disabled, so you must type the path to the file. You can type an absolute path or a relative path, but the Job Server must be able to access it. For example, a path on UNIX might be /usr/data/abc.txt. A path on Windows might be C:DATAabc.txt. Note: In the Windows operating system, files are not case-sensitive; however, file names are case sensitive in the UNIX environment. (For example, abc.txt and aBc.txt would be two different files in the same UNIX directory.) To reduce the risk of typing errors, you can telnet to the Job Server (UNIX or Windows) computer and find the full path name of the file you want to use. Then, copy and paste the path name from the telnet application directly into the Root directory text box in the file format editor. You cannot use the Windows Explorer to determine the exact file location on Windows. 127 2011-06-09
- 128. File formats 3. If the file type is delimited, set the appropriate column delimiter for the sample file. You can choose from the drop-down list or specify Unicode delimiters by directly typing the Unicode character code in the form of /XXXX, where XXXX is a decimal Unicode character code. For example, /44 is the Unicode character for the comma (,) character. 4. Under Input/Output, set Skip row header to Yes if you want to use the first row in the file to designate field names. The file format editor will show the column names in the Data Preview area and create the metadata structure automatically. 5. Edit the metadata structure as needed. For both delimited and fixed-width files, you can edit the metadata structure in the Column Attributes work area: a. b. c. d. e. f. Right-click to insert or delete fields. Rename fields. Set data types. Enter field lengths for the Blob and VarChar data type. Enter scale and precision information for Numeric and Decimal data types. Enter Format field information for appropriate data types, if desired. This format information overrides the default format set in the Properties-Values area for that data type. g. Enter the Content Type information. You do not need to specify columns for files used as targets. If you have added a column while creating a new format, the content type may auto-fill based on the field name. If an appropriate content type cannot be automatically filled, then it will default to blank. For fixed-width files, you can also edit the metadata structure in the Data Preview area: a. Click to select and highlight columns. b. Right-click to insert or delete fields. Note: The Data Preview pane cannot display blob data. 6. Click Save & Close to save the file format template and close the file format editor. 6.3.3 Replicating and renaming file formats After you create one file format schema, you can quickly create another file format object with the same schema by replicating the existing file format and renaming it. To save time in creating file format objects, replicate and rename instead of configuring from scratch. 6.3.3.1 To create a file format from an existing file format 128 2011-06-09
- 129. File formats 1. In the Formats tab of the object library, right-click an existing file format and choose Replicate from the menu. The File Format Editor opens, displaying the schema of the copied file format. 2. Double-click to select the Name property value (which contains the same name as the original file format object). 3. Type a new, unique name for the replicated file format. Note: You must enter a new name for the replicated file. The software does not allow you to save the replicated file with the same name as the original (or any other existing File Format object). Also, this is your only opportunity to modify the Name property value. Once saved, you cannot modify the name again. 4. Edit other properties as desired. Look for properties available when the file format editor is in source mode or target mode. 5. To save and view your new file format schema, click Save. To terminate the replication process (even after you have changed the name and clicked Save), click Cancel or press the Esc button on your keyboard. 6. Click Save & Close. Related Topics • Reference Guide: File format 6.3.4 To create a file format from an existing flat table schema 1. From the Query editor, right-click a schema and select Create File format. The File Format editor opens populated with the schema you selected. 2. Edit the new schema as appropriate and click Save & Close. The software saves the file format in the repository. You can access it from the Formats tab of the object library. 6.3.5 To create a specific source or target file 1. Select a flat file format template on the Formats tab of the local object library. 2. Drag the file format template to the data flow workspace. 129 2011-06-09
- 130. File formats 3. Select Make Source to define a source file format, or select Make Target to define a target file format. 4. Click the name of the file format object in the workspace to open the file format editor. 5. Enter the properties specific to the source or target file. Look for properties available when the file format editor is in source mode or target mode. Under File name(s), be sure to specify the file name and location in the File and Location properties. Note: You can use variables as file names. 6. Connect the file format object to other objects in the data flow as appropriate. Related Topics • Reference Guide: File format • Setting file names at run-time using variables 6.4 Editing file formats You can modify existing file format templates to match changes in the format or structure of a file. You cannot change the name of a file format template. For example, if you have a date field in a source or target file that is formatted as mm/dd/yy and the data for this field changes to the format dd-mm-yy due to changes in the program that generates the source file, you can edit the corresponding file format template and change the date format information. For specific source or target file formats, you can edit properties that uniquely define that source or target such as the file name and location. Caution: If the template is used in other jobs (usage is greater than 0), changes that you make to the template are also made in the files that use the template. 6.4.1 To edit a file format template 1. In the object library Formats tab, double-click an existing flat file format (or right-click and choose Edit). The file format editor opens with the existing format values. 2. Edit the values as needed. 130 2011-06-09
- 131. File formats Look for properties available when the file format editor is in source mode or target mode. Caution: If the template is used in other jobs (usage is greater than 0), changes that you make to the template are also made in the files that use the template. 3. Click Save. Related Topics • Reference Guide: File format 6.4.2 To edit a source or target file 1. From the workspace, click the name of a source or target file. The file format editor opens, displaying the properties for the selected source or target file. 2. Edit the desired properties. Look for properties available when the file format editor is in source mode or target mode. To change properties that are not available in source or target mode, you must edit the file's file format template. Any changes you make to values in a source or target file editor override those on the original file format. 3. Click Save. Related Topics • Reference Guide: File format 6.4.3 Change multiple column properties Use these steps when you are creating a new file format or editing an existing one. 1. Select the "Format" tab in the Object Library. 2. Right-click on an existing file format listed under Flat Files and choose Edit. The "File Format Editor "opens. 3. In the column attributes area (upper right pane) select the multiple columns that you want to change. • To choose a series of columns, select the first column and press the keyboard "Shift" key and select the last column. 131 2011-06-09
- 132. File formats • To choose non-consecutive columns hold down the keyboard "Control" key and select the columns. 4. Right click and choose Properties. The "Multiple Columns Properties "window opens. 5. Change the Data Type and/or the Content Type and click Ok. The Data Type and Content Type of the selected columns change based on your settings. 6.5 File format features The software offers several capabilities for processing files. 6.5.1 Reading multiple files at one time The software can read multiple files with the same format from a single directory using a single source object. 6.5.1.1 To specify multiple files to read 1. Open the editor for your source file format 2. Under Data File(s) in the file format editor, set the Location of the source files to Local or Job Server. 3. Set the root directory in Root directory. Note: If your Job Server is on a different computer than the Designer, you cannot use Browse to specify the root directory. You must type the path. You can type an absolute path or a relative path, but the Job Server must be able to access it. 4. Under File name(s), enter one of the following: • A list of file names separated by commas, or • A file name containing a wild card character (* or ?). For example: 1999????.txt might read files from the year 1999 132 2011-06-09
- 133. File formats *.txt reads all files with the txt extension from the specified Root directory 6.5.2 Identifying source file names You might want to identify the source file for each row in your target in the following situations: • You specified a wildcard character to read multiple source files at one time • You load from different source files on different runs 6.5.2.1 To identify the source file for each row in the target 1. Under Source Information in the file format editor, set Include file name to Yes. This option generates a column named DI_FILENAME that contains the name of the source file. 2. In the Query editor, map the DI_FILENAME column from Schema In to Schema Out. 3. When you run the job, the DI_FILENAME column for each row in the target contains the source file name. 6.5.3 Number formats The dot (.) and the comma (,) are the two most common formats used to determine decimal and thousand separators for numeric data types. When formatting files in the software, data types in which these symbols can be used include Decimal, Numeric, Float, and Double. You can use either symbol for the thousands indicator and either symbol for the decimal separator. For example: 2,098.65 or 2.089,65. 133 2011-06-09
- 134. File formats Format Description {none} The software expects that the number contains only the decimal separator. The reading of the number data and this decimal separator is determined by Data Service Job Server Locale Region. Comma (,) is the decimal separator when is Data Service Locale is set to a country that uses commas (for example, Germany or France). Dot (.) is the decimal separator when Locale is set to country that uses dots (for example, USA, India, and UK). In this format, the software will return an error if a number contains a thousand separator. When the software writes the data, it only uses the Job Server Locale decimal separator. It does not use thousand separators. #,##0.0 The software expects that the decimal separator of a number will be a dot (.) and the thousand separator will be a comma (,). When the software loads the data to a flat file, it uses a comma (,) as the thousand separator and a dot (.) as decimal separator. #.##0,0 The software expects that the decimal separator of a number will be a comma (,) and the thousand separator will be dot (.). When the software loads the data to a flat file, it uses a dot (.) as the thousand separator and comma (,) as decimal separator. Leading and trailing decimal signs are also supported. For example: +12,000.00 or 32.32-. 6.5.4 Ignoring rows with specified markers The file format editor provides a way to ignore rows containing a specified marker (or markers) when reading files. For example, you might want to ignore comment line markers such as # and //. Associated with this feature, two special characters — the semicolon (;) and the backslash () — make it possible to define multiple markers in your ignore row marker string. Use the semicolon to delimit each marker, and use the backslash to indicate special characters as markers (such as the backslash and the semicolon). The default marker value is an empty string. When you specify the default value, no rows are ignored. 6.5.4.1 To specify markers for rows to ignore 1. Open the file format editor from the Object Library or by opening a source object in the workspace. 2. Find Ignore row marker(s) under the Format Property. 134 2011-06-09
- 135. File formats 3. Click in the associated text box and enter a string to indicate one or more markers representing rows that the software should skip during file read and/or metadata creation. The following table provides some ignore row marker(s) examples. (Each value is delimited by a semicolon unless the semicolon is preceded by a backslash.) Marker Value(s) Row(s) Ignored None (this is the default value) abc Any that begin with the string abc abc;def;hi Any that begin with abc or def or hi abc;; Any that begin with abc or ; abc;;; Any that begin with abc or or ; 6.5.5 Date formats at the field level You can specify a date format at the field level to overwrite the default date, time, or date-time formats set in the Properties-Values area. For example, when the Data Type is set to Date, you can edit the value in the corresponding Format field to a different date format such as: • yyyy.mm.dd • mm/dd/yy • dd.mm.yy 6.5.6 Parallel process threads Data Services can use parallel threads to read and load files to maximize performance. To specify parallel threads to process your file format: 1. Open the file format editor in one of the following ways: • • In the Formats tab in the Object Library, right-click a file format name and click Edit. In the workspace, double-click the source or target object. 2. Find Parallel process threads under the "General" Property. 3. Specify the number of threads to read or load this file format. 135 2011-06-09
- 136. File formats For example, if you have four CPUs on your Job Server computer, enter the number 4 in the Parallel process threads box. Related Topics • Performance Optimization Guide: Using Parallel Execution, File multi-threading 6.5.7 Error handling for flat-file sources During job execution, the software processes rows from flat-file sources one at a time. You can configure the File Format Editor to identify rows in flat-file sources that contain the following types of errors: • Data-type conversion errors — For example, a field might be defined in the File Format Editor as having a data type of integer but the data encountered is actually varchar. • Row-format errors — For example, in the case of a fixed-width file, the software identifies a row that does not match the expected width value. These error-handling properties apply to flat-file sources only. Related Topics • Reference Guide: File format 6.5.7.1 Error-handling options In the File Format Editor, the Error Handling set of properties allows you to choose whether or not to have the software perform the following actions: • check for either of the two types of flat-file source error • write the invalid row(s) to a specified error file • stop processing the source file after reaching a specified number of invalid rows • log data-type conversion or row-format warnings to the error log; if so, you can limit the number of warnings to log without stopping the job 6.5.7.2 About the error file 136 2011-06-09
- 137. File formats If enabled, the error file will include both types of errors. The format is a semicolon-delimited text file. You can have multiple input source files for the error file. The file resides on the same computer as the Job Server. Entries in an error file have the following syntax: source file path and name; row number in source file; Data Services error; column number where the error occurred; all columns from the invalid row The following entry illustrates a row-format error: d:/acl_work/in_test.txt;2;-80104: 1-3-A column delimiter was seen after column number <3> for row number <2> in file <d:/acl_work/in_test.txt>. The total number of columns defined is <3>, so a row delimiter should be seen after column number <3>. Please check the file for bad data, or redefine the input schema for the file by editing the file format in the UI.;3;defg;234;def where 3 indicates an error occurred after the third column, and defg;234;def are the three columns of data from the invalid row. Note: If you set the file format's Parallel process thread option to any value greater than 0 or {none}, the row number in source file value will be -1. 6.5.7.3 Configuring the File Format Editor for error handling 6.5.7.3.1 To capture data-type conversion or row-format errors 1. In the object library, click the Formats tab. 2. Expand Flat Files, right-click a format, and click Edit. 3. The File Format Editor opens. 4. To capture data-type conversion errors, under the Error Handling properties for Capture data conversion errors, click Yes. 5. To capture errors in row formats, for Capture row format errors click Yes. 6. Click Save or Save & Close. 6.5.7.3.2 To write invalid rows to an error file 1. In the object library, click the Formats tab. 2. Expand Flat Files, right-click a format, and click Edit. The File Format Editor opens. 3. Under the Error Handling properties, click Yes for either or both of the Capture data conversion errors or Capture row format errors properties. 4. For Write error rows to file, click Yes. Two more fields appear: Error file root directory and Error file name. 5. Type an Error file root directory in which to store the error file. 137 2011-06-09
- 138. File formats If you type a directory path here, then enter only the file name in the Error file name property. 6. Type an Error file name. If you leave Error file root directory blank, then type a full path and file name here. 7. Click Save or Save & Close. For added flexibility when naming the error file, you can enter a variable that is set to a particular file with full path name. Use variables to specify file names that you cannot otherwise enter such as those that contain multibyte characters 6.5.7.3.3 To limit to the number of invalid rows processed before stopping the job 1. In the object library, click the Formats tab. 2. Expand Flat Files, right-click a format, and click Edit. The File Format Editor opens. 3. Under the Error Handling properties, click Yes for either or both the Capture data conversion errors or Capture row format errors properties. 4. For Maximum errors to stop job, type a number. Note: This property was previously known as Bad rows limit. 5. Click Save or Save & Close. 6.5.7.3.4 To log data-type conversion warnings in the error log 1. In the object library, click the Formats tab. 2. Expand Flat Files, right-click a format, and click Edit. The File Format Editor opens. 3. Under the Error Handling properties, for click Yes. 4. Click Save or Save & Close. 6.5.7.3.5 To log row-format warnings in the error log 1. In the object library, click the Formats tab. 2. Expand Flat Files, right-click a format, and click Edit. The File Format Editor opens. 3. Under the Error Handling properties, for click Yes. 4. Click Save or Save & Close. 6.5.7.3.6 To limit to the number of warning messages to log If you choose to log either data-type or row-format warnings, you can limit the total number of warnings to log without interfering with job execution. 138 2011-06-09
- 139. File formats 1. In the object library, click the Formats tab. 2. Expand Flat Files, right-click a format, and click Edit. The File Format Editor opens. 3. Under the Error Handling properties, for click Yes. 4. For Maximum warnings to log, type a number. 5. Click Save or Save & Close. 6.6 File transfers The software can read and load files using a third-party file transfer program for flat files. You can use third-party (custom) transfer programs to: • Incorporate company-standard file-transfer applications as part of the software job execution • Provide high flexibility and security for files transferred across a firewall The custom transfer program option allows you to specify: • A custom transfer program (invoked during job execution) • Additional arguments, based on what is available in your program, such as: • Connection data • Encryption/decryption mechanisms • Compression mechanisms 6.6.1 Custom transfer system variables for flat files When you set custom transfer options for external file sources and targets, some transfer information, like the name of the remote server that the file is being transferred to or from, may need to be entered literally as a transfer program argument. You can enter other information using the following system variables: Data entered for: User name 139 Is substituted for this variable if it is defined in the Arguments field $AW_USER 2011-06-09
- 140. File formats Data entered for: Is substituted for this variable if it is defined in the Arguments field Password $AW_PASSWORD Local directory $AW_LOCAL_DIR File(s) $AW_FILE_NAME By using these variables as custom transfer program arguments, you can collect connection information entered in the software and use that data at run-time with your custom transfer program. For example, the following custom transfer options use a Windows command file (Myftp.cmd) with five arguments. Arguments 1 through 4 are system variables: • User and Password variables are for the external server • The Local Directory variable is for the location where the transferred files will be stored in the software • The File Name variable is for the names of the files to be transferred Argument 5 provides the literal external server name. Note: If you do not specify a standard output file (such as ftp.out in the example below), the software writes the standard output into the job's trace log. @echo off set set set set set USER=%1 PASSWORD=%2 LOCAL_DIR=%3 FILE_NAME=%4 LITERAL_HOST_NAME=%5 set INP_FILE=ftp.inp echo echo echo echo echo %USER%>%INP_FILE% %PASSWORD%>>%INP_FILE% lcd %LOCAL_DIR%>>%INP_FILE% get %FILE_NAME%>>%INP_FILE% bye>>%INP_FILE% ftp -s%INPT_FILE% %LITERAL_HOST_NAME%>ftp.out 6.6.2 Custom transfer options for flat files Of the custom transfer program options, only the Program executable option is mandatory. 140 2011-06-09
- 141. File formats Entering User Name, Password, and Arguments values is optional. These options are provided for you to specify arguments that your custom transfer program can process (such as connection data). You can also use Arguments to enable or disable your program's built-in features such as encryption/decryption and compression mechanisms. For example, you might design your transfer program so that when you enter -sSecureTransportOn or -CCompressionYES security or compression is enabled. Note: Available arguments depend on what is included in your custom transfer program. See your custom transfer program documentation for a valid argument list. You can use the Arguments box to enter a user name and password. However, the software also provides separate User name and Password boxes. By entering the $AW_USER and $AW_PASSWORD variables as Arguments and then using the User and Password boxes to enter literal strings, these extra boxes are useful in two ways: • You can more easily update users and passwords in the software both when you configure the software to use a transfer program and when you later export the job. For example, when you migrate the job to another environment, you might want to change login information without scrolling through other arguments. • You can use the mask and encryption properties of the Password box. Data entered in the Password box is masked in log files and on the screen, stored in the repository, and encrypted by Data Services. Note: The software sends password data to the custom transfer program in clear text. If you do not allow clear passwords to be exposed as arguments in command-line executables, then set up your custom program to either: • Pick up its password from a trusted location • Inherit security privileges from the calling program (in this case, the software) 6.6.3 Setting custom transfer options The custom transfer option allows you to use a third-party program to transfer flat file sources and targets. You can configure your custom transfer program in the File Format Editor window. Like other file format settings, you can override custom transfer program settings if they are changed for a source or target in a particular data flow. You can also edit the custom transfer option when exporting a file format. 6.6.3.1 To configure a custom transfer program in the file format editor 141 2011-06-09
- 142. File formats 1. Select the Formats tab in the object library. 2. Right-click Flat Files in the tab and select New. The File Format Editor opens. 3. Select either the Delimited or the Fixed width file type. Note: While the custom transfer program option is not supported by SAP application file types, you can use it as a data transport method for an SAP ABAP data flow. 4. Enter a format name. 5. Select Yes for the Custom transfer program option. 6. Expand "Custom Transfer" and enter the custom transfer program name and arguments. 7. Complete the other boxes in the file format editor window. In the Data File(s) section, specify the location of the file in the software. To specify system variables for Root directory and File(s) in the Arguments box: • Associate the system variable $AW_LOCAL_DIR with the local directory argument of your custom transfer program. • Associate the system variable $AW_FILE_NAME with the file name argument of your custom transfer program. For example, enter: -l$AW_LOCAL_DIR$AW_FILE_NAME When the program runs, the Root directory and File(s) settings are substituted for these variables and read by the custom transfer program. Note: The flag -l used in the example above is a custom program flag. Arguments you can use as custom program arguments in the software depend upon what your custom transfer program expects. 8. Click Save. Related Topics • Supplement for SAP: Custom Transfer method • Reference Guide: File format 6.6.4 Design tips Keep the following concepts in mind when using the custom transfer options: • 142 Variables are not supported in file names when invoking a custom transfer program for the file. 2011-06-09
- 143. File formats • You can only edit custom transfer options in the File Format Editor (or Datastore Editor in the case of SAP application) window before they are exported. You cannot edit updates to file sources and targets at the data flow level when exported. After they are imported, you can adjust custom transfer option settings at the data flow level. They override file format level settings. When designing a custom transfer program to work with the software, keep in mind that: • The software expects the called transfer program to return 0 on success and non-zero on failure. • The software provides trace information before and after the custom transfer program executes. The full transfer program and its arguments with masked password (if any) is written in the trace log. When "Completed Custom transfer" appears in the trace log, the custom transfer program has ended. • If the custom transfer program finishes successfully (the return code = 0), the software checks the following: • For an ABAP data flow, if the transport file does not exist in the local directory, it throws an error and the software stops. • For a file source, if the file or files to be read by the software do not exist in the local directory, the software writes a warning message into the trace log. • If the custom transfer program throws an error or its execution fails (return code is not 0), then the software produces an error with return code and stdout/stderr output. • If the custom transfer program succeeds but produces standard output, the software issues a warning, logs the first 1,000 bytes of the output produced, and continues processing. • The custom transfer program designer must provide valid option arguments to ensure that files are transferred to and from the local directory (specified in the software). This might require that the remote file and directory name be specified as arguments and then sent to the Designer interface using system variables. Related Topics • Supplement for SAP: Custom Transfer method 6.7 Creating COBOL copybook file formats When creating a COBOL copybook format, you can: • create just the format, then configure the source after you add the format to a data flow, or • create the format and associate it with a data file at the same time This section also describes how to: • • 143 create rules to identify which records represent which schemas using a field ID option identify the field that contains the length of the schema's record using a record length field option 2011-06-09
- 144. File formats Related Topics • Reference Guide: Import or Edit COBOL copybook format options • Reference Guide: COBOL copybook source options • Reference Guide: Data Types, Conversion to or from internal data types 6.7.1 To create a new COBOL copybook file format 1. In the local object library, click the Formats tab, right-click COBOL copybooks, and click New. The Import COBOL copybook window opens. 2. Name the format by typing a name in the Format name field. 3. On the Format tab for File name, specify the COBOL copybook file format to import, which usually has the extension .cpy. During design, you can specify a file in one of the following ways: • For a file located on the computer where the Designer runs, you can use the Browse button. • For a file located on the computer where the Job Server runs, you must type the path to the file. You can type an absolute path or a relative path, but the Job Server must be able to access it. 4. Click OK. The software adds the COBOL copybook to the object library. 5. The COBOL Copybook schema name(s) dialog box displays. If desired, select or double-click a schema name to rename it. 6. Click OK. When you later add the format to a data flow, you can use the options in the source editor to define the source. Related Topics • Reference Guide: COBOL copybook source options 6.7.2 To create a new COBOL copybook file format and a data file 1. In the local object library, click the Formats tab, right-click COBOL copybooks, and click New. The Import COBOL copybook window opens. 2. Name the format by typing a name in the Format name field. 144 2011-06-09
- 145. File formats 3. On the Format tab for File name, specify to the COBOL copybook file format to import, which usually has the extension .cpy. During design, you can specify a file in one of the following ways: • For a file located on the computer where the Designer runs, you can use the Browse button. • For a file located on the computer where the Job Server runs, you must type the path to the file. You can type an absolute path or a relative path, but the Job Server must be able to access it. 4. Click the Data File tab. 5. For Directory, type or browse to the directory that contains the COBOL copybook data file to import. If you include a directory path here, then enter only the file name in the Name field. 6. Specify the COBOL copybook data file Name. If you leave Directory blank, then type a full path and file name here. During design, you can specify a file in one of the following ways: • For a file located on the computer where the Designer runs, you can use the Browse button. • For a file located on the computer where the Job Server runs, you must type the path to the file. You can type an absolute path or a relative path, but the Job Server must be able to access it. 7. If the data file is not on the same computer as the Job Server, click the Data Access tab. Select FTP or Custom and enter the criteria for accessing the data file. 8. Click OK. 9. The COBOL Copybook schema name(s) dialog box displays. If desired, select or double-click a schema name to rename it. 10. Click OK. The Field ID tab allows you to create rules for indentifying which records represent which schemas. Related Topics • Reference Guide: Import or Edit COBOL copybook format options 6.7.3 To create rules to identify which records represent which schemas 1. In the local object library, click the Formats tab, right-click COBOL copybooks, and click Edit. The Edit COBOL Copybook window opens. 2. In the top pane, select a field to represent the schema. 3. Click the Field ID tab. 4. On the Field ID tab, select the check box Use field <schema name.field name> as ID. 5. Click Insert below to add an editable value to the Values list. 145 2011-06-09
- 146. File formats 6. 7. 8. 9. Type a value for the field. Continue (adding) inserting values as necessary. Select additional fields and insert values as necessary. Click OK. 6.7.4 To identify the field that contains the length of the schema's record 1. In the local object library, click the Formats tab, right-click COBOL copybooks, and click Edit. The Edit COBOL Copybook window opens. 2. Click the Record Length Field tab. 3. For the schema to edit, click in its Record Length Field column to enable a drop-down menu. 4. Select the field (one per schema) that contains the record's length. The offset value automatically changes to the default of 4; however, you can change it to any other numeric value. The offset is the value that results in the total record length when added to the value in the Record length field. 5. Click OK. 6.8 Creating Microsoft Excel workbook file formats on UNIX platforms This section describes how to use a Microsoft Excel workbook as a source with a Job Server on a UNIX platform. To create Microsoft Excel workbook file formats on Windows, refer to the Reference Guide. To access the workbook, you must create and configure an adapter instance in the Administrator. The following procedure provides an overview of the configuration process. For details about creating adapters, refer to the Management Console Guide. Also consider the following requirements: • To import the workbook, it must be available on a Windows file system. You can later change the location of the actual file to use for processing in the Excel workbook file format source editor. See the Reference Guide. • To reimport or view data in the Designer, the file must be available on Windows. • Entries in the error log file might be represented numerically for the date and time fields. Additionally, Data Services writes the records with errors to the output (in Windows, these records are ignored). 146 2011-06-09
- 147. File formats Related Topics • Reference Guide: Excel workbook format • Management Console Guide: Adapters • Reference Guide: Excel workbook source options 6.8.1 To create a Microsoft Excel workbook file format on UNIX 1. Using the Server Manager ($LINK_DIR/bin/svrcfg), ensure the UNIX Job Server can support adapters. See the Installation Guide for UNIX. 2. Ensure a repository associated with the Job Server has been added to the Administrator. To add a repository to the Administrator, see the Management Console Guide. 3. In the Administrator, add an adapter to access Excel workbooks. See the Management Console Guide. You can only configure one Excel adapter per Job Server. Use the following options: • On the Installed Adapters tab, select MSExcelAdapter. • On the Adapter Configuration tab for the Adapter instance name, type BOExcelAdapter (required and case sensitive). You may leave all other options at their default values except when processing files larger than 1 MB. In that case, change the Additional Java Launcher Options value to -Xms64m -Xmx512 or -Xms128m -Xmx1024m (the default is -Xms64m -Xmx256m). Note that Java memory management can prevent processing very large files (or many smaller files). 4. Start the adapter. 5. In the Designer on the "Formats" tab of the object library, create the file format by importing the Excel workbook. For details, see the Reference Guide. Related Topics • Management Console Guide: Adding repositories • Management Console Guide: Adding and configuring adapter instances • Reference Guide: Excel workbook format 6.9 Creating Web log file formats Web logs are flat files generated by Web servers and are used for business intelligence. Web logs typically track details of Web site hits such as: • 147 Client domain names or IP addresses 2011-06-09
- 148. File formats • User names • Timestamps • Requested action (might include search string) • Bytes transferred • Referred address • Cookie ID Web logs use a common file format and an extended common file format. Common Web log format: 151.99.190.27 - - [01/Jan/1997:13:06:51 -0600] "GET /~bacuslab HTTP/1.0" 301 -4 Extended common Web log format: saturn5.cun.com - - [25/JUN/1998:11:19:58 -0500] "GET /wew/js/mouseover.html HTTP/1.0" 200 1936 "http://av.yahoo.com/bin/query?p=mouse+over+javascript+source+code&hc=0" "Mozilla/4.02 [en] (x11; U; SunOS 5.6 sun4m)" The software supports both common and extended common Web log formats as sources. The file format editor also supports the following: • • Dash as NULL indicator Time zone in date-time, e.g. 01/Jan/1997:13:06:51 –0600 The software includes several functions for processing Web log data: • • • Word_ext function Concat_data_time function WL_GetKeyValue function Related Topics • Word_ext function • Concat_date_time function • WL_GetKeyValue function 6.9.1 Word_ext function The word_ext is a string function that extends the word function by returning the word identified by its position in a delimited string. This function is useful for parsing URLs or file names. Format word_ext(string, word_number, separator(s)) A negative word number means count from right to left 148 2011-06-09
- 149. File formats Examples word_ext('www.bodi.com', 2, '.') returns 'bodi'. word_ext('www.cs.wisc.edu', -2, '.') returns 'wisc'. word_ext('www.cs.wisc.edu', 5, '.') returns NULL. word_ext('aaa+=bbb+=ccc+zz=dd', 4, '+=') returns 'zz'. If 2 separators are specified (+=), the function looks for either one. word_ext(',,,,,aaa,,,,bb,,,c ', 2, '.') returns 'bb'. This function skips consecutive delimiters. 6.9.2 Concat_date_time function The concat_date_time is a date function that returns a datetime from separate date and time inputs. Format concat_date_time(date, time) Example concat_date_time(MS40."date",MS40."time") 6.9.3 WL_GetKeyValue function The WL_GetKeyValue is a custom function (written in the Scripting Language) that returns the value of a given keyword. It is useful for parsing search strings. Format WL_GetKeyValue(string, keyword) Example A search in Google for bodi B2B is recorded in a Web log as: GET "http://www.google.com/search?hl=en&lr=&safe=off&q=bodi+B2B&btnG=Google+Search" WL_GetKeyValue('http://www.google.com/search?hl=en&lr=&safe=off&q=bodi+B2B&btnG=Google+Search','q') returns 'bodi+B2B'. 6.10 Unstructured file formats 149 2011-06-09
- 150. File formats Unstructured file formats are a type of flat file format. To create them, see Creating file formats. To read files that contain unstructured content, create a file format as a source that reads one or more files from a directory. At runtime, the source object in the data flow produces one row per file and contains a reference to each file to access its content. In the data flow, you can use a Text Data Processing transform such as Entity Extraction to process unstructured text or employ another transform to manipulate the data. The unstructured file format types include: • • Unstructured text: Use this format to process a directory of text-based files such as text, HTML, or XML. Data Services stores each file's content using the long data type. Unstructured binary: Use this format to read binary documents. Data Services stores each file's content using the blob data type. For example, you could use the unstructured binary file format to move a directory of graphic files on disk into a database table. Suppose you want to associate employee photos with the corresponding employee data that is stored in a database. The data flow would include the unstructured binary file format source, a Query transform that associates the employee photo with the employee data using the employee's ID number for example, and the database target table. Related Topics • Creating file formats • Reference Guide: Objects, File format • Text Data Processing overview 150 2011-06-09
- 151. Data Flows Data Flows This section describes the fundamantals of data flows including data flow objects, using lookups, data flow execution, and auditing. 7.1 What is a data flow? Data flows extract, transform, and load data. Everything having to do with data, including reading sources, transforming data, and loading targets, occurs inside a data flow. The lines connecting objects in a data flow represent the flow of data through data transformation steps. After you define a data flow, you can add it to a job or work flow. From inside a work flow, a data flow can send and receive information to and from other objects through input and output parameters. 7.1.1 Naming data flows Data flow names can include alphanumeric characters and underscores (_). They cannot contain blank spaces. 7.1.2 Data flow example Suppose you want to populate the fact table in your data warehouse with new data from two tables in your source transaction database. 151 2011-06-09
- 152. Data Flows Your data flow consists of the following: • Two source tables • A join between these tables, defined in a query transform • A target table where the new rows are placed You indicate the flow of data through these components by connecting them in the order that data moves through them. The resulting data flow looks like the following: 7.1.3 Steps in a data flow Each icon you place in the data flow diagram becomes a step in the data flow. You can use the following objects as steps in a data flow: • • • source target transforms The connections you make between the icons determine the order in which the software completes the steps. Related Topics • Source and target objects • Transforms 7.1.4 Data flows as steps in work flows Data flows are closed operations, even when they are steps in a work flow. Data sets created within a data flow are not available to other steps in the work flow. A work flow does not operate on data sets and cannot provide more data to a data flow; however, a work flow can do the following: 152 2011-06-09
- 153. Data Flows • Call data flows to perform data movement operations • Define the conditions appropriate to run data flows • Pass parameters to and from data flows 7.1.5 Intermediate data sets in a data flow Each step in a data flow—up to the target definition—produces an intermediate result (for example, the results of a SQL statement containing a WHERE clause), which flows to the next step in the data flow. The intermediate result consists of a set of rows from the previous operation and the schema in which the rows are arranged. This result is called a data set. This data set may, in turn, be further "filtered" and directed into yet another data set. 7.1.6 Operation codes Each row in a data set is flagged with an operation code that identifies the status of the row. The operation codes are as follows: Operation code Description Creates a new row in the target. NORMAL 153 All rows in a data set are flagged as NORMAL when they are extracted from a source. If a row is flagged as NORMAL when loaded into a target, it is inserted as a new row in the target. 2011-06-09
- 154. Data Flows Operation code Description Creates a new row in the target. INSERT Rows can be flagged as INSERT by transforms in the data flow to indicate that a change occurred in a data set as compared with an earlier image of the same data set. The change is recorded in the target separately from the existing data. Is ignored by the target. Rows flagged as DELETE are not loaded. DELETE Rows can be flagged as DELETE only by the Map_Operation transform. Overwrites an existing row in the target. UPDATE Rows can be flagged as UPDATE by transforms in the data flow to indicate that a change occurred in a data set as compared with an earlier image of the same data set. The change is recorded in the target in the same row as the existing data. 7.1.7 Passing parameters to data flows Data does not flow outside a data flow, not even when you add a data flow to a work flow. You can, however, pass parameters into and out of a data flow. Parameters evaluate single values rather than sets of values. When a data flow receives parameters, the steps inside the data flow can reference those parameters as variables. Parameters make data flow definitions more flexible. For example, a parameter can indicate the last time a fact table was updated. You can use this value in a data flow to extract only rows modified since the last update. The following figure shows the parameter last_update used in a query to determine the data set used to load the fact table. Related Topics • Variables and Parameters 154 2011-06-09
- 155. Data Flows 7.2 Creating and defining data flows You can create data flows using objects from • • the object library the tool palette After creating a data flow, you can change its properties. Related Topics • To change properties of a data flow 7.2.1 To define a new data flow using the object library 1. In the object library, go to the Data Flows tab. 2. Select the data flow category, right-click and select New. 3. Select the new data flow. 4. Drag the data flow into the workspace for a job or a work flow. 5. Add the sources, transforms, and targets you need. 7.2.2 To define a new data flow using the tool palette 1. Select the data flow icon in the tool palette. 2. Click the workspace for a job or work flow to place the data flow. You can add data flows to batch and real-time jobs. When you drag a data flow icon into a job, you are telling the software to validate these objects according the requirements of the job type (either batch or real-time). 3. Add the sources, transforms, and targets you need. 7.2.3 To change properties of a data flow 155 2011-06-09
- 156. Data Flows 1. Right-click the data flow and select Properties. The Properties window opens for the data flow. 2. Change desired properties of a data flow. 3. Click OK. This table describes the various properties you can set for the data flow. Option Description Execute only once When you specify that a data flow should only execute once, a batch job will never re-execute that data flow after the data flow completes successfully, except if the data flow is contained in a work flow that is a recovery unit that re-executes and has not completed successfully elsewhere outside the recovery unit. It is recommended that you do not mark a data flow as Execute only once if a parent work flow is a recovery unit. Use database links Database links are communication paths between one database server and another. Database links allow local users to access data on a remote database, which can be on the local or a remote computer of the same or different database type. Degree of parallelism Degree Of Parallelism (DOP) is a property of a data flow that defines how many times each transform within a data flow replicates to process a parallel subset of data. Cache type You can cache data to improve performance of operations such as joins, groups, sorts, filtering, lookups, and table comparisons. You can select one of the following values for the Cache type option on your data flow Properties window: • In-Memory: Choose this value if your data flow processes a small amount of data that can fit in the available memory. • Pageable: This value is the default. Related Topics • Performance Optimization Guide: Maximizing Push-Down Operations, Database link support for push-down operations across datastores • Performance Optimization Guide: Using parallel Execution, Degree of parallelism • Performance Optimization Guide: Using Caches • Reference Guide: Objects, Data flow 7.3 Source and target objects A data flow directly reads and loads data using two types of objects: 156 2011-06-09
- 157. Data Flows Source objects— Define sources from which you read data Target objects— Define targets to which you write (or load) data Related Topics • Source objects • Target objects 7.3.1 Source objects Source objects represent data sources read from data flows. Source object Description Software access Table A file formatted with columns and rows as used in relational databases Direct or through adapter Template table A template table that has been created and saved in another data flow (used in development). Direct File A delimited or fixed-width flat file Direct Document A file with an application- specific format (not readable by SQL or XML parser) Through adapter XML file A file formatted with XML tags Direct XML message Used as a source in real-time jobs. Direct You can also use IDoc messages as real-time sources for SAP applications. Related Topics • Template tables • Real-time source and target objects • Supplement for SAP: IDoc sources in real-time jobs 7.3.2 Target objects Target objects represent data targets that can be written to in data flows. 157 2011-06-09
- 158. Data Flows Target object Description Software access Table A file formatted with columns and rows as used in relational databases Direct or through adapter Template table A table whose format is based on the output of the preceding transform (used in development) Direct File A delimited or fixed-width flat file Direct Document A file with an application- specific format (not readable by SQL or XML parser) Through adapter XML file A file formatted with XML tags Direct XML template file An XML file whose format is based on the preceding transform output (used in development, primarily for debugging data flows) Direct XML message See Real-time source and target objects Outbound message See Real-time source and target objects You can also use IDoc messages as real-time sources for SAP applications. Related Topics • Supplement for SAP: IDoc targets in real-time jobs 7.3.3 Adding source or target objects to data flows Fulfill the following prerequisites before using a source or target object in a data flow: For Tables accessed directly from a database Define a database datastore and import table metadata. Template tables Define a database datastore. Files 158 Prerequisite Define a file format and import the file 2011-06-09
- 159. Data Flows For Prerequisite XML files and messages Import an XML file format Objects accessed through an adapter Define an adapter datastore and import object metadata. Related Topics • Database datastores • Template tables • File formats • To import a DTD or XML Schema format • Adapter datastores 7.3.3.1 To add a source or target object to a data flow 1. Open the data flow in which you want to place the object. 2. If the object library is not already open, select Tools > Object Library to open it. 3. Select the appropriate object library tab: Choose the Formats tab for flat files, DTDs, or XML Schemas, or choose the Datastores tab for database and adapter objects. 4. Select the object you want to add as a source or target. (Expand collapsed lists by clicking the plus sign next to a container icon.) For a new template table, select the Template Table icon from the tool palette. For a new XML template file, select the Template XML icon from the tool palette. 5. Drop the object in the workspace. 6. For objects that can be either sources or targets, when you release the cursor, a popup menu appears. Select the kind of object to make. For new template tables and XML template files, when you release the cursor, a secondary window appears. Enter the requested information for the new template object. Names can include alphanumeric characters and underscores (_). Template tables cannot have the same name as an existing table within a datastore. 7. The source or target object appears in the workspace. 8. Click the object name in the workspace The software opens the editor for the object. Set the options you require for the object. 159 2011-06-09
- 160. Data Flows Note: Ensure that any files that reference flat file, DTD, or XML Schema formats are accessible from the Job Server where the job will be run and specify the file location relative to this computer. 7.3.4 Template tables During the initial design of an application, you might find it convenient to use template tables to represent database tables. With template tables, you do not have to initially create a new table in your DBMS and import the metadata into the software. Instead, the software automatically creates the table in the database with the schema defined by the data flow when you execute a job. After creating a template table as a target in one data flow, you can use it as a source in other data flows. Though a template table can be used as a source table in multiple data flows, it can only be used as a target in one data flow. Template tables are particularly useful in early application development when you are designing and testing a project. If you modify and save the data transformation operation in the data flow where the template table is a target, the schema of the template table automatically changes. Any updates to the schema are automatically made to any other instances of the template table. During the validation process, the software warns you of any errors such as those resulting from changing the schema. 7.3.4.1 To create a target template table 1. Use one of the following methods to open the Create Template window: • From the tool palette: • • • • From the object library: • • • Click the template table icon. Click inside a data flow to place the template table in the workspace. On the Create Template window, select a datastore. Expand a datastore. Click the template table icon and drag it to the workspace. From the object library: • • Expand a datastore. Click the template table icon and drag it to the workspace. 2. On the Create Template window, enter a table name. 3. Click OK. The table appears in the workspace as a template table icon. 160 2011-06-09
- 161. Data Flows 4. Connect the template table to the data flow as a target (usually a Query transform). 5. In the Query transform, map the Schema In columns that you want to include in the target table. 6. From the Project menu select Save. In the workspace, the template table's icon changes to a target table icon and the table appears in the object library under the datastore's list of tables. After you are satisfied with the design of your data flow, save it. When the job is executed, software uses the template table to create a new table in the database you specified when you created the template table. Once a template table is created in the database, you can convert the template table in the repository to a regular table. 7.3.5 Converting template tables to regular tables You must convert template tables to regular tables to take advantage of some features such as bulk loading. Other features, such as exporting an object, are available for template tables. Note: Once a template table is converted, you can no longer alter the schema. 7.3.5.1 To convert a template table into a regular table from the object library 1. Open the object library and go to the Datastores tab. 2. Click the plus sign (+) next to the datastore that contains the template table you want to convert. A list of objects appears. 3. Click the plus sign (+) next to Template Tables. The list of template tables appears. 4. Right-click a template table you want to convert and select Import Table. The software converts the template table in the repository into a regular table by importing it from the database. To update the icon in all data flows, choose View > Refresh. In the datastore object library, the table is now listed under Tables rather than Template Tables. 7.3.5.2 To convert a template table into a regular table from a data flow 161 2011-06-09
- 162. Data Flows 1. Open the data flow containing the template table. 2. Right-click on the template table you want to convert and select Import Table. After a template table is converted into a regular table, you can no longer change the table's schema. 7.4 Adding columns within a data flow Within a data flow, the Propagate Column From command adds an existing column from an upstream source or transform through intermediate objects to the selected endpoint. Columns are added in each object with no change to the data type or other attributes. When there is more than one possible path between the starting point and ending point, you can specify the route for the added columns. Column propagation is a pull-through operation. The Propagate Column From command is issued from the object where the column is needed. The column is pulled from the selected upstream source or transform and added to each of the intermediate objects as well as the selected endpoint object. For example, in the data flow below, the Employee source table contains employee name information as well as employee ID, job information, and hire dates. The Name_Cleanse transform is used to standardize the employee names. Lastly, the data is output to an XML file called Employee_Names. After viewing the output in the Employee_Names table, you realize that the middle initial (minit column) should be included in the output. You right-click the top-level schema of the Employee_Names table and select Propagate Column From. The "Propagate Column to Employee_Names" window appears. In the left pane of the "Propagate Column to Employee_Names" window, select the Employee source table from the list of objects. The list of output columns displayed in the right pane changes to display the columns in the schema of the selected object. Select the MINIT column as the column you want to pull through from the source, and then click Propagate. The minit column schema is carried through the Query and Name_Cleanse transforms to the Em ployee_Names table. Characteristics of propagated columns are as follows: • The Propagate Column From command can be issued from the top-level schema of either a transform or a target. • Columns are added in each object with no change to the data type or other attributes. Once a column is added to the schema of an object, the column functions in exactly the same way as if it had been created manually. • The propagated column is added at the end of the schema list in each object. 162 2011-06-09
- 163. Data Flows • • • • The output column name is auto-generated to avoid naming conflicts with existing columns. You can edit the column name, if desired. Only columns included in top-level schemas can be propagated. Columns in nested schemas cannot be propagated. A column can be propagated more than once. Any existing columns are shown in the right pane of the "Propagate Column to" window in the "Already Exists In" field. Each additional column will have a unique name. Multiple columns can be selected and propagated in the same operation. Note: You cannot propagate a column through a Hierarchy_Flattening transform or a Table_Comparison transform. 7.4.1 To add columns within a data flow Within a data flow, the Propagate Column From command adds an existing column from an upstream source or transform through intermediate objects to a selected endpoint. Columns are added in each object with no change to the data type or other attributes. To add columns within a data flow: 1. In the downstream object where you want to add the column (the endpoint), right-click the top-level schema and click Propagate Column From. The Propagate Column From can be issued from the top-level schema in a transform or target object. 2. In the left pane of the "Propagate Column to" window, select the upstream object that contains the column you want to map. The available columns in that object are displayed in the right pane along with a list of any existing mappings from that column. 3. In the right pane, select the column you wish to add and click either Propagate or Propagate and Close. One of the following occurs: • If there is a single possible route, the selected column is added through the intermediate transforms to the downstream object. • If there is more than one possible path through intermediate objects, the "Choose Route to" dialog displays. This may occur when your data flow contains a Query transform with multiple input objects. Select the path you prefer and click OK. 7.4.2 Propagating columns in a data flow containing a Merge transform 163 2011-06-09
- 164. Data Flows In valid data flows that contain two or more sources which are merged using a Merge transform, the schema of the inputs into the Merge transform must be identical. All sources must have the same schema, including: • the same number of columns • the same column names • like columns must have the same data type In order to maintain a valid data flow when propagating a column through a Merge transform, you must make sure to meet this restriction. When you propagate a column and a Merge transform falls between the starting point and ending point, a message warns you that after the propagate operation completes the data flow will be invalid because the input schemas in the Merge transform will not be identical. If you choose to continue with the column propagation operation, you must later add columns to the input schemas in the Merge transform so that the data flow is valid. For example, in the data flow shown below, the data from each source table is filtered and then the results are merged in the Merge transform. If you choose to propagate a column from the SALES(Pubs.DBO) source to the CountrySales target, the column would be added to the TableFilter schema but not to the FileFilter schema, resulting in differing input schemas in the Merge transform and an invalid data flow. In order to maintain a valid data flow, when propagating a column through a Merge transform you may want to follow a multi-step process: 1. Ensure that the column you want to propagate is available in the schemas of all the objects that lead into the Merge transform on the upstream side. This ensures that all inputs to the Merge transform are identical and the data flow is valid. 2. Propagate the column on the downstream side of the Merge transform to the desired endpoint. 7.5 Lookup tables and the lookup_ext function Lookup tables contain data that other tables reference. Typically, lookup tables can have the following kinds of columns: 164 2011-06-09
- 165. Data Flows • • • Lookup column—Use to match a row(s) based on the input values. You apply operators such as =, >, <, ~ to identify a match in a row. A lookup table can contain more than one lookup column. Output column—The column returned from the row that matches the lookup condition defined for the lookup column. A lookup table can contain more than one output column. Return policy column—Use to specify the data to return in the case where multiple rows match the lookup condition(s). Use the lookup_ext function to retrieve data from a lookup table based on user-defined lookup conditions that match input data to the lookup table data. Not only can the lookup_ext function retrieve a value in a table or file based on the values in a different source table or file, but it also provides extended functionality that lets you do the following: • Return multiple columns from a single lookup • Choose from more operators, including pattern matching, to specify a lookup condition • Specify a return policy for your lookup • Call lookup_ext in scripts and custom functions (which also lets you reuse the lookup(s) packaged inside scripts) • Define custom SQL using the SQL_override parameter to populate the lookup cache, which is useful for narrowing large quantities of data to only the sections relevant for your lookup(s) • Call lookup_ext using the function wizard in the query output mapping to return multiple columns in a Query transform • Choose a caching strategy, for example decide to cache the whole lookup table in memory or dynamically generate SQL for each input record • Use lookup_ext with memory datastore tables or persistent cache tables. The benefits of using persistent cache over memory tables for lookup tables are: • Multiple data flows can use the same lookup table that exists on persistent cache. • The software does not need to construct the lookup table each time a data flow uses it. • Persistent cache has no memory constraints because it is stored on disk and the software quickly pages it into memory. • • Use pageable cache (which is not available for the lookup and lookup_seq functions) Use expressions in lookup tables and return the resulting values For a description of the related functions lookup and lookup_seq, see the Reference Guide. Related Topics • Reference Guide: Functions and Procedures, lookup_ext • Performance Optimization Guide: Using Caches, Caching data 7.5.1 Accessing the lookup_ext editor Lookup_ext has its own graphic editor. You can invoke the editor in two ways: • 165 Add a new function call inside a Query transform—Use this option if you want the lookup table to return more than one column 2011-06-09
- 166. Data Flows • From the Mapping tab in a query or script function 7.5.1.1 To add a new function call 1. In the Query transform "Schema out" pane, without selecting a specific output column right-click in the pane and select New Function Call. 2. Select the "Function category" Lookup Functions and the "Function name"  lookup_ext. 3. Click Next to invoke the editor. In the Output section, you can add multiple columns to the output schema. An advantage of using the new function call is that after you close the lookup_ext function window, you can reopen the graphical editor to make modifications (right-click the function name in the schema and select Modify Function Call). 7.5.1.2 To invoke the lookup_ext editor from the Mapping tab 1. Select the output column name. 2. On the "Mapping" tab, click Functions. 3. Select the "Function category"Lookup Functions and the "Function name"lookup_ext. 4. Click Next to invoke the editor. In the Output section, "Variable" replaces "Output column name". You can define one output column that will populate the selected column in the output schema. When lookup_ext returns more than one output column, use variables to store the output values, or use lookup_ext as a new function call as previously described in this section. With functions used in mappings, the graphical editor isn't available, but you can edit the text on the "Mapping" tab manually. 7.5.2 Example: Defining a simple lookup_ext function This procedure describes the process for defining a simple lookup_ext function using a new function call. The associated example illustrates how to use a lookup table to retrieve department names for employees. For details on all the available options for the lookup_ext function, see the Reference Guide. 1. In a data flow, open the Query editor. 166 2011-06-09
- 167. Data Flows 2. From the "Schema in" pane, drag the ID column to the "Schema out" pane. 3. Select the ID column in the "Schema out" pane, right-click, and click New Function Call. Click Insert Below. 4. Select the "Function category"Lookup Functions and the "Function name"lookup_ext and click Next. The lookup_ext editor opens. 5. In the "Lookup_ext - Select Parameters" window, select a lookup table: a. Next to the Lookup table text box, click the drop-down arrow and double-click the datastore, file format, or current schema that includes the table. b. Select the lookup table and click OK. In the example, the lookup table is a file format called ID_lookup.txt that is in D:Data. 6. For the Cache spec, the default of PRE_LOAD_CACHE is useful when the number of rows in the table is small or you expect to access a high percentage of the table values. NO_CACHE reads values from the lookup table for every row without caching values. Select DEMAND_LOAD_CACHE when the number of rows in the table is large and you expect to frequently access a low percentage of table values or when you use the table in multiple lookups and the compare conditions are highly selective, resulting in a small subset of data. 7. To provide more resources to execute the lookup_ext function, select Run as a separate process. This option creates a separate child data flow process for the lookup_ext function when the software executes the data flow. 8. Define one or more conditions. For each, add a lookup table column name (select from the drop-down list or drag from the "Parameter" pane), select the appropriate operator, and enter an expression by typing, dragging, pasting, or using the Smart Editor (click the icon in the right column). In the example, the condition is ID_DEPT = Employees.ID_DEPT. 9. Define the output. For each output column: a. Add a lookup table column name. b. Optionally change the default value from NULL. c. Specify the "Output column name" by typing, dragging, pasting, or using the Smart Editor (click the icon in the right column). In the example, the output column is ID_DEPT_NAME. 10. If multiple matches are possible, specify the ordering and set a return policy (default is MAX) to select one match. To order the output, enter the column name(s) in the "Order by" list. Example: The following example illustrates how to use the lookup table ID_lookup.txt to retrieve department names for employees. The Employees table is as follows: 167 2011-06-09
- 168. Data Flows ID NAME ID_DEPT SSN111111111 Employee1 10 SSN222222222 Employee2 10 TAXID333333333 Employee3 20 The lookup table ID_lookup.txt is as follows: ID_DEPT ID_PATTERN ID_RETURN ID_DEPT_NAME 10 ms(SSN*) =substr(ID_Pattern,4,20) Payroll 20 ms(TAXID*) =substr(ID_Pattern,6,30) Accounting The lookup_ext editor would be configured as follows. Related Topics • Example: Defining a complex lookup_ext function 168 2011-06-09
- 169. Data Flows 7.5.3 Example: Defining a complex lookup_ext function This procedure describes the process for defining a complex lookup_ext function using a new function call. The associated example uses the same lookup and input tables as in the Example: Defining a simple lookup_ext function This example illustrates how to extract and normalize employee ID numbers. For details on all the available options for the lookup_ext function, see the Reference Guide. 1. In a data flow, open the Query editor. 2. From the "Schema in" pane, drag the ID column to the "Schema out" pane. Do the same for the Name column. 3. In the "Schema out" pane, right-click the Name column and click New Function Call. Click Insert Below. 4. Select the "Function category"Lookup Functions and the "Function name"lookup_ext and click Next. 5. In the "Lookup_ext - Select Parameters" window, select a lookup table: In the example, the lookup table is in the file format ID_lookup.txt that is in D:Data. 6. Define one or more conditions. In the example, the condition is ID_PATTERN ~ Employees.ID. 7. Define the output. For each output column: a. Add a lookup table column name. b. If you want the software to interpret the column in the lookup table as an expression and return the calculated value, select the Expression check box. c. Optionally change the default value from NULL. d. Specify the "Output column name"(s) by typing, dragging, pasting, or using the Smart Editor (click the icon in the right column). In the example, the output columns are ID_RETURN and ID_DEPT_NAME. Example: In this example, you want to extract and normalize employee Social Security numbers and tax identification numbers that have different prefixes. You want to remove the prefixes, thereby normalizing the numbers. You also want to identify the department from where the number came. The data flow has one source table Employees, a query configured with lookup_ext, and a target table. Configure the lookup_ext editor as in the following graphic. 169 2011-06-09
- 170. Data Flows The lookup condition is ID_PATTERN ~ Employees.ID. The software reads each row of the source table Employees, then checks the lookup table ID_lookup.txt for all rows that satisfy the lookup condition. The operator ~ means that the software will apply a pattern comparison to Employees.ID. When it encounters a pattern in ID_lookup.ID_PATTERN that matches Employees.ID, the software applies the expression in ID_lookup.ID_RETURN. In this example, Employee1 and Employee2 both have IDs that match the pattern ms(SSN*) in the lookup table. the software then applies the expression =sub str(ID_PATTERN,4,20) to the data, which extracts from the matched string (Employees.ID) a substring of up to 20 characters starting from the 4th position. The results for Employee1 and Employee2 are 111111111 and 222222222, respectively. For the output of the ID_RETURN lookup column, the software evaluates ID_RETURN as an expression because the Expression box is checked. In the lookup table, the column ID_RETURN contains the expression =substr(ID_PATTERN,4,20). ID_PATTERN in this expression refers to the lookup table column ID_PATTERN. When the lookup condition ID_PATTERN ~ Employees.ID is true, the software evaluates the expression. Here the software substitutes the placeholder ID_PATTERN with the actual Employees.ID value. The output also includes the ID_DEPT_NAME column, which the software returns as a literal value (because the Expression box is not checked). The resulting target table is as follows: 170 2011-06-09
- 171. Data Flows ID NAME ID_RETURN ID_DEPT_NAME SSN111111111 Employee1 111111111 Payroll SSN222222222 Employee2 222222222 Payroll TAXID333333333 Employee3 333333333 Accounting Related Topics • Reference Guide: Functions and Procedures, lookup_ext • Accessing the lookup_ext editor • Example: Defining a simple lookup_ext function • Reference Guide: Functions and Procedures, match_simple 7.6 Data flow execution A data flow is a declarative specification from which the software determines the correct data to process. For example in data flows placed in batch jobs, the transaction order is to extract, transform, then load data into a target. Data flows are similar to SQL statements. The specification declares the desired output. The software executes a data flow each time the data flow occurs in a job. However, you can specify that a batch job execute a particular data flow only one time. In that case, the software only executes the first occurrence of the data flow; the software skips subsequent occurrences in the job. You might use this feature when developing complex batch jobs with multiple paths, such as jobs with try/catch blocks or conditionals, and you want to ensure that the software only executes a particular data flow one time. Related Topics • Creating and defining data flows 7.6.1 Push down operations to the database server From the information in the data flow specification, the software produces output while optimizing performance. For example, for SQL sources and targets, the software creates database-specific SQL statements based on a job's data flow diagrams. To optimize performance, the software pushes down as many transform operations as possible to the source or target database and combines as many 171 2011-06-09
- 172. Data Flows operations as possible into one request to the database. For example, the software tries to push down joins and function evaluations. By pushing down operations to the database, the software reduces the number of rows and operations that the engine must process. Data flow design influences the number of operations that the software can push to the source or target database. Before running a job, you can examine the SQL that the software generates and alter your design to produce the most efficient results. You can use the Data_Transfer transform to pushdown resource-intensive operations anywhere within a data flow to the database. Resource-intensive operations include joins, GROUP BY, ORDER BY, and DISTINCT. Related Topics • Performance Optimization Guide: Maximizing push-down operations • Reference Guide: Data_Transfer 7.6.2 Distributed data flow execution The software provides capabilities to distribute CPU-intensive and memory-intensive data processing work (such as join, grouping, table comparison and lookups) across multiple processes and computers. This work distribution provides the following potential benefits: • Better memory management by taking advantage of more CPU resources and physical memory • Better job performance and scalability by using concurrent sub data flow execution to take advantage of grid computing You can create sub data flows so that the software does not need to process the entire data flow in memory at one time. You can also distribute the sub data flows to different job servers within a server group to use additional memory and CPU resources. Use the following features to split a data flow into multiple sub data flows: • Run as a separate process option on resource-intensive operations that include the following: • • • • • • • • • 172 Hierarchy_Flattening transform Associate transform Country ID transform Global Address Cleanse transform Global Suggestion Lists transform Match Transform United States Regulatory Address Cleanse transform User-Defined transform Query operations that are CPU-intensive and memory-intensive: • Join • GROUP BY 2011-06-09
- 173. Data Flows • • • • • • ORDER BY DISTINCT Table_Comparison transform Lookup_ext function Count_distinct function Search_replace function If you select the Run as a separate process option for multiple operations in a data flow, the software splits the data flow into smaller sub data flows that use separate resources (memory and computer) from each other. When you specify multiple Run as a separate process options, the sub data flow processes run in parallel. • Data_Transfer transform With this transform, the software does not need to process the entire data flow on the Job Server computer. Instead, the Data_Transfer transform can push down the processing of a resource-intensive operation to the database server. This transform splits the data flow into two sub data flows and transfers the data to a table in the database server to enable the software to push down the operation. Related Topics • Performance Optimization Guide: Splitting a data flow into sub data flows • Performance Optimization Guide: Data_Transfer transform for push-down operations 7.6.3 Load balancing You can distribute the execution of a job or a part of a job across multiple Job Servers within a Server Group to better balance resource-intensive operations. You can specify the following values on the Distribution level option when you execute a job: • Job level - A job can execute on an available Job Server. • Data flow level - Each data flow within a job can execute on an available Job Server. • Sub data flow level - An resource-intensive operation (such as a sort, table comparison, or table lookup) within a data flow can execute on an available Job Server. Related Topics • Performance Optimization Guide: Using grid computing to distribute data flows execution 7.6.4 Caches 173 2011-06-09
- 174. Data Flows The software provides the option to cache data in memory to improve operations such as the following in your data flows. • Joins — Because an inner source of a join must be read for each row of an outer source, you might want to cache a source when it is used as an inner source in a join. • Table comparisons — Because a comparison table must be read for each row of a source, you might want to cache the comparison table. • Lookups — Because a lookup table might exist on a remote database, you might want to cache it in memory to reduce access times. The software provides the following types of caches that your data flow can use for all of the operations it contains: • In-memory Use in-memory cache when your data flow processes a small amount of data that fits in memory. • Pageable cache Use a pageable cache when your data flow processes a very large amount of data that does not fit in memory. If you split your data flow into sub data flows that each run on a different Job Server, each sub data flow can use its own cache type. Related Topics • Performance Optimization Guide: Using Caches 7.7 Audit Data Flow overview You can audit objects within a data flow to collect run time audit statistics. You can perform the following tasks with this auditing feature: • Collect audit statistics about data read into a job, processed by various transforms, and loaded into targets. • Define rules about the audit statistics to determine if the correct data is processed. • Generate notification of audit failures. • Query the audit statistics that persist in the repository. For a full description of auditing data flows, see Using Auditing . 174 2011-06-09
- 175. Transforms Transforms Transforms operate on data sets by manipulating input sets and producing one or more output sets. By contrast, functions operate on single values in specific columns in a data set. Many built-in transforms are available from the object library on the Transforms tab. The following is a list of available transforms. The transforms that you can use depend on the software package that you have purchased. (If a transform belongs to a package that you have not purchased, it is disabled and cannot be used in a job.) Transform Category Transform Description Data Integrator Data_Transfer Allows a data flow to split its processing into two sub data flows and push down resource-consuming operations to the database server. Date_Generation Generates a column filled with date values based on the start and end dates and increment that you provide. Effective_Date Generates an additional "effective to" column based on the primary key's "effective date." Hierarchy_Flattening Flattens hierarchical data into relational tables so that it can participate in a star schema. Hierarchy flattening can be both vertical and horizontal. History_Preserving Converts rows flagged as UPDATE to UPDATE plus INSERT, so that the original values are preserved in the target. You specify in which column to look for updated data. Key_Generation Generates new keys for source data, starting from a value based on existing keys in the table you specify. Map_CDC_Operation Sorts input data, maps output data, and resolves before- and after-images for UPDATE rows. While commonly used to support Oracle changed-data capture, this transform supports any data stream if its input requirements are met. 175 2011-06-09
- 176. Transforms Transform Category Rotates the values in specified columns to rows. (Also see Reverse Pivot.) Reverse Pivot (Rows to Columns) Rotates the values in specified rows to columns. Table_Comparison Compares two data sets and produces the difference between them as a data set with rows flagged as INSERT and UPDATE. XML_Pipeline Processes large XML inputs in small batches. Associate Combine the results of two or more Match transforms or two or more Associate transforms, or any combination of the two, to find matches across match sets. Country ID Parses input data and then identifies the country of destination for each record. Data Cleanse Identifies and parses name, title, and firm data, phone numbers, Social Security numbers, dates, and e-mail addresses. It can assign gender, add prenames, generate Match standards, and convert input sources to a standard format. It can also parse and manipulate various forms of international data, as well as operational and product data. DSF2 Walk Sequencer Adds delivery sequence information to your data, which you can use with presorting software to qualify for walk-sequence discounts. Geocoder Uses geographic coordinates, addresses, and pointof-interest (POI) data to append address, latitude and longitude, census, and other information to your records. Global Address Cleanse Identifies, parses, validates, and corrects global address data, such as primary number, primary name, primary type, directional, secondary identifier, and secondary number. Global Suggestion Lists Completes and populates addresses with minimal data, and it can offer suggestions for possible matches. Match 176 Description Pivot (Columns to Rows) Data Quality Transform Identifies matching records based on your business rules. Also performs candidate selection, unique ID, best record, and other operations. 2011-06-09
- 177. Transforms Transform Category Identifies, parses, validates, and corrects USA address data according to the U.S. Coding Accuracy Support System (CASS). User-Defined Does just about anything that you can write Python code to do. You can use the User-Defined transform to create new records and data sets, or populate a field with a specific value, just to name a few possibilities. Case Simplifies branch logic in data flows by consolidating case or decision making logic in one transform. Paths are defined in an expression table. Map_Operation Allows conversions between operation codes. Merge Unifies rows from two or more sources into a single target. Query Retrieves a data set that satisfies conditions that you specify. A query transform is similar to a SQL SELECT statement. Row_Generation Generates a column filled with integer values starting at zero and incrementing by one to the end value you specify. SQL Performs the indicated SQL query operation. Validation Text Data Processing Description USA Regulatory Address Cleanse Platform Transform Ensures that the data at any stage in the data flow meets your criteria. You can filter out or replace data that fails your criteria. Entity_Extraction Extracts information (entities and facts) from any text, HTML, or XML content. Related Topics • Reference Guide: Transforms 8.1 To add a transform to a data flow You can use the Designer to add transforms to data flows. 1. Open a data flow object. 177 2011-06-09
- 178. Transforms 2. Open the object library if it is not already open and click the Transforms tab. 3. Select the transform or transform configuration that you want to add to the data flow. 4. Drag the transform or transform configuration icon into the data flow workspace. If you selected a transform that has available transform configurations, a drop-down menu prompts you to select a transform configuration. 5. Draw the data flow connections. To connect a source to a transform, click the square on the right edge of the source and drag the cursor to the arrow on the left edge of the transform. Continue connecting inputs and outputs as required for the transform. • • The input for the transform might be the output from another transform or the output from a source; or, the transform may not require source data. You can connect the output of the transform to the input of another transform or target. 6. Double-click the name of the transform. This opens the transform editor, which lets you complete the definition of the transform. 7. Enter option values. To specify a data column as a transform option, enter the column name as it appears in the input schema or drag the column name from the input schema into the option box. Related Topics • To add a Query transform to a data flow • To add a Data Quality transform to a data flow • To add a text data processing transform to a data flow 8.2 Transform editors After adding a transform to a data flow, you configure it using the transform's editor. Transform editor layouts vary. The most commonly used transform is the Query transform, which has two panes: • • An input schema area and/or output schema area A options area (or parameters area) that lets you to set all the values the transform requires Data Quality transforms, such as Match and Data Cleanse, use a transform editor that lets you set options and map input and output fields. The Entity Extraction transform editor lets you set extraction options and map input and output fields. 178 2011-06-09
- 179. Transforms Related Topics • Query Editor • Data Quality transform editors • Entity Extraction transform editor 8.3 Transform configurations A transform configuration is a transform with preconfigured best practice input fields, best practice output fields, and options that can be used in multiple data flows. These are useful if you repeatedly use a transform with specific options and input and output fields. Some transforms, such as Data Quality transforms, have read-only transform configurations that are provided when Data Services is installed. You can also create your own transform configuration, either by replicating an existing transform configuration or creating a new one. You cannot perform export or multi-user operations on read-only transform configurations. In the Transform Configuration Editor window, you set up the default options, best practice input fields, and best practice output fields for your transform configuration. After you place an instance of the transform configuration in a data flow, you can override these preset defaults. If you edit a transform configuration, that change is inherited by every instance of the transform configuration used in data flows, unless a user has explicitly overridden the same option value in an instance. Related Topics • To create a transform configuration • To add a user-defined field 8.3.1 To create a transform configuration 1. In the Transforms tab of the "Local Object Library," right-click a transform and select New to create a new transform configuration, or right-click an existing transform configuration and select Replicate. If New or Replicate is not available from the menu, then the selected transform type cannot have transform configurations. The "Transform Configuration Editor" window opens. 2. In Transform Configuration Name, enter the name of the transform configuration. 3. In the Options tab, set the option values to determine how the transform will process your data. The available options depend on the type of transform that you are creating a configuration for. 179 2011-06-09
- 180. Transforms For the Associate, Match, and User-Defined transforms, options are not editable in the Options tab. You must set the options in the Associate Editor, Match Editor, or User-Defined Editor, which are accessed by clicking the Edit Options button. If you change an option value from its default value, a green triangle appears next to the option name to indicate that you made an override. 4. To designate an option as "best practice," select the Best Practice checkbox next to the option's value. Designating an option as best practice indicates to other users who use the transform configuration which options are typically set for this type of transform. Use the filter to display all options or just those options that are designated as best practice options. 5. Click the Verify button to check whether the selected option values are valid. If there are any errors, they are displayed at the bottom of the window. 6. In the Input Best Practices tab, select the input fields that you want to designate as the best practice input fields for the transform configuration. The transform configurations provided with Data Services do not specify best practice input fields, so that it doesn't appear that one input schema is preferred over other input schemas. For example, you may map the fields in your data flow that contain address data whether the address data resides in discrete fields, multiline fields, or a combination of discrete and multiline fields. These input fields will be the only fields displayed when the Best Practice filter is selected in the Input tab of the transform editor when the transform configuration is used within a data flow. 7. For Associate, Match, and User-Defined transform configurations, you can create user-defined input fields. Click the Create button and enter the name of the input field. 8. In the Output Best Practices tab, select the output fields that you want to designate as the best practice output fields for the transform configuration. These output fields will be the only fields displayed when the Best Practice filter is selected in the Output tab of the transform editor when the transform configuration is used within a data flow. 9. Click OK to save the transform configuration. The transform configuration is displayed in the "Local Object Library" under the base transform of the same type. You can now use the transform configuration in data flows. Related Topics • Reference Guide: Transforms, Transform configurations 8.3.2 To add a user-defined field For some transforms, such as the Associate, Match, and User-Defined transforms, you can create user-defined input fields rather than fields that are recognized by the transform. These transforms use user-defined fields because they do not have a predefined set of input fields. 180 2011-06-09
- 181. Transforms You can add a user-defined field either to a single instance of a transform in a data flow or to a transform configuration so that it can be used in all instances. In the User-Defined transform, you can also add user-defined output fields. 1. In the Transforms tab of the "Local Object Library," right-click an existing Associate, Match, or UserDefined transform configuration and select Edit. The "Transform Configuration Editor" window opens. 2. In the Input Best Practices tab, click the Create button and enter the name of the input field. 3. Click OK to save the transform configuration. When you create a user-defined field in the transform configuration, it is displayed as an available field in each instance of the transform used in a data flow. You can also create user-defined fields within each transform instance. Related Topics • Data Quality transform editors 8.4 The Query transform The Query transform is by far the most commonly used transform, so this section provides an overview. The Query transform can perform the following operations: • • • • • • • Choose (filter) the data to extract from sources Join data from multiple sources Map columns from input to output schemas Perform transformations and functions on the data Perform data nesting and unnesting Add new columns, nested schemas, and function results to the output schema Assign primary keys to output columns Related Topics • Nested Data • Reference Guide: Transforms 8.4.1 To add a Query transform to a data flow 181 2011-06-09
- 182. Transforms Because it is so commonly used, the Query transform icon is included in the tool palette, providing an easier way to add a Query transform. 1. Click the Query icon in the tool palette. 2. Click anywhere in a data flow workspace. 3. Connect the Query to inputs and outputs. Note: • • • • The inputs for a Query can include the output from another transform or the output from a source. The outputs from a Query can include input to another transform or input to a target. You can change the content type for the columns in your data by selecting a different type from the output content type list. If you connect a target table to a Query with an empty output schema, the software automatically fills the Query's output schema with the columns from the target table, without mappings. 8.4.2 Query Editor The Query Editor is a graphical interface for performing query operations. It contains the following areas: input schema area (upper left), output schema area (upper right), and a parameters area (lower tabbed area). The icon indicates that the tab contains user-defined entries or that there is at least one join pair (FROM tab only). The input and output schema areas can contain: Columns, Nested schemas, and Functions (output only). The "Schema In" and "Schema Out" lists display the currently selected schema in each area. The currently selected output schema is called the current schema and determines the following items: • • The output elements that can be modified (added, mapped, or deleted) The scope of the Select through Order by tabs in the parameters area The current schema is highlighted while all other (non-current) output schemas are gray. 8.4.2.1 To change the current output schema You can change the current output schema in the following ways: • • • 182 Select a schema from the Output list so that it is highlighted. Right-click a schema, column, or function in the Output Schema area and select Make Current. Double-click one of the non-current (grayed-out) elements in the Output Schema area. 2011-06-09
- 183. Transforms 8.4.2.2 To modify the output schema contents You can modify the output schema in several ways: • • • • Drag and drop (or copy and paste) columns or nested schemas from the input schema area to the output schema area to create simple mappings. Use right-click menu options on output elements to: • Add new output columns and schemas. • Use function calls to generate new output columns. • Assign or reverse primary key settings on output columns. Primary key columns are flagged by a key icon. • Unnest or re-nest schemas. Use the Mapping tab to provide complex column mappings. Drag and drop input schemas and columns into the output schema to enable the editor. Use the function wizard and the smart editor to build expressions. When the text editor is enabled, you can access these features using the buttons above the editor. Use the Select through Order By tabs to provide additional parameters for the current schema (similar to SQL SELECT statement clauses). You can drag and drop schemas and columns into these areas. Tab name Description Select Specifies whether to output only distinct rows (discarding any identical duplicate rows). From Lists all input schemas. Allows you to specify join pairs and join conditions as well as enter join rank and cache for each input schema. The resulting SQL FROM clause is displayed. Specifies conditions that determine which rows are output. Enter the conditions in SQL syntax, like a WHERE clause in a SQL SELECT statement. For example: Where TABLE1.EMPNO = TABLE2.EMPNO AND TABLE1.EMPNO > 1000 OR TABLE2.EMPNO < 9000 Use the Functions, Domains, and smart editor buttons for help building expressions. Group By 183 Specifies how the output rows are grouped (if required). 2011-06-09
- 184. Transforms Tab name Order By • Description Specifies how the output rows are sequenced (if required). Use the Find tab to locate input and output elements containing a specific word or term. 8.5 Data Quality transforms Data Quality transforms are a set of transforms that help you improve the quality of your data. The transforms can parse, standardize, correct, and append information to your customer and operational data. Data Quality transforms include the following transforms: • • • • • • • • • Associate Country ID Data Cleanse DSF2 Walk Sequencer Global Address Cleanse Global Suggestion Lists Match USA Regulatory Address Cleanse User-Defined Related Topics • Reference Guide: Transforms 8.5.1 To add a Data Quality transform to a data flow Data Quality transforms cannot be directly connected to an upstream transform that contains or generates nested tables. This is common in real-time data flows, especially those that perform matching. To connect these transforms, you must insert either a Query transform or an XML Pipeline transform between the transform with the nested table and the Data Quality transform. 1. Open a data flow object. 2. Open the object library if it is not already open. 3. Go to the Transforms tab. 184 2011-06-09
- 185. Transforms 4. Expand the Data Quality transform folder and select the transform or transform configuration that you want to add to the data flow. 5. Drag the transform or transform configuration icon into the data flow workspace. If you selected a transform that has available transform configurations, a drop-down menu prompts you to select a transform configuration. 6. Draw the data flow connections. To connect a source or a transform to another transform, click the square on the right edge of the source or upstream transform and drag the cursor to the arrow on the left edge of the Data Quality transform. • The input for the transform might be the output from another transform or the output from a source; or, the transform may not require source data. • You can connect the output of the transform to the input of another transform or target. 7. Double-click the name of the transform. This opens the transform editor, which lets you complete the definition of the transform. 8. In the input schema, select the input fields that you want to map and drag them to the appropriate field in the Input tab. This maps the input field to a field name that is recognized by the transform so that the transform knows how to process it correctly. For example, an input field that is named "Organization" would be mapped to the Firm field. When content types are defined for the input, these columns are automatically mapped to the appropriate input fields. You can change the content type for the columns in your data by selecting a different type from the output content type list. 9. For the Associate, Match, and User-Defined transforms, you can add user-defined fields to the Input tab. You can do this in two ways: • Click the first empty row at the bottom of the table and press F2 on your keyboard. Enter the name of the field. Select the appropriate input field from the drop-down box to map the field. • Drag the appropriate input field to the first empty row at the bottom of the table. To rename the user-defined field, click the name, press F2 on your keyboard, and enter the new name. 10. In the Options tab, select the appropriate option values to determine how the transform will process your data. • Make sure that you map input fields before you set option values, because in some transforms, the available options and option values depend on the mapped input fields. • For the Associate, Match, and User-Defined transforms, options are not editable in the Options tab. You must set the options in the Associate Editor, Match Editor, and User-Defined Editor. You can access these editors either by clicking the Edit Options button in the Options tab or by right-clicking the transform in the data flow. If you change an option value from its default value, a green triangle appears next to the option name to indicate that you made an override. 11. In the Output tab, double-click the fields that you want to output from the transform. Data Quality transforms can generate fields in addition to the input fields that the transform processes, so you can output many fields. 185 2011-06-09
- 186. Transforms Make sure that you set options before you map output fields. The selected fields appear in the output schema. The output schema of this transform becomes the input schema of the next transform in the data flow. 12. If you want to pass data through the transform without processing it, drag fields directly from the input schema to the output schema. 13. To rename or resize an output field, double-click the output field and edit the properties in the "Column Properties" window. Related Topics • Reference Guide: Data Quality Fields • Data Quality transform editors 8.5.2 Data Quality transform editors The Data Quality editors, graphical interfaces for setting input and output fields and options, contain the following areas: input schema area (upper left), output schema area (upper right), and the parameters area (lower tabbed area). The parameters area contains three tabs: Input, Options, and Output. Generally, it is considered best practice to complete the tabs in this order, because the parameters available in a tab may depend on parameters selected in the previous tab. Input schema area The input schema area displays the input fields that are output from the upstream transform in the data flow. Output schema area The output schema area displays the fields that the transform outputs, and which become the input fields for the downstream transform in the data flow. Input tab The Input tab displays the available field names that are recognized by the transform. You map these fields to input fields in the input schema area. Mapping input fields to field names that the transform recognizes tells the transform how to process that field. Options tab The Options tab contain business rules that determine how the transform processes your data. Each transform has a different set of available options. If you change an option value from its default value, a green triangle appears next to the option name to indicate that you made an override. In the Associate, Match, and User-Defined transforms, you cannot edit the options directly in the Options tab. Instead you must use the Associate, Match, and User-Defined editors, which you can access from the Edit Options button. 186 2011-06-09
- 187. Transforms Output tab The Output tab displays the field names that can be output by the transform. Data Quality transforms can generate fields in addition to the input fields that that transform processes, so that you can output many fields. These mapped output fields are displayed in the output schema area. Filter and sort The Input, Options, and Output tabs each contain filters that determine which fields are displayed in the tabs. Filter Description Best Practice Displays the fields or options that have been designated as a best practice for this type of transform. However, these are merely suggestions; they may not meet your needs for processing or outputting your data. The transform configurations provided with the software do not specify best practice input fields. In Use Displays the fields that have been mapped to an input field or output field. All Displays all available fields. The Output tab has additional filter and sort capabilities that you access by clicking the column headers. You can filter each column of data to display one or more values, and also sort the fields in ascending or descending order. Icons in the column header indicate whether the column has a filter or sort applied to it. Because you can filter and sort on multiple columns, they are applied from left to right. The filter and sort menu is not available if there is only one item type in the column. Embedded help The embedded help is the place to look when you need more information about Data Services transforms and options. The topic changes to help you with the context you're currently in. When you select a new transform or a new option group, the topic updates to reflect that selection. You can also navigate to other topics by using hyperlinks within the open topic. Note: To view option information for the Associate, Match, and User-Defined transforms, you will need to open their respective editors by selecting the transform in the data flow and then choosing Tools > <transform> Editor. Related Topics • Associate, Match, and User-Defined transform editors 187 2011-06-09
- 188. Transforms 8.5.2.1 Associate, Match, and User-Defined transform editors The Associate, Match, and User-Defined transforms each have their own editor in which you can add option groups and edit options. The editors for these three transforms look and act similarly, and in some cases even share the same option groups. The editor window is divided into four areas: 1. Option Explorer — In this area, you select the option groups, or operations, that are available for the transform. To display an option group that is hidden, right-click the option group it belongs to and select the name of the option group from the menu. 2. Option Editor — In this area, you specify the value of the option. 3. Buttons — Use these to add, remove and order option groups. 4. Embedded help — The embedded help displays additional information about using the current editor screen. Related Topics • Reference Guide: Transforms, Associate • Reference Guide: Transforms, Match • Reference Guide: Transforms, User-Defined 188 2011-06-09
- 189. Transforms 8.5.2.2 Ordered options editor Some transforms allow you to choose and specify the order of multiple values for a single option. One example is the parser sequence option of the Data Cleanse transform. To configure an ordered option: 1. Click the Add and Remove buttons to move option values between the Available and Selected values lists. Note: Remove all values. To clear the Selected values list and move all option values to the Available values list, click Remove All. 2. Select a value in the Available values list, and click the up and down arrow buttons to change the position of the value in the list. 3. Click OK to save your changes to the option configuration. The values are listed in the Designer and separated by pipe characters. 8.6 Text Data Processing transforms Text Data Processing transform help you extract specific information from your text. You can parses large volumes of documents, identifying “entities” such as customers, products, locations, and financial information relevant to your organization. The following sections provide an overview of this fucntionality and the Entity Extraction transform. 8.6.1 Text Data Processing overview Text Data Processing analyzes text and automatically identifies and extracts entities, including people, dates, places, organizations and so on, in multiple languages. It looks for patterns, activities, events, and relationships among entities and enables their extraction. Extracting such information from text tells you what the text is about — this information can be used within applications for information management, data integration, and data quality; business intelligence; query, analytics and reporting; search, navigation, document and content management; among other usage scenarios. Text Data Processing goes beyond conventional character matching tools for information retrieval, which can only seek exact matches for specific strings. It understands semantics of words. In addition to known entity matching, it performs a complementary function of new entity discovery. To customize 189 2011-06-09
- 190. Transforms entity extraction, the software enables you to specify your own list of entities in a custom dictionary. These dictionaries enable you to store entities and manage name variations. Known entity names can be standardized using a dictionary. It also performs normalization of certain numeric expressions, such as dates. Text Data Processing automates extraction of key information from text sources to reduce manual review and tagging. This in turn can reduce cost towards understanding important insights hidden in text. Access to relevant information from unstructured text can help streamline operations and reduce unnecessary costs. In Data Services, text data processing refers to a set of transforms that extracts information from unstructured data and creates structured data that can be used by various business intelligence tools. 8.6.2 Entity Extraction transform overview Text data processing is accomplished in the software using the following transform: • Entity Extraction - Extracts entities and facts from unstructured text. Extraction involves processing and analyzing text, finding entities of interest, assigning them to the appropriate type, and presenting this metadata in a standard format. By using dictionaries and rules, you can customize your extraction output to include entities defined in them. Extraction applications are as diverse as your information needs. Some examples of information that can be extracted using this transform include: • • • • • Co-occurrence and associations of brand names, company names, people, supplies, and more. Competitive and market intelligence such as competitors’ activities, merger and acquisition events, press releases, contact information, and so on. A person’s associations, activities, or role in a particular event. Customer claim information, defect reports, or patient information such as adverse drug effects. Various alphanumeric patterns such as ID numbers, contract dates, profits, and so on. 8.6.2.1 Entities and Facts overview Entities denote names of people, places, and things that can be extracted. Entities are defined as a pairing of a name and its type. Type indicates the main category of an entity. Entities can be further broken down into subentities. A subentity is an embedded entity of the same semantic type as the containing entity. The subentity has a prefix that matches that of the larger, containing entity. Here are some examples of entities and subentities: • Eiffel Tower is an entity with name "Eiffel Tower" and type PLACE. • Mr. Joe Smith is an entity with name "Mr. Joe Smith" and type PERSON. For this entity, there are three subentities. 190 2011-06-09
- 191. Transforms • • • "Mr." is associated with subentity PERSON_PRE. Joe is associated with subentity PERSON_GIV. Smith is associated with subentity PERSON_FAM. Entities can also have subtypes. A subtype indicates further classification of an entity; it is a hierarchical specification of an entity type that enables the distinction between different semantic varieties of the same entity type. A subtype can be described as a sub-category of an entity. Here are some examples of entities and subtypes: • • • Airbus is an entity of type VEHICLE with a subtype AIR. Mercedes-Benz coupe is an entity of type VEHICLE with a subtype LAND. SAP is an entity of type ORGANIZATION with a subtype COMMERCIAL. Facts denote a pattern that creates an expression to extract information such as sentiments, events, or relationships. Facts are extracted using custom extraction rules. Fact is an umbrella term covering extractions of more complex patterns including one or more entities, a relationship between one or more entities, or some sort of predicate about an entity. Facts provide context of how different entities are connected in the text. Entities by themselves only show that they are present in a document, but facts provide information on how these entities are related. Fact types identify the category of a fact; for example, sentiments and requests. A subfact is a key piece of information embedded within a fact. A subfact type can be described as a category associated with the subfact. Here are some examples of facts and fact types: • SAP acquired Business Objects in a friendly takeover. This is an event of type merger and acquisition (M&A). • Mr. Joe Smith is very upset with his airline bookings. This is a fact of type SENTIMENT. How extraction works The extraction process uses its inherent knowledge of the semantics of words and the linguistic context in which these words occur to find entities and facts. It creates specific patterns to extract entities and facts based on system rules. You can add entries in a dictionary as well as write custom rules to customize extraction output. The following sample text and sample output shows how unstructured content can be transformed into structured information for further processing and analysis. Example: Sample text and extraction information "Mr. Jones is very upset with Green Insurance. The offer for his totaled vehicle is too low. He states that Green offered him $1250.00 but his car is worth anywhere from $2500 and $4500. Mr. Jones would like Green's comprehensive coverage to be in line with other competitors." This sample text when processed with the extraction transform would identify and group the information in a logical way (identifying entities, subentities, subtypes, facts, fact types, subfacts, and subfact types) that can be further processed. The following tables show information tagged as entities, entity types, subentities, subentity types, subtypes, facts, fact types, subfacts, and subfact types from the sample text: 191 2011-06-09
- 192. Transforms Enti ties Entity Type Mr. Jones PERSON Subtype Green PERSON_FAM ORGANIZATION 1250 USD, 2500 USD, 4500 USD PERSON_PRE Jones ORGANIZATION Subentity Type Mr. Green Insurance Subentities CURRENCY COMMERCIAL Note: The CURRENCY entities are normalized to display USD instead of a $ sign. Facts Subfact Subfact Type Mr. Jones is very upset with Green Insurance. SENTIMENT very upset StrongNegativeSentiment Jones would like that Green's comprehensive coverage to be in line with other competitors. 192 Fact Type REQUEST 2011-06-09
- 193. Transforms 8.6.2.2 Dictionary overview A text data processing dictionary is a user-defined repository of entities. It is an easy-to-use customization tool that specifies a list of entities that the extraction transform should always extract while processing text. The information is classified under the standard form and the variant of an entity. A standard form may have one or more variants embedded under it; variants are other commonly known names of an entity. For example, United Parcel Service of America is the standard form for that company, and United Parcel Service and UPS are both variants for the same company. While each standard form must have a type, variants can optionally have their own type; for example, while United Parcel Service of America is associated with a standard form type ORGANIZATION, you might define a variant type ABBREV to include abbreviations. A dictionary structure can help standardize references to an entity. Related Topics • Text Data Processing Extraction Customization Guide: Using Dictionaries 8.6.2.3 Rule overview A text data processing rule defines custom patterns to extract entities, relationships, events, and other larger extractions that are together referred to as facts. You write custom extraction rules to perform extraction that is customized to your specific needs. Related Topics • Text Data Processing Extraction Customization Guide: Using Extraction Rules 8.6.3 Using the Entity Extraction transform The Entity Extraction transform can extract information from any text, HTML, or XML content and generate output. You can use the output in several ways based on your work flow. You can use it as an input to another transform or write to multiple output sources such as a database table or a flat file. The output is generated in UTF-16 encoding. The following list provides some scenarios on when to use the transform alone or in combination with other Data Services transforms. 193 2011-06-09
- 194. Transforms • • • Searching for specific information and relationships from a large amount of text related to a broad domain. For example, a company is interested in analysing customer feedback received in free form text after a new product launch. Linking structured information from unstructured text together with existing structured information to make new connections. For example, a law enforcement department is trying to make connections between various crimes and people involved using their own database and information available in various reports in text format. Analyzing and reporting on product quality issues such as excessive repairs and returns for certain products. For example, you may have structured information about products, parts, customers, and suppliers in a database, while important information pertaining to problems may be in notes: fields of maintenance records, repair logs, product escalations, and support center logs. To identify the issues, you need to make connections between various forms of data. 8.6.4 Differences between text data processing and data cleanse transforms The Entity Extraction transform provides functionality similar to the Data Cleanse transform in certain cases, especially with respect to customization capabilities. This section describes the differences between the two and which transform to use to meet your goals. The Text Data Processing transform is for making sense of unstructured content and the Data Cleanse transform is for standardizing and cleansing structured data. The following table describes some of the main differences. In many cases, using a combination of Text Data Processing and Data Cleanse transforms will generate the data that is best suited for your business intelligence analyses and reports. Criteria Data Cleanse Input type Unstructured text that requires linguistic parsing to generate relevant information. Structured data represented as fields in records. Input size More than 5KB of text. Less than 5KB of text. Input scope Normally broad domain with many variations. Specific data domain with limited variations. Matching task Content discovery, noise reduction, pattern matching, and relationship between different entities. Dictionary lookup, pattern matching. Potential usage 194 Text Data Processing Identifies potentially meaningful information from unstructured content and extracts it into a format that can be stored in a repository. Ensures quality of data for matching and storing into a repository such as Meta Data Management. 2011-06-09
- 195. Transforms Criteria Text Data Processing Data Cleanse Output Creates annotations about the source text in the form of entities, entity types, facts, and their offset, length, and so on. Input is not altered. Creates parsed and standardized fields. Input is altered if desired. 8.6.5 Using multiple transforms You can include multiple transforms in the same dataflow to perform various analytics on unstructured information. For example, to extract names and addresses embedded in some text and validate the information before running analytics on the extracted information, you could: • • • Use the Entity Extraction transform to process text containing names and addresses and extract different entities. Pass the extraction output to the Case transform to identify which rows represent names and which rows represent addresses Use the Data Cleanse transform to standardize the extracted names and use the Global Address Cleanse transform to validate and correct the extracted address data. Note: To generate the correct data, include the standard_form and type fields in the Entity Extraction transform output schema; map the type field in the Case transform based on the entity type such as PERSON, ADDRESS, etc. Next, map any PERSON entities from the Case transform to the Data Cleanse transform and map any ADDRESS entities to the Global Address Cleanse transform. 8.6.6 Examples for using the Entity Extraction transform This section describes some examples for employing the Entity Extraction transform. The scenario is that a human resources department wants to analyze résumés received in a variety of formats. The formats include: • A text file as an attachment to an email • A text résumé pasted into a field on the company's Web site • Updates to résumé content that the department wants to process in real time 195 2011-06-09
- 196. Transforms Example: Text file email attachment The human resources department frequently receives résumés as attachments to emails from candidates. They store these attachments in a separate directory on a server. To analyze and process data from these text files: 1. Configure an Unstructured text file format that points to the directory of résumés. 2. Build a data flow with the unstructured text file format as the source, an Entity Extraction transform, and a target. 3. Configure the transform to process and analyze the text. Example: Text résumé pasted into a field on a Web site The human resources department's online job application form includes a field into which applicants can paste their résumés. This field is captured in a database table column. To analyze and process data from the database: 1. Configure a connection to the database via a datastore. 2. Build a data flow with the database table as the source, an Entity Extraction transform, and a target. 3. Configure the transform to process and analyze the text. Example: Updated content to be processed in real time Suppose the human resources department is seeking a particular qualification in an applicant. When the applicant updates her résumé in the company's Web-based form with the desired qualification, the HR manager wants to be immediately notified. Use a real-time job to enable this functionality. To analyze and process the data in real time: 1. Add a real-time job including begin and end markers and a data flow. Connect the objects. 2. Build the data flow with a message source, an Entity Extraction transform, and a message target. 3. Configure the transform to process and analyze the text. Related Topics • Unstructured file formats • Database datastores • Real-time Jobs 8.6.7 To add a text data processing transform to a data flow 1. Open a data flow object. 2. Open the local object library if it is not already open. 196 2011-06-09
- 197. Transforms 3. Go to the Transforms tab. 4. Expand the Text Data Processing transform folder and select the transform or transform configuration that you want to add to the data flow. 5. Drag the transform or transform configuration icon into the data flow workspace. If you selected a transform that has available transform configurations, a drop-down menu prompts you to select a transform configuration. 6. Draw the data flow connections. To connect a source or a transform to another transform, click the square on the right edge of the source or upstream transform and drag the cursor to the arrow on the left edge of the text data processing transform. • The input for the transform might be the output from another transform or the output from a source. • You can connect the output of the transform to the input of another transform or target. 7. Double-click the name of the transform. This opens the transform editor, which lets you complete the definition of the transform. 8. In the input schema, select the input field that you want to map and drag it to the appropriate field in the Input tab. This maps the input field to a field name that is recognized by the transform so that the transform knows how to process it correctly. For example, an input field that is named Content would be mapped to the TEXT input field. 9. In the Options tab, select the appropriate option values to determine how the transform will process your data. Make sure that you map input fields before you set option values. If you change an option value from its default value, a green triangle appears next to the option name to indicate that you made an override. 10. In the Output tab, double-click the fields that you want to output from the transform. The transforms can generate fields in addition to the input fields that the transform processes, so you can output many fields. Make sure that you set options before you map output fields. The selected fields appear in the output schema. The output schema of this transform becomes the input schema of the next transform in the data flow. 11. If you want to pass data through the transform without processing it, drag fields directly from the input schema to the output schema. 12. To rename or resize an output field, double-click the output field and edit the properties in the "Column Properties" window. Related Topics • Entity Extraction transform editor • Reference Guide: Entity Extraction transform, Input fields • Reference Guide: Entity Extraction transform, Output fields • Reference Guide: Entity Extraction transform, Extraction options 197 2011-06-09
- 198. Transforms 8.6.8 Entity Extraction transform editor The Entity Extraction transform options specify various parameters to process content using the transform. Filtering options, under different extraction options, enable you to limit the entities extracted to specific entities from a dictionary, the system files, rules, or a combination of them. Extraction options are divided into the following categories: • Common This option is set to specify that the Entity Extraction transform is to be run as a separate process. • Languages Mandatory option. Use this option to specify the language for the extraction process. The Entity Types filtering option is optional and you may select it when you select the language to limit your extraction output. • Processing Options Use these options to specify parameters to be used when processing the content. • Dictionaries Use this option to specify different dictionaries to be used for processing the content. To use the Entity Types filtering option, you must specify the Dictionary File. Note: Text Data Processing includes the dictionary schema file extraction-dictionary.xsd. By default, this file is installed in the LINK_DIR/bin folder, where LINK_DIR is your Data Services installation directory. Refer to this schema to create your own dictionary files. • Rules Use this option to specify different rule files to be used for processing the content. To use the Rule Names filtering option, you must specify the Rule File. If you do not specify any filtering options, the extraction output will contain all entities extracted using entity types defined in the selected language, dictionary file(s), and rule name(s) in the selected rule file(s). Note: Selecting a dictionary file or a rule file in the extraction process is optional. The extraction output will include the entities from them if they are specified. Related Topics • Importing XML Schemas • Reference Guide: Entity Extraction transform, Extraction options • Text Data Processing Extraction Customization Guide: Using Dictionaries 198 2011-06-09
- 199. Transforms 8.6.9 Using filtering options The filtering options under different extraction options control the output generated by the Entity Extraction transform. Using these options, you can limit the entities extracted to specific entities from a dictionary, the system files, rules, or a combination of them. For example, you are processing customer feedback fields for an automobile company and are interested in looking at the comments related to one specific model. Using the filtering options, you can control your output to extract data only related to that model. Filtering options are divided into three categories: • • • The Filter By Entity Types option under the Languages option group - Use this option to limit extraction output to include only selected entities for this language. The Filter By Entity Types option under the Dictionary option group - Use this option to limit extraction output to include only entities defined in a dictionary. The Filter By Rules Names option under the Rules option group - Use this option to limit extraction output to include only entities and facts returned by the specific rules. The following table describes information contained in the extraction output based on the combination of these options: Lan guages Dictio naries Rules Entity Types Entity Types Rule Names Yes No No Entities (extracted using the entity types) selected in the filter. No Entities (extracted using the entity types) defined in the selected language and entity types selected from the dictionaries filter. No Entities (extracted using the entity types) defined in the filters for the selected language and any specified dictionaries. Yes Entities (extracted using the entity types) defined in the selected language and any rule names selected in the filter from any specified rule files. No Yes No 199 Yes Yes No Extraction Output Content Note If multiple dictionaries are specified that contain the same entity type but it is only selected as a filter for one of them, entities of this type will also be returned from the other dictionary. If multiple rule files are specified that contain the same rule name but it is only selected as a filter for one of them, entities and facts of this type will also be returned from the other rule file. 2011-06-09
- 200. Transforms Lan guages No Yes Yes Dictio naries Yes No Yes Rules Extraction Output Content Yes Entities (extracted using entity types) defined in the selected language, entity types selected from the dictionaries filter, and any rule names selected in the filter from any specified rule files. Yes Entities (extracted using entity types) defined in the filters for the selected language and any rule names selected in the filter from any specified rule files. Yes Entities (extracted using entity types) defined in the filters for the selected language, entity types selected from the dictionaries filter, and any rule names selected in the filter from any specified rule files. Note The extraction process filters the output using the union of the extracted entities or facts for the selected language, the dictionaries, and the rule files. If you change your selection for the language, dictionaries, or rules, any filtering associated with that option will only be cleared by clicking the Filter by... option. You must select new filtering choices based on the changed selection. Note: • • If you are using multiple dictionaries (or rules) and have set filtering options for some of the selected dictionaries (or rules), the extraction process combines the dictionaries internally, and output is filtered using the union of the entity types selected for each dictionary and rule names selected for each rule file. The output will identify the source as a dictionary (or rule) file and not the individual name of a dictionary (or rule) file. If you select the Dictionary Only option under the Processing Options group, with a valid dictionary file, the entity types defined for the language are not included in the extraction output, but any extracted rule file entities and facts are included. Related Topics • Entity Extraction transform editor 200 2011-06-09
- 201. Work Flows Work Flows Related Topics • What is a work flow? • Steps in a work flow • Order of execution in work flows • Example of a work flow • Creating work flows • Conditionals • While loops • Try/catch blocks • Scripts 9.1 What is a work flow? A work flow defines the decision-making process for executing data flows. For example, elements in a work flow can determine the path of execution based on a value set by a previous job or can indicate an alternative path if something goes wrong in the primary path. Ultimately, the purpose of a work flow is to prepare for executing data flows and to set the state of the system after the data flows are complete. Jobs (introduced in Projects) are special work flows. Jobs are special because you can execute them. Almost all of the features documented for work flows also apply to jobs, with one exception: jobs do not have parameters. 201 2011-06-09
- 202. Work Flows 9.2 Steps in a work flow Work flow steps take the form of icons that you place in the work space to create a work flow diagram. The following objects can be elements in work flows: • Work flows • Data flows • Scripts • Conditionals • While loops • Try/catch blocks Work flows can call other work flows, and you can nest calls to any depth. A work flow can also call itself. The connections you make between the icons in the workspace determine the order in which work flows execute, unless the jobs containing those work flows execute in parallel. 9.3 Order of execution in work flows Steps in a work flow execute in a left-to-right sequence indicated by the lines connecting the steps. Here is the diagram for a work flow that calls three data flows: Note that Data_Flow1 has no connection from the left but is connected on the right to the left edge of Data_Flow2 and that Data_Flow2 is connected to Data_Flow3. There is a single thread of control connecting all three steps. Execution begins with Data_Flow1 and continues through the three data flows. Connect steps in a work flow when there is a dependency between the steps. If there is no dependency, the steps need not be connected. In that case, the software can execute the independent steps in the work flow as separate processes. In the following work flow, the software executes data flows 1 through 3 in parallel: 202 2011-06-09
- 203. Work Flows To execute more complex work flows in parallel, define each sequence as a separate work flow, then call each of the work flows from another work flow as in the following example: You can specify that a job execute a particular work flow or data flow only one time. In that case, the software only executes the first occurrence of the work flow or data flow; the software skips subsequent occurrences in the job. You might use this feature when developing complex jobs with multiple paths, such as jobs with try/catch blocks or conditionals, and you want to ensure that the software only executes a particular work flow or data flow one time. 9.4 Example of a work flow Suppose you want to update a fact table. You define a data flow in which the actual data transformation takes place. However, before you move data from the source, you want to determine when the fact table was last updated so that you only extract rows that have been added or changed since that date. You need to write a script to determine when the last update was made. You can then pass this date to the data flow as a parameter. In addition, you want to check that the data connections required to build the fact table are active when data is read from them. To do this in the software, you define a try/catch block. If the connections are not active, the catch runs a script you wrote, which automatically sends mail notifying an administrator of the problem. Scripts and error detection cannot execute in the data flow. Rather, they are steps of a decision-making process that influences the data flow. This decision-making process is defined as a work flow, which looks like the following: 203 2011-06-09
- 204. Work Flows The software executes these steps in the order that you connect them. 9.5 Creating work flows You can create work flows using one of two methods: • Object library • Tool palette After creating a work flow, you can specify that a job only execute the work flow one time, even if the work flow appears in the job multiple times. 9.5.1 To create a new work flow using the object library 1. Open the object library. 2. Go to the Work Flows tab. 3. Right-click and choose New. 4. Drag the work flow into the diagram. 5. Add the data flows, work flows, conditionals, try/catch blocks, and scripts that you need. 9.5.2 To create a new work flow using the tool palette 1. Select the work flow icon in the tool palette. 2. Click where you want to place the work flow in the diagram. If more than one instance of a work flow appears in a job, you can improve execution performance by running the work flow only one time. 9.5.3 To specify that a job executes the work flow one time 204 2011-06-09
- 205. Work Flows When you specify that a work flow should only execute once, a job will never re-execute that work flow after the work flow completes successfully, except if the work flow is contained in a work flow that is a recovery unit that re-executes and has not completed successfully elsewhere outside the recovery unit. It is recommended that you not mark a work flow as Execute only once if the work flow or a parent work flow is a recovery unit. 1. Right click on the work flow and select Properties. The Properties window opens for the work flow. 2. Select the Execute only once check box. 3. Click OK. Related Topics • Reference Guide: Work flow 9.6 Conditionals Conditionals are single-use objects used to implement if/then/else logic in a work flow. Conditionals and their components (if expressions, then and else diagrams) are included in the scope of the parent control flow's variables and parameters. To define a conditional, you specify a condition and two logical branches: Conditional branch Description If A Boolean expression that evaluates to TRUE or FALSE. You can use functions, variables, and standard operators to construct the expression. Then Work flow elements to execute if the If expression evaluates to TRUE. Else (Optional) Work flow elements to execute if the If expression evaluates to FALSE. Define the Then and Else branches inside the definition of the conditional. A conditional can fit in a work flow. Suppose you use a Windows command file to transfer data from a legacy system into the software. You write a script in a work flow to run the command file and return a success flag. You then define a conditional that reads the success flag to determine if the data is available for the rest of the work flow. 205 2011-06-09
- 206. Work Flows To implement this conditional in the software, you define two work flows—one for each branch of the conditional. If the elements in each branch are simple, you can define them in the conditional editor itself. Both the Then and Else branches of the conditional can contain any object that you can have in a work flow including other work flows, nested conditionals, try/catch blocks, and so on. 9.6.1 To define a conditional 1. Define the work flows that are called by the Then and Else branches of the conditional. It is recommended that you define, test, and save each work flow as a separate object rather than constructing these work flows inside the conditional editor. 2. Open the work flow in which you want to place the conditional. 3. Click the icon for a conditional in the tool palette. 4. Click the location where you want to place the conditional in the diagram. The conditional appears in the diagram. 5. Click the name of the conditional to open the conditional editor. 6. Click if. 7. Enter the Boolean expression that controls the conditional. Continue building your expression. You might want to use the function wizard or smart editor. 8. After you complete the expression, click OK. 9. Add your predefined work flow to the Then box. 206 2011-06-09
- 207. Work Flows To add an existing work flow, open the object library to the Work Flows tab, select the desired work flow, then drag it into the Then box. 10. (Optional) Add your predefined work flow to the Else box. If the If expression evaluates to FALSE and the Else box is blank, the software exits the conditional and continues with the work flow. 11. After you complete the conditional, choose DebugValidate. The software tests your conditional for syntax errors and displays any errors encountered. 12. The conditional is now defined. Click the Back button to return to the work flow that calls the conditional. 9.7 While loops Use a while loop to repeat a sequence of steps in a work flow as long as a condition is true. This section discusses: • Design considerations • Defining a while loop • Using a while loop with View Data 9.7.1 Design considerations The while loop is a single-use object that you can use in a work flow. The while loop repeats a sequence of steps as long as a condition is true. 207 2011-06-09
- 208. Work Flows Typically, the steps done during the while loop result in a change in the condition so that the condition is eventually no longer satisfied and the work flow exits from the while loop. If the condition does not change, the while loop will not end. For example, you might want a work flow to wait until the system writes a particular file. You can use a while loop to check for the existence of the file using the file_exists function. As long as the file does not exist, you can have the work flow go into sleep mode for a particular length of time, say one minute, before checking again. Because the system might never write the file, you must add another check to the loop, such as a counter, to ensure that the while loop eventually exits. In other words, change the while loop to check for the existence of the file and the value of the counter. As long as the file does not exist and the counter is less than a particular value, repeat the while loop. In each iteration of the loop, put the work flow in sleep mode and then increment the counter. 208 2011-06-09
- 209. Work Flows 9.7.2 Defining a while loop You can define a while loop in any work flow. 9.7.2.1 To define a while loop 1. Open the work flow where you want to place the while loop. 2. Click the while loop icon on the tool palette. 3. Click the location where you want to place the while loop in the workspace diagram. The while loop appears in the diagram. 4. Click the while loop to open the while loop editor. 5. In the While box at the top of the editor, enter the condition that must apply to initiate and repeat the steps in the while loop. Alternatively, click to open the expression editor, which gives you more space to enter an expression and access to the function wizard. Click OK after you enter an expression in the editor. 6. Add the steps you want completed during the while loop to the workspace in the while loop editor. You can add any objects valid in a work flow including scripts, work flows, and data flows. Connect these objects to represent the order that you want the steps completed. 209 2011-06-09
- 210. Work Flows Note: Although you can include the parent work flow in the while loop, recursive calls can create an infinite loop. 7. After defining the steps in the while loop, choose Debug > Validate. The software tests your definition for syntax errors and displays any errors encountered. 8. Close the while loop editor to return to the calling work flow. 9.7.3 Using a while loop with View Data When using View Data, a job stops when the software has retrieved the specified number of rows for all scannable objects. Depending on the design of your job, the software might not complete all iterations of a while loop if you run a job in view data mode: • If the while loop contains scannable objects and there are no scannable objects outside the while loop (for example, if the while loop is the last object in a job), then the job will complete after the scannable objects in the while loop are satisfied, possibly after the first iteration of the while loop. • If there are scannable objects after the while loop, the while loop will complete normally. Scanned objects in the while loop will show results from the last iteration. • If there are no scannable objects following the while loop but there are scannable objects completed in parallel to the while loop, the job will complete as soon as all scannable objects are satisfied. The while loop might complete any number of iterations. 9.8 Try/catch blocks A try/catch block is a combination of one try object and one or more catch objects that allow you to specify alternative work flows if errors occur while the software is executing a job. Try/catch blocks: • • • "Catch" groups of exceptions "thrown" by the software, the DBMS, or the operating system. Apply solutions that you provide for the exceptions groups or for specific errors within a group. Continue execution. Try and catch objects are single-use objects. Here's the general method to implement exception handling: 1. Insert a try object before the steps for which you are handling errors. 2. Insert a catch object in the work flow after the steps. 3. In the catch object, do the following: 210 2011-06-09
- 211. Work Flows • • • Select one or more groups of errors that you want to catch. Define the actions that a thrown exception executes. The actions can be a single script object, a data flow, a workflow, or a combination of these objects. Optional. Use catch functions inside the catch block to identify details of the error. If an exception is thrown during the execution of a try/catch block and if no catch object is looking for that exception, then the exception is handled by normal error logic. The following work flow shows a try/catch block surrounding a data flow: In this case, if the data flow BuildTable causes any system-generated exceptions specified in the catch Catch_A, then the actions defined in Catch_A execute. The action initiated by the catch object can be simple or complex. Here are some examples of possible exception actions: • • • Send the error message to an online reporting database or to your support group. Rerun a failed work flow or data flow. Run a scaled-down version of a failed work flow or data flow. Related Topics • Defining a try/catch block • Categories of available exceptions • Example: Catching details of an error • Reference Guide: Objects, Catch 9.8.1 Defining a try/catch block To define a try/catch block: 1. Open the work flow that will include the try/catch block. 2. Click the try icon in the tool palette. 3. Click the location where you want to place the try in the diagram. The try icon appears in the diagram. Note: There is no editor for a try; the try merely initiates the try/catch block. 4. Click the catch icon in the tool palette. 5. Click the location where you want to place the catch object in the work space. 211 2011-06-09
- 212. Work Flows The catch object appears in the work space. 6. Connect the try and catch objects to the objects they enclose. 7. Click the name of the catch object to open the catch editor. 8. Select one or more groups from the list of Exceptions. To select all exception groups, click the check box at the top. 9. Define the actions to take for each exception group and add the actions to the catch work flow box. The actions can be an individual script, a data flow, a work flow, or any combination of these objects. a. It is recommended that you define, test, and save the actions as a separate object rather than constructing them inside the catch editor. b. If you want to define actions for specific errors, use the following catch functions in a script that the work flow executes: • error_context() • error_message() • error_number() • error_timestamp() c. To add an existing work flow to the catch work flow box, open the object library to the Work Flows tab, select the desired work flow, and drag it into the box. 10. After you have completed the catch, choose Validation > Validate > All Objects in View. The software tests your definition for syntax errors and displays any errors encountered. 11. Click the Back button to return to the work flow that calls the catch. 12. If you want to catch multiple exception groups and assign different actions to each exception group, repeat steps 4 through 11 for each catch in the work flow. Note: In a sequence of catch blocks, if one catch block catches an exception, the subsequent catch blocks will not be executed. For example, if your work flow has the following sequence and Catch1 catches an exception, then Catch2 and CatchAll will not execute. Try > DataFlow1 > Catch1 > Catch2 > CatchAll If any error in the exception group listed in the catch occurs during the execution of this try/catch block, the software executes the catch work flow. Related Topics • Categories of available exceptions • Example: Catching details of an error • Reference Guide: Objects, Catch 9.8.2 Categories of available exceptions 212 2011-06-09
- 213. Work Flows Categories of available exceptions include: • • • • • • • • • • • • • Execution errors (1001) Database access errors (1002) Database connection errors (1003) Flat file processing errors (1004) File access errors (1005) Repository access errors (1006) SAP system errors (1007) System resource exception (1008) SAP BW execution errors (1009) XML processing errors (1010) COBOL copybook errors (1011) Excel book errors (1012) Data Quality transform errors (1013) 9.8.3 Example: Catching details of an error This example illustrates how to use the error functions in a catch script. Suppose you want to catch database access errors and send the error details to your support group. 1. In the catch editor, select the exception group that you want to catch. In this example, select the checkbox in front of Database access errors (1002). 2. In the work flow area of the catch editor, create a script object with the following script: mail_to('[email protected]', 'Data Service error number' || error_number(), 'Error message: ' || error_message(),20,20); print('DBMS Error: ' || error_message()); 3. This sample catch script includes the mail_to function to do the following: • Specify the email address of your support group. • Send the error number that the error_number() function returns for the exception caught. • Send the error message that the error_message() function returns for the exception caught. 4. The sample catch script includes a print command to print the error message for the database error. Related Topics • Reference Guide: Objects, Catch error functions • Reference Guide: Objects, Catch scripts 213 2011-06-09
- 214. Work Flows 9.9 Scripts Scripts are single-use objects used to call functions and assign values to variables in a work flow. For example, you can use the SQL function in a script to determine the most recent update time for a table and then assign that value to a variable. You can then assign the variable to a parameter that passes into a data flow and identifies the rows to extract from a source. A script can contain the following statements: • • • • • Function calls If statements While statements Assignment statements Operators The basic rules for the syntax of the script are as follows: • • • • • Each line ends with a semicolon (;). Variable names start with a dollar sign ($). String values are enclosed in single quotation marks ('). Comments start with a pound sign (#). Function calls always specify parameters even if the function uses no parameters. For example, the following script statement determines today's date and assigns the value to the variable $TODAY: $TODAY = sysdate(); You cannot use variables unless you declare them in the work flow that calls the script. Related Topics • Reference Guide: Data Services Scripting Language 9.9.1 To create a script 1. Open the work flow. 2. Click the script icon in the tool palette. 3. Click the location where you want to place the script in the diagram. The script icon appears in the diagram. 4. Click the name of the script to open the script editor. 214 2011-06-09
- 215. Work Flows 5. Enter the script statements, each followed by a semicolon. The following example shows a script that determines the start time from the output of a custom function. AW_StartJob ('NORMAL','DELTA', $G_STIME,$GETIME); $GETIME =to_date( sql('ODS_DS','SELECT to_char(MAX(LAST_UPDATE) , 'YYYY-MM-DDD HH24:MI:SS') FROM EMPLOYEE'), 'YYYY_MMM_DDD_HH24:MI:SS'); Click the function button to include functions in your script. 6. After you complete the script, select Validation > Validate. The software tests your script for syntax errors and displays any errors encountered. 7. Click the ... button and then save to name and save your script. The script is saved by default in <LINKDIR>/BusinessObjects Data Services/ DataQuality/Samples. 9.9.2 Debugging scripts using the print function The software has a debugging feature that allows you to print: • The values of variables and parameters during execution • The execution path followed within a script You can use the print function to write the values of parameters and variables in a work flow to the trace log. For example, this line in a script: print('The value of parameter $x: [$x]'); produces the following output in the trace log: The following output is being printed via the Print function in <Session job_name>. The value of parameter $x: value Related Topics • Reference Guide: Functions and Procedures, print 215 2011-06-09
- 217. Nested Data Nested Data This section discusses nested data and how to use them in the software. 10.1 What is nested data? Real-world data often has hierarchical relationships that are represented in a relational database with master-detail schemas using foreign keys to create the mapping. However, some data sets, such as XML documents and SAP ERP IDocs, handle hierarchical relationships through nested data. The software maps nested data to a separate schema implicitly related to a single row and column of the parent schema. This mechanism is called Nested Relational Data Modelling (NRDM). NRDM provides a way to view and manipulate hierarchical relationships within data flow sources, targets, and transforms. Sales orders are often presented using nesting: the line items in a sales order are related to a single header and are represented using a nested schema. Each row of the sales order data set contains a nested line item schema. 10.2 Representing hierarchical data You can represent the same hierarchical data in several ways. Examples include: • Multiple rows in a single data set Order data set 217 2011-06-09
- 218. Nested Data Order No ShipTo1 ShipTo2 Item Qty ItemPrice 9999 1001 123 State St Town, CA 001 2 10 9999 • CustID 1001 123 State St Town, CA 002 4 5 Multiple data sets related by a join Order header data set OrderNo CustID ShipTo1 ShipTo2 9999 1001 123 State St Town, CA Line-item data set OrderNo Item Qty ItemPrice 9999 001 2 10 9999 002 4 5 WHERE Header.OrderNo=LineItem.OrderNo • Nested data Using the nested data method can be more concise (no repeated information), and can scale to present a deeper level of hierarchical complexity. For example, columns inside a nested schema can also contain columns. There is a unique instance of each nested schema for each row at each level of the relationship. Order data set Generalizing further with nested data, each row at each level can have any number of columns containing nested schemas. 218 2011-06-09
- 219. Nested Data Order data set You can see the structure of nested data in the input and output schemas of sources, targets, and transforms in data flows. Nested schemas appear with a schema icon paired with a plus sign, which indicates that the object contains columns. The structure of the schema shows how the data is ordered. • • LineItems is a nested schema. The minus sign in front of the schema icon indicates that the column list is open. • 219 Sales is the top-level schema. CustInfo is a nested schema with the column list closed. 2011-06-09
- 220. Nested Data 10.3 Formatting XML documents The software allows you to import and export metadata for XML documents (files or messages), which you can use as sources or targets in jobs. XML documents are hierarchical. Their valid structure is stored in separate format documents. The format of an XML file or message (.xml) can be specified using either an XML Schema (.xsd for example) or a document type definition (.dtd). When you import a format document's metadata, it is structured into the software's internal schema for hierarchical documents which uses the nested relational data model (NRDM). Related Topics • Importing XML Schemas • Specifying source options for XML files • Mapping optional schemas • Using Document Type Definitions (DTDs) • Generating DTDs and XML Schemas from an NRDM schema 10.3.1 Importing XML Schemas The software supports WC3 XML Schema Specification 1.0. For an XML document that contains information to place a sales order—order header, customer, and line items—the corresponding XML Schema includes the order structure and the relationship between data. 220 2011-06-09
- 221. Nested Data Message with data OrderNo CustID ShipTo1 ShipTo2 9999 1001 123 State St LineItems Town, CA Item ItemQty ItemPrice 001 2 10 002 4 5 Each column in the XML document corresponds to an ELEMENT or attribute definition in the XML schema. Corresponding XML schema <?xml version="1.0"?> <xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema"> <xs:element name="Order"> <xs:complexType> <xs:sequence> <xs:element name="OrderNo" type="xs:string" /> <xs:element name="CustID" type="xs:string" /> <xs:element name="ShipTo1" type="xs:string" /> <xs:element name="ShipTo2" type="xs:string" /> <xs:element maxOccurs="unbounded" name="LineItems"> <xs:complexType> <xs:sequence> <xs:element name="Item" type="xs:string" /> <xs:element name="ItemQty" type="xs:string" /> <xs:element name="ItemPrice" type="xs:string" /> </xs:sequence> </xs:complexType> </xs:element> </xs:sequence> </xs:complexType> </xs:element> </xs:schema> Related Topics • Reference Guide: XML schema 10.3.1.1 Importing XML schemas Import the metadata for each XML Schema you use. The object library lists imported XML Schemas in the Formats tab. When importing an XML Schema, The software reads the defined elements and attributes, then imports the following: • • 221 Document structure Namespace 2011-06-09
- 222. Nested Data • Table and column names • Data type of each column • Content type of each column • Nested table and column attributes While XML Schemas make a distinction between elements and attributes, the software imports and converts them all to nested table and column attributes. Related Topics • Reference Guide: XML schema 10.3.1.1.1 To import an XML Schema 1. From the object library, click the Format tab. 2. Right-click the XML Schemas icon. 3. Enter the settings for the XML schemas that you import. When importing an XML Schema: • Enter the name you want to use for the format in the software. • Enter the file name of the XML Schema or its URL address. Note: If your Job Server is on a different computer than the Designer, you cannot use Browse to specify the file path. You must type the path. You can type an absolute path or a relative path, but the Job Server must be able to access it. • If the root element name is not unique within the XML Schema, select a name in the Namespace drop-down list to identify the imported XML Schema. Note: When you import an XML schema for a real-time web service job, you should use a unique target namespace for the schema. When Data Services generates the WSDL file for a real-time job with a source or target schema that has no target namespace, it adds an automatically generated target namespace to the types section of the XML schema. This can reduce performance because Data Services must suppress the namespace information from the web service request during processing, and then reattach the proper namespace information before returning the response to the client. • • If the XML Schema contains recursive elements (element A contains B, element B contains A), specify the number of levels it has by entering a value in the Circular level box. This value must match the number of recursive levels in the XML Schema's content. Otherwise, the job that uses this XML Schema will fail. • 222 In the Root element name drop-down list, select the name of the primary node you want to import. The software only imports elements of the XML Schema that belong to this node or any subnodes. You can set the software to import strings as a varchar of any size. Varchar 1024 is the default. 2011-06-09
- 223. Nested Data 4. Click OK. After you import an XML Schema, you can edit its column properties such as data type using the General tab of the Column Properties window. You can also view and edit nested table and column attributes from the Column Properties window. 10.3.1.1.2 To view and edit nested table and column attributes for XML Schema 1. From the object library, select the Formats tab. 2. Expand the XML Schema category. 3. Double-click an XML Schema name. The XML Schema Format window appears in the workspace. The Type column displays the data types that the software uses when it imports the XML document metadata. 4. Double-click a nested table or column and select Attributes to view or edit XML Schema attributes. Related Topics • Reference Guide: XML schema 10.3.1.2 Importing abstract types An XML schema uses abstract types to force substitution for a particular element or type. • When an element is defined as abstract, a member of the element's substitution group must appear in the instance document. • When a type is defined as abstract, the instance document must use a type derived from it (identified by the xsi:type attribute). For example, an abstract element PublicationType can have a substitution group that consists of complex types such as MagazineType, BookType, and NewspaperType. The default is to select all complex types in the substitution group or all derived types for the abstract type, but you can choose to select a subset. 10.3.1.2.1 To limit the number of derived types to import for an abstract type 1. On the Import XML Schema Format window, when you enter the file name or URL address of an XML Schema that contains an abstract type, the Abstract type button is enabled. For example, the following excerpt from an xsd defines the PublicationType element as abstract with derived types BookType and MagazineType: <xsd:complexType name="PublicationType" abstract="true"> <xsd:sequence> <xsd:element name="Title" type="xsd:string"/> 223 2011-06-09
- 224. Nested Data <xsd:element name="Author" type="xsd:string" minOccurs="0" maxOccurs="unbounded"/> <xsd:element name="Date" type="xsd:gYear"/> </xsd:sequence> </xsd:complexType> <xsd:complexType name="BookType"> <xsd:complexContent> <xsd:extension base="PublicationType"> <xsd:sequence> <xsd:element name="ISBN" type="xsd:string"/> <xsd:element name="Publisher" type="xsd:string"/> </xsd:sequence> </xsd:extension> /xsd:complexContent> </xsd:complexType> <xsd:complexType name="MagazineType"> <xsd:complexContent> <xsd:restriction base="PublicationType"> <xsd:sequence> <xsd:element name="Title" type="xsd:string"/> <xsd:element name="Author" type="xsd:string" minOccurs="0" maxOccurs="1"/> <xsd:element name="Date" type="xsd:gYear"/> </xsd:sequence> </xsd:restriction> </xsd:complexContent> </xsd:complexType> 2. To select a subset of derived types for an abstract type, click the Abstract type button and take the following actions: a. From the drop-down list on the Abstract type box, select the name of the abstract type. b. Select the check boxes in front of each derived type name that you want to import. c. Click OK. Note: When you edit your XML schema format, the software selects all derived types for the abstract type by default. In other words, the subset that you previously selected is not preserved. 10.3.1.3 Importing substitution groups An XML schema uses substitution groups to assign elements to a special group of elements that can be substituted for a particular named element called the head element. The list of substitution groups can have hundreds or even thousands of members, but an application typically only uses a limited number of them. The default is to select all substitution groups, but you can choose to select a subset. 10.3.1.3.1 To limit the number of substitution groups to import 1. On the Import XML Schema Format window, when you enter the file name or URL address of an XML Schema that contains substitution groups, the Substitution Group button is enabled. For example, the following excerpt from an xsd defines the PublicationType element with substitution groups MagazineType, BookType, AdsType, and NewspaperType: <xsd:element name="Publication" type="PublicationType"/> <xsd:element name="BookStore"> <xsd:complexType> <xsd:sequence> <xsd:element ref="Publication" maxOccurs="unbounded"/> </xsd:sequence> </xsd:complexType> 224 2011-06-09
- 225. Nested Data </xsd:element> <xsd:element name="Magazine" type="MagazineType" substitutionGroup="Publication"/> <xsd:element name="Book" type="BookType" substitutionGroup="Publication"/> <xsd:element name="Ads" type="AdsType" substitutionGroup="Publication"/> <xsd:element name="Newspaper" type="NewspaperType" substitutionGroup="Publication"/> 2. Click the Substitution Group button and take the following actions a. From the drop-down list on the Substitution group box, select the name of the substitution group. b. Select the check boxes in front of each substitution group name that you want to import. c. Click OK. Note: When you edit your XML schema format, the software selects all elements for the substitution group by default. In other words, the subset that you previously selected is not preserved. 10.3.2 Specifying source options for XML files After you import metadata for XML documents (files or messages), you create a data flow to use the XML documents as sources or targets in jobs. 10.3.2.1 Creating a data flow with a source XML file 10.3.2.1.1 To create a data flow with a source XML file 1. From the object library, click the Format tab. 2. Expand the XML Schema and drag the XML Schema that defines your source XML file into your data flow. 3. Place a query in the data flow and connect the XML source to the input of the query. 4. Double-click the XML source in the work space to open the XML Source File Editor. 5. You must specify the name of the source XML file in the XML file text box. Related Topics • Reading multiple XML files at one time • Identifying source file names • Reference Guide: XML file source 225 2011-06-09
- 226. Nested Data 10.3.2.2 Reading multiple XML files at one time The software can read multiple files with the same format from a single directory using a single source object. 10.3.2.2.1 To read multiple XML files at one time 1. Open the editor for your source XML file 2. In XML File on the Source tab, enter a file name containing a wild card character (* or ?). For example: D:orders1999????.xml might read files from the year 1999 D:orders*.xml reads all files with the xml extension from the specified directory Related Topics • Reference Guide: XML file source 10.3.2.3 Identifying source file names You might want to identify the source XML file for each row in your source output in the following situations: 1. You specified a wildcard character to read multiple source files at one time 2. You load from a different source file on different days 10.3.2.3.1 To identify the source XML file for each row in the target 1. In the XML Source File Editor, select Include file name column which generates a column DI_FILENAME to contain the name of the source XML file. 2. In the Query editor, map the DI_FILENAME column from Schema In to Schema Out. 3. When you run the job, the target DI_FILENAME column will contain the source XML file name for each row in the target. 10.3.3 Mapping optional schemas 226 2011-06-09
- 227. Nested Data You can quickly specify default mapping for optional schemas without having to manually construct an empty nested table for each optional schema in the Query transform. Also, when you import XML schemas (either through DTDs or XSD files), the software automatically marks nested tables as optional if the corresponding option was set in the DTD or XSD file. The software retains this option when you copy and paste schemas into your Query transforms. This feature is especially helpful when you have very large XML schemas with many nested levels in your jobs. When you make a schema column optional and do not provide mapping for it, the software instantiates the empty nested table when you run the job. While a schema element is marked as optional, you can still provide a mapping for the schema by appropriately programming the corresponding sub-query block with application logic that specifies how the software should produce the output. However, if you modify any part of the sub-query block, the resulting query block must be complete and conform to normal validation rules required for a nested query block. You must map any output schema not marked as optional to a valid nested query block. The software generates a NULL in the corresponding PROJECT list slot of the ATL for any optional schema without an associated, defined sub-query block. 10.3.3.1 To make a nested table "optional" 1. Right-click a nested table and select Optional to toggle it on. To toggle it off, right-click the nested table again and select Optional again. 2. You can also right-click a nested table and select Properties, then go to the Attributes tab and set the Optional Table attribute value to yes or no. Click Apply and OK to set. Note: If the Optional Table value is something other than yes or no, this nested table cannot be marked as optional. When you run a job with a nested table set to optional and you have nothing defined for any columns and nested tables beneath that table, the software generates special ATL and does not perform user interface validation for this nested table. Example: CREATE NEW Query ( EMPNO int KEY , ENAME varchar(10), JOB varchar (9) NT1 al_nested_table ( DEPTNO int KEY , DNAME varchar (14), NT2 al_nested_table (C1 int) ) SET("Optional Table" = 'yes') ) AS SELECT EMP.EMPNO, EMP.ENAME, EMP.JOB, NULL FROM EMP, DEPT; Note: You cannot mark top-level schemas, unnested tables, or nested tables containing function calls optional. 227 2011-06-09
- 228. Nested Data 10.3.4 Using Document Type Definitions (DTDs) The format of an XML document (file or message) can be specified by a document type definition (DTD). The DTD describes the data contained in the XML document and the relationships among the elements in the data. For an XML document that contains information to place a sales order—order header, customer, and line items—the corresponding DTD includes the order structure and the relationship between data. Message with data OrderNo CustID ShipTo1 ShipTo2 9999 1001 123 State St LineItems Town, CA Item ItemQty ItemPrice 001 2 10 002 4 5 Each column in the XML document corresponds to an ELEMENT definition. Corresponding DTD Definition <?xml encoding="UTF-8"?> <!ELEMENT Order (OrderNo, CustID, ShipTo1, ShipTo2, LineItems+)> <!ELEMENT OrderNo (#PCDATA)> <!ELEMENT CustID (#PCDATA)> <!ELEMENT ShipTo1 (#PCDATA)> <!ELEMENT ShipTo2 (#PCDATA)> <!ELEMENT LineItems (Item, ItemQty, ItemPrice)> <!ELEMENT Item (#PCDATA)> <!ELEMENT ItemQty (#PCDATA)> <!ELEMENT ItemPrice (#PCDATA)> Import the metadata for each DTD you use. The object library lists imported DTDs in the Formats tab. You can import metadata from either an existing XML file (with a reference to a DTD) or DTD file. If you import the metadata from an XML file, the software automatically retrieves the DTD for that XML file. When importing a DTD, the software reads the defined elements and attributes. The software ignores other parts of the definition, such as text and comments. This allows you to modify imported XML data and edit the data type as needed. Related Topics • Reference Guide: DTD 228 2011-06-09
- 229. Nested Data 10.3.4.1 To import a DTD or XML Schema format 1. From the object library, click the Format tab. 2. Right-click the DTDs icon and select New. 3. Enter settings into the Import DTD Format window: • In the DTD definition name box, enter the name you want to give the imported DTD format in the software. • Enter the file that specifies the DTD you want to import. Note: If your Job Server is on a different computer than the Designer, you cannot use Browse to specify the file path. You must type the path. You can type an absolute path or a relative path, but the Job Server must be able to access it. • If importing an XML file, select XML for the File type option. If importing a DTD file, select the DTD option. • In the Root element name box, select the name of the primary node you want to import. The software only imports elements of the DTD that belong to this node or any subnodes. • If the DTD contains recursive elements (element A contains B, element B contains A), specify the number of levels it has by entering a value in the Circular level box. This value must match the number of recursive levels in the DTD's content. Otherwise, the job that uses this DTD will fail. • You can set the software to import strings as a varchar of any size. Varchar 1024 is the default. 4. Click OK. After you import a DTD, you can edit its column properties such as data type using the General tab of the Column Properties window. You can also view and edit DTD nested table and column attributes from the Column Properties window. 10.3.4.2 To view and edit nested table and column attributes for DTDs 1. From the object library, select the Formats tab. 2. Expand the DTDs category. 3. Double-click a DTD name. The DTD Format window appears in the workspace. 4. Double-click a nested table or column. 229 2011-06-09
- 230. Nested Data The Column Properties window opens. 5. Select the Attributes tab to view or edit DTD attributes. 10.3.5 Generating DTDs and XML Schemas from an NRDM schema You can right-click any schema from within a query editor in the Designer and generate a DTD or an XML Schema that corresponds to the structure of the selected schema (either NRDM or relational). This feature is useful if you want to stage data to an XML file and subsequently read it into another data flow. 1. Generate a DTD/XML Schema. 2. Use the DTD/XML Schema to setup an XML format 3. Use the XML format to set up an XML source for the staged file. The DTD/XML Schema generated will be based on the following information: • Columns become either elements or attributes based on whether the XML Type attribute is set to ATTRIBUTE or ELEMENT. • If the Required attribute is set to NO, the corresponding element or attribute is marked optional. • Nested tables become intermediate elements. • The Native Type attribute is used to set the type of the element or attribute. • While generating XML Schemas, the MinOccurs and MaxOccurs values will be set based on the Minimum Occurrence and Maximum Occurrence attributes of the corresponding nested table. No other information is considered while generating the DTD or XML Schema. Related Topics • Reference Guide: DTD • Reference Guide: XML schema 10.4 Operations on nested data This section discusses the operations that you can perform on nested data. 230 2011-06-09
- 231. Nested Data 10.4.1 Overview of nested data and the Query transform With relational data, a Query transform allows you to execute a SELECT statement. The mapping between input and output schemas defines the project list for the statement. When working with nested data, the Query transform provides an interface to perform SELECT statements at each level of the relationship that you define in the output schema. You use the Query transform to manipulate nested data. If you want to extract only part of the nested data, you can use the XML_Pipeline transform. Without nested schemas, the Query transform assumes that the FROM clause in the SELECT statement contains the data sets that are connected as inputs to the query object. When working with nested data, you must explicitly define the FROM clause in a query. The software assists by setting the top-level inputs as the default FROM clause values for the top-level output schema. The other SELECT statement elements defined by the query work the same with nested data as they do with flat data. However, because a SELECT statement can only include references to relational data sets, a query that includes nested data includes a SELECT statement to define operations for each parent and child schema in the output. The Query Editor contains a tab for each clause of the query: • SELECT provides an option to specify distinct rows to output (discarding any identical duplicate rows). • FROM lists all input schemas and allows you to specify join pairs and conditions. The parameters you enter for the following tabs apply only to the current schema (displayed in the Schema Out text box at the top right of the Query Editor): • WHERE • GROUP BY • ORDER BY Related Topics • Query Editor • Reference Guide: XML_Pipeline 10.4.2 FROM clause construction The FROM clause is located at the bottom of the FROM tab. It automatically populates with the information included in the Input Schema(s) section at the top, and the Join Pairs section in the middle 231 2011-06-09
- 232. Nested Data of the tab. You can change the FROM clause by changing the selected schema in the Input Schema(s) area, and the Join Pairs section. Schemas selected in the Input Schema(s) section (and reflected in the FROM clause), including columns containing nested schemas, are available to be included in the output. When you include more than one schema in the Input Schema(s) section (By selecting the "From" check box), you can specify join pairs and join conditions as well as enter join rank and cache for each input schema. FROM clause descriptions and the behavior of the query are exactly the same with nested data as with relational data. The current schema allows you to distinguish multiple SELECT statements from each other within a single query. However, because the SELECT statements are dependent upon each other, and because the user interface makes it easy to construct arbitrary data sets, determining the appropriate FROM clauses for multiple levels of nesting can be complex. A FROM clause can contain: • Any top-level schema from the input • Any schema that is a column of a schema in the FROM clause of the parent schema • Any join conditions from the join pairs The FROM clause forms a path that can start at any level of the output. The first schema in the path must always be a top-level schema from the input. The data that a SELECT statement from a lower schema produces differs depending on whether or not a schema is included in the FROM clause at the top-level. The next two examples use the sales order data set to illustrate scenarios where FROM clause values change the data resulting from the query. Related Topics • To modify the output schema contents 10.4.2.1 Example: FROM clause includes all top-level inputs To include detailed customer information for all of the orders in the output, join the Order_Status_In schema at the top-level with the Cust schema. Include both input schemas at the top-level in the FROM clause to produce the appropriate data. When you select both input schemas in the Input schema(s) area of the FROM tab, they automatically appear in the FROM clause. 232 2011-06-09
- 233. Nested Data Observe the following points in the Query Editor above: • The Input schema(s) table in the FROM tab includes the two top-level schemas Order_Status_In and Cust (this is also reflected in the FROM clause). • The Schema Out pane shows the nested schema, cust_info, and the columns Cust_ID, Customer_name, and Address. 10.4.2.2 Example: Lower level FROM clause contains top-level input Suppose you want the detailed information from one schema to appear for each row in a lower level of another schema. For example, the input includes a top-level Materials schema and a nested LineItems schema, and you want the output to include detailed material information for each line item. The graphic below illustrates how this is set up in Designer. 233 2011-06-09
- 234. Nested Data The example on the left shows the following setup: • • The Input Schema area in the FROM tab shows the nested schema LineItems selected. The FROM tab shows the FROM Clause “FROM "Order".LineItems”. The example on the right shows the following setup: • • • • The Materials.Description schema is mapped to LineItems.Item output schema. The Input schema(s) Materials and Order.LineItems are selected in the Input Schema area in the FROM tab (the From column has a check mark). A Join Pair is created joining the nested Order.LineItems schema with the top-level Materials schema using a left outer join type. A Join Condition is added where the Item field under the nested schema LineItems is equal to the Item field in the top-level Materials schema. The resulting FROM Clause: "Order".LineItems.Item = Materials.Item 10.4.3 Nesting columns 234 2011-06-09
- 235. Nested Data When you nest rows of one schema inside another, the data set produced in the nested schema is the result of a query against the first one using the related values from the second one. For example, if you have sales-order information in a header schema and a line-item schema, you can nest the line items under the header schema. The line items for a single row of the header schema are equal to the results of a query including the order number: SELECT * FROM LineItems WHERE Header.OrderNo = LineItems.OrderNo You can use a query transform to construct a nested data set from relational data. When you indicate the columns included in the nested schema, specify the query used to define the nested data set for each row of the parent schema. 10.4.3.1 To construct a nested data set Follow the steps below to set up a nested data set. 1. Create a data flow with the input sources that you want to include in the nested data set. 2. Place a Query transform and a target table in the data flow. Connect the sources to the input of the query. 3. Open the Query transform and set up the select list, from clause, and where clause to describe the SELECT statement that the query executes to determine the top-level data set. • Select list: Map the input schema items to the output schema by draging the columns from the input schema to the output schema. You can also include new columns or include mapping expressions for the columns. • FROM clause: Include the input sources in the list on the FROM tab, and include any joins and join conditions required to define the data. • WHERE clause: Include any filtering required to define the data set for the top-level output. 4. Create a new schema in the output. Right-click in the Schema Out area of the Query Editor and choose New Output Schema. A new schema icon appears in the output, nested under the top-level schema. You can also drag an entire schema from the input to the output. 235 2011-06-09
- 236. Nested Data 5. Change the current output schema to the nested schema by right-clicking the nested schema and selecting Make Current. The Query Editor changes to display the new current schema. 6. Indicate the FROM clause, select list, and WHERE clause to describe the SELECT statement that the query executes to determine the top-level data set. • FROM clause: If you created a new output schema, you need to drag schemas from the input to populate the FROM clause. If you dragged an existing schema from the input to the top-level output, that schema is automatically mapped and listed in the From tab. • Select list: Only columns are available that meet the requirements for the FROM clause. • WHERE clause: Only columns are available that meet the requirements for the FROM clause. 7. If the output requires it, nest another schema at this level. Repeat steps 4 through 6 in this current schema for as many nested schemas that you want to set up. 8. If the output requires it, nest another schema under the top level. Make the top-level schema the current schema. Related Topics • Query Editor • FROM clause construction • To modify the output schema contents 10.4.4 Using correlated columns in nested data Correlation allows you to use columns from a higher-level schema to construct a nested schema. In a nested-relational model, the columns in a nested schema are implicitly related to the columns in the parent row. To take advantage of this relationship, you can use columns from the parent schema in the construction of the nested schema. The higher-level column is a correlated column. Including a correlated column in a nested schema can serve two purposes: • The correlated column is a key in the parent schema. Including the key in the nested schema allows you to maintain a relationship between the two schemas after converting them from the nested data model to a relational model. • The correlated column is an attribute in the parent schema. Including the attribute in the nested schema allows you to use the attribute to simplify correlated queries against the nested data. To include a correlated column in a nested schema, you do not need to include the schema that includes the column in the FROM clause of the nested schema. 236 2011-06-09
- 237. Nested Data 10.4.4.1 To used a correlated column in a nested schema 1. Create a data flow with a source that includes a parent schema with a nested schema. For example, the source could be an order header schema that has a LineItems column that contains a nested schema. 2. Connect a query to the output of the source. 3. In the query editor, copy all columns of the parent schema to the output. In addition to the top-level columns, the software creates a column called LineItems that contains a nested schema that corresponds to the LineItems nested schema in the input. 4. Change the current schema to the LineItems schema. (For information on setting the current schema and completing the parameters, see Query Editor.) 5. Include a correlated column in the nested schema. Correlated columns can include columns from the parent schema and any other schemas in the FROM clause of the parent schema. For example, drag the OrderNo column from the Header schema into the LineItems schema. Including the correlated column creates a new output column in the LineItems schema called OrderNo and maps it to the Order.OrderNo column. The data set created for LineItems includes all of the LineItems columns and the OrderNo. If the correlated column comes from a schema other than the immediate parent, the data in the nested schema includes only the rows that match both the related values in the current row of the parent schema and the value of the correlated column. You can always remove the correlated column from the lower-level schema in a subsequent query transform. 10.4.5 Distinct rows and nested data The Distinct rows option in Query transforms removes any duplicate rows at the top level of a join. This is particularly useful to avoid cross products in joins that produce nested output. 10.4.6 Grouping values across nested schemas 237 2011-06-09
- 238. Nested Data When you specify a Group By clause for a schema with a nested schema, the grouping operation combines the nested schemas for each group. For example, to assemble all the line items included in all the orders for each state from a set of orders, you can set the Group By clause in the top level of the data set to the state column (Order.State) and create an output schema that includes State column (set to Order.State) and LineItems nested schema. The result is a set of rows (one for each state) that has the State column and the LineItems nested schema that contains all the LineItems for all the orders for that state. 10.4.7 Unnesting nested data Loading a data set that contains nested schemas into a relational (non-nested) target requires that the nested rows be unnested. For example, a sales order may use a nested schema to define the relationship between the order header and the order line items. To load the data into relational schemas, the multi-level must be unnested. Unnesting a schema produces a cross-product of the top-level schema (parent) and the nested schema (child). 238 2011-06-09
- 239. Nested Data It is also possible that you would load different columns from different nesting levels into different schemas. A sales order, for example, may be flattened so that the order number is maintained separately with each line item and the header and line item information loaded into separate schemas. The software allows you to unnest any number of nested schemas at any depth. No matter how many levels are involved, the result of unnesting schemas is a cross product of the parent and child schemas. When more than one level of unnesting occurs, the inner-most child is unnested first, then the result—the cross product of the parent and the inner-most child—is then unnested from its parent, and so on to the top-level schema. Unnesting all schemas (cross product of all data) might not produce the results you intend. For example, if an order includes multiple customer values such as ship-to and bill-to addresses, flattening a sales order by unnesting customer and line-item schemas produces rows of data that might not be useful for processing the order. 239 2011-06-09
- 240. Nested Data 10.4.7.1 To unnest nested data 1. Create the output that you want to unnest in the output schema of a query. Data for unneeded columns or schemas might be more difficult to filter out after the unnesting operation. You can use the Cut command to remove columns or schemas from the top level; to remove nested schemas or columns inside nested schemas, make the nested schema the current schema, and then cut the unneeded columns or nested columns. 2. For each of the nested schemas that you want to unnest, right-click the schema name and choose Unnest. The output of the query (the input to the next step in the data flow) includes the data in the new relationship, as the following diagram shows. 240 2011-06-09
- 241. Nested Data 10.4.8 Transforming lower levels of nested data Nested data included in the input to transforms (with the exception of a query or XML_Pipeline transform) passes through the transform without being included in the transform's operation. Only the columns at the first level of the input data set are available for subsequent transforms. 10.4.8.1 To transform values in lower levels of nested schemas 1. Take one of the following actions to obtain the nested data • Use a query transform to unnest the data. • Use an XML_Pipeline transform to select portions of the nested data. • Perform the transformation. 2. Nest the data again to reconstruct the nested relationships. Related Topics • Unnesting nested data • Reference Guide: XML_Pipeline 10.5 XML extraction and parsing for columns In addition to extracting XML message and file data, representing it as NRDM data during transformation, then loading it to an XML message or file, you can also use the software to extract XML data stored in a source table or flat file column, transform it as NRDM data, then load it to a target or flat file column. More and more database vendors allow you to store XML in one column. The field is usually a varchar, long, or clob. The software's XML handling capability also supports reading from and writing to such fields. The software provides four functions to support extracting from and loading to columns: • • load_to_xml • long_to_varchar • 241 extract_from_xml varchar_to_long 2011-06-09
- 242. Nested Data The extract_from_xml function gets the XML content stored in a single column and builds the corresponding NRDM structure so that the software can transform it. This function takes varchar data only. To enable extracting and parsing for columns, data from long and clob columns must be converted to varchar before it can be transformed by the software. • The software converts a clob data type input to varchar if you select the Import unsupported data types as VARCHAR of size option when you create a database datastore connection in the Datastore Editor. • If your source uses a long data type, use the long_to_varchar function to convert data to varchar. Note: The software limits the size of the XML supported with these methods to 100K due to the current limitation of its varchar data type. There are plans to lift this restriction in the future. The function load_to_xml generates XML from a given NRDM structure in the software, then loads the generated XML to a varchar column. If you want a job to convert the output to a long column, use the varchar_to_long function, which takes the output of the load_to_xml function as input. 10.5.1 Sample scenarios The following scenarios describe how to use functions to extract XML data from a source column and load it into a target column. Related Topics • Extracting XML data from a column into the software • Loading XML data into a column of the data type long • Extracting data quality XML strings using extract_from_xml function 10.5.1.1 Extracting XML data from a column into the software This scenario uses long_to_varchar and extract_from_xml functions to extract XML data from a column with data of the type long. 1. First, assume you have previously performed the following steps: a. Imported an Oracle table that contains a column named Content with the data type long, which contains XML data for a purchase order. b. Imported the XML Schema PO.xsd, which provides the format for the XML data, into the repository. c. Created a Project, a job, and a data flow for your design. 242 2011-06-09
- 243. Nested Data d. Opened the data flow and dropped the source table with the column named content in the data flow. 2. From this point: a. Create a query with an output column of data type varchar, and make sure that its size is big enough to hold the XML data. b. Name this output column content. c. In the Map section of the query editor, open the Function Wizard, select the Conversion function type, then select the long_to_varchar function and configure it by entering its parameters. long_to_varchar(content, 4000) The second parameter in this function (4000 in this case) is the maximum size of the XML data stored in the table column. Use this parameter with caution. If the size is not big enough to hold the maximum XML data for the column, the software will truncate the data and cause a runtime error. Conversely, do not enter a number that is too big, which would waste computer memory at runtime. d. In the query editor, map the source table column to a new output column. e. Create a second query that uses the function extract_from_xml to extract the XML data. To invoke the function extract_from_xml, right-click the current context in the query, choose New Function Call. When the Function Wizard opens, select Conversion and extract_from_xml. Note: You can only use the extract_from_xml function in a new function call. Otherwise, this function is not displayed in the function wizard. f. Enter values for the input parameters. • The first is the XML column name. Enter content, which is the output column in the previous query that holds the XML data • The second parameter is the DTD or XML Schema name. Enter the name of the purchase order schema (in this case PO) • The third parameter is Enable validation. Enter 1 if you want the software to validate the XML with the specified Schema. Enter 0 if you do not. g. Click Next. h. For the function, select a column or columns that you want to use on output. Imagine that this purchase order schema has five top-level elements: orderDate, shipTo, billTo, comment, and items. You can select any number of the top-level columns from an XML schema, which include either scalar or NRDM column data. The return type of the column is defined in the schema. If the function fails due to an error when trying to produce the XML output, the software returns NULL for scalar columns and empty nested tables for NRDM columns. The extract_from_xml function also adds two columns: • 243 AL_ERROR_NUM — returns error codes: 0 for success and a non-zero integer for failures 2011-06-09
- 244. Nested Data • AL_ERROR_MSG — returns an error message if AL_ERROR_NUM is not 0. Returns NULL if AL_ERROR_NUM is 0 Choose one or more of these columns as the appropriate output for the extract_from_xml function. i. Click Finish. The software generates the function call in the current context and populates the output schema of the query with the output columns you specified. With the data converted into the NRDM structure, you are ready to do appropriate transformation operations on it. For example, if you want to load the NRDM structure to a target XML file, create an XML file target and connect the second query to it. Note: If you find that you want to modify the function call, right-click the function call in the second query and choose Modify Function Call. In this example, to extract XML data from a column of data type long, we created two queries: the first query to convert the data using the long_to_varchar function and the second query to add the extract_from_xml function. Alternatively, you can use just one query by entering the function expression long_to_varchar directly into the first parameter of the function extract_from_xml. The first parameter of the function extract_from_xml can take a column of data type varchar or an expression that returns data of type varchar. If the data type of the source column is not long but varchar, do not include the function long_to_varchar in your data flow. 10.5.1.2 Loading XML data into a column of the data type long This scenario uses the load_to_xml function and the varchar_to_long function to convert an NRDM structure to scalar data of the varchar type in an XML format and load it to a column of the data type long. In this example, you want to convert an NRDM structure for a purchase order to XML data using the function load_to_xml, and then load the data to an Oracle table column called content, which is of the long data type. Because the function load_to_xml returns a value of varchar data type, you use the function varchar_to_long to convert the value of varchar data type to a value of the data type long. 1. Create a query and connect a previous query or source (that has the NRDM structure of a purchase order) to it. In this query, create an output column of the data type varchar called content. Make sure the size of the column is big enough to hold the XML data. 2. From the Mapping area open the function wizard, click the category Conversion Functions, and then select the function load_to_xml. 244 2011-06-09
- 245. Nested Data 3. Click Next. 4. Enter values for the input parameters. The function load_to_xml has seven parameters. 5. Click Finish. In the mapping area of the Query window, notice the function expression: load_to_xml(PO, 'PO', 1, '<?xml version="1.0" encoding = "UTF-8" ?>', NULL, 1, 4000) In this example, this function converts the NRDM structure of purchase order PO to XML data and assigns the value to output column content. 6. Create another query with output columns matching the columns of the target table. a. Assume the column is called content and it is of the data type long. b. Open the function wizard from the mapping section of the query and select the Conversion Functions category c. Use the function varchar_to_long to map the input column content to the output column content. The function varchar_to_long takes only one input parameter. d. Enter a value for the input parameter. varchar_to_long(content) 7. Connect this query to a database target. Like the example using the extract_from_xml function, in this example, you used two queries. You used the first query to convert an NRDM structure to XML data and to assign the value to a column of varchar data type. You used the second query to convert the varchar data type to long. You can use just one query if you use the two functions in one expression: varchar_to_long( load_to_xml(PO, 'PO', 1, '<?xml version="1.0" encoding = "UTF-8" ?>', NULL, 1, 4000) ) If the data type of the column in the target database table that stores the XML data is varchar, there is no need for varchar_to_long in the transformation. Related Topics • Reference Guide: Functions and Procedure 10.5.1.3 Extracting data quality XML strings using extract_from_xml function This scenario uses the extract_from_xml function to extract XML data from the Geocoder, Global Suggestion Lists, Global Address Cleanse, and USA Regulatory Address Cleanse transforms. 245 2011-06-09
- 246. Nested Data The Geocoder transform, Global Suggestion Lists transform, and the suggestion list functionality in the Global Address Cleanse and USA Regulatory Address Cleanse transforms can output a field that contains an XML string. The transforms output the following fields that can contain XML. Transform XML output field Geocoder Result_List Global Address Cleanse Suggestion_List Global Suggestion List USA Regulatory Address Cleanse Output field description Contains an XML output string when multiple records are returned for a search. The content depends on the available data. Contains an XML output string that includes all of the suggestion list component field values specified in the transform options. To output these fields as XML, you must choose XML as the output style in the transform options. To use the data contained within the XML strings (for example, in a web application that uses the job published as a web service), you must extract the data. There are two methods that you can use to extract the data: 1. Insert a Query transform using the extract_from_xml function. With this method, you insert a Query transform into the dataflow after the Geocoder, Global Suggestion Lists, Global Address Cleanse, or USA Regulatory Address Cleanse transform. Then you use the extract_from_xml function to parse the nested output data. This method is considered a best practice, because it provides parsed output data that is easily accessible to an integrator. 2. Develop a simple data flow that does not unnest the nested data. With this method, you simply output the output field that contains the XML string without unnesting the nested data. This method allows the application developer, or integrator, to dynamically select the output components in the final output schema before exposing it as a web service. The application developer must work closely with the data flow designer to understand the data flow behind a real-time web service. The application developer must understand the transform options and specify what to return from the return address suggestion list, and then unnest the XML output string to generate discrete address elements. 10.5.1.3.1 To extract data quality XML strings using extract_from_xml function 1. Create an XSD file for the output. 2. In the Format tab of the Local Object Library, create an XML Schema for your output XSD. 246 2011-06-09
- 247. Nested Data 3. In the Format tab of the Local Object Library, create an XML Schema for the gac_sugges tion_list.xsd, global_suggestion_list.xsd,urac_suggestion_list.xsd, or re sult_list.xsd. 4. In the data flow, include the following field in the Schema Out of the transform: • For the Global Address Cleanse, Global Suggestion Lists, and USA Regulatory Address Cleanse transforms, include the Suggestion_List field. • For the Geocoder transform, include the Result_List field 5. Add a Query transform after the Global Address Cleanse, Global Suggestion Lists,USA Regulatory Address Cleanse, or Geocoder transform. Complete it as follows. 6. Pass through all fields except the Suggestion_List or Result_List field from the Schema In to the Schema Out. To do this, drag fields directly from the input schema to the output schema. 7. In the Schema Out, right-click the Query node and select New Output Schema. Enter Suggestion_List or Result_List as the schema name (or whatever the field name is in your output XSD). 8. In the Schema Out, right-click the Suggestion_List or Result_List field and select Make Current. 9. In the Schema Out, right-click the Suggestion_List or Result_List list field and select New Function Call. 10. Select extract_from_xml from the Conversion Functions category and click Next. In the Define Input Parameter(s) window, enter the following information and click Next. • XML field name—Select the Suggestion_List or Result_List field from the upstream transform. • DTD or Schema name—Select the XML Schema that you created for the gac_suggestion_list.xsd, urac_suggestion_list.xsd, or result_list.xsd. • Enable validation—Enter 1 to enable validation. 11. Select LIST or RECORD from the left parameter list and click the right arrow button to add it to the Selected output parameters list. 12. Click Finish. The Schema Out includes the suggestion list/result list fields within the Suggestion_List or Result_List field. 13. Include the XML Schema for your output XML following the Query. Open the XML Schema to validate that the fields are the same in both the Schema In and the Schema Out. 14. If you are extracting data from a Global Address Cleanse, Global Suggestion Lists, or USA Regulatory Address Cleanse transform, and have chosen to output only a subset of the available suggestion list output fields in the Options tab, insert a second Query transform to specify the fields that you want to output. This allows you to select the output components in the final output schema before it is exposed as a web service. 247 2011-06-09
- 249. Real-time Jobs Real-time Jobs The software supports real-time data transformation. Real-time means that the software can receive requests from ERP systems and Web applications and send replies immediately after getting the requested data from a data cache or a second application. You define operations for processing on-demand messages by building real-time jobs in the Designer. 11.1 Request-response message processing The message passed through a real-time system includes the information required to perform a business transaction. The content of the message can vary: • It could be a sales order or an invoice processed by an ERP system destined for a data cache. • It could be an order status request produced by a Web application that requires an answer from a data cache or back-office system. The Access Server constantly listens for incoming messages. When a message is received, the Access Server routes the message to a waiting process that performs a predefined set of operations for the message type. The Access Server then receives a response for the message and replies to the originating application. Two components support request-response message processing: • Access Server — Listens for messages and routes each message based on message type. • Real-time job — Performs a predefined set of operations for that message type and creates a response. Processing might require that additional data be added to the message from a data cache or that the message data be loaded to a data cache. The Access Server returns the response to the originating application. 249 2011-06-09
- 250. Real-time Jobs 11.2 What is a real-time job? The Designer allows you to define the processing of real-time messages using a real-time job. You create a different real-time job for each type of message your system can produce. 11.2.1 Real-time versus batch Like a batch job, a real-time job extracts, transforms, and loads data. Real-time jobs "extract" data from the body of the message received and from any secondary sources used in the job. Each real-time job can extract data from a single message type. It can also extract data from other sources such as tables or files. The same powerful transformations you can define in batch jobs are available in real-time jobs. However, you might use transforms differently in real-time jobs. For example, you might use branches and logic controls more often than you would in batch jobs. If a customer wants to know when they can pick up their order at your distribution center, you might want to create a CheckOrderStatus job using a look-up function to count order items and then a case transform to provide status in the form of strings: "No items are ready for pickup" or "X items in your order are ready for pickup" or "Your order is ready for pickup". Also in real-time jobs, the software writes data to message targets and secondary targets in parallel. This ensures that each message receives a reply as soon as possible. Unlike batch jobs, real-time jobs do not execute in response to a schedule or internal trigger; instead, real-time jobs execute as real-time services started through the Administrator. Real-time services then wait for messages from the Access Server. When the Access Server receives a message, it passes the message to a running real-time service designed to process this message type. The real-time service processes the message and returns a response. The real-time service continues to listen and process messages on demand until it receives an instruction to shut down. 250 2011-06-09
- 251. Real-time Jobs 11.2.2 Messages How you design a real-time job depends on what message you want it to process. Typical messages include information required to implement a particular business operation and to produce an appropriate response. For example, suppose a message includes information required to determine order status for a particular order. The message contents might be as simple as the sales order number. The corresponding real-time job might use the input to query the right sources and return the appropriate product information. In this case, the message contains data that can be represented as a single column in a single-row table. In a second case, a message could be a sales order to be entered into an ERP system. The message might include the order number, customer information, and the line-item details for the order. The message processing could return confirmation that the order was submitted successfully. In this case, the message contains data that cannot be represented in a single table; the order header information can be represented by a table and the line items for the order can be represented by a second table. The software represents the header and line item data in the message in a nested relationship. 251 2011-06-09
- 252. Real-time Jobs When processing the message, the real-time job processes all of the rows of the nested table for each row of the top-level table. In this sales order, both of the line items are processed for the single row of header information. Real-time jobs can send only one row of data in a reply message (message target). However, you can structure message targets so that all data is contained in a single row by nesting tables within columns of a single, top-level table. The software data flows support the nesting of tables within other tables. Related Topics • Nested Data 11.2.3 Real-time job examples These examples provide a high-level description of how real-time jobs address typical real-time scenarios. Later sections describe the actual objects that you would use to construct the logic in the Designer. 11.2.3.1 Loading transactions into a back-office application A real-time job can receive a transaction from a Web application and load it to a back-office application (ERP, SCM, legacy). Using a query transform, you can include values from a data cache to supplement the transaction before applying it against the back-office application (such as an ERP system). 252 2011-06-09
- 253. Real-time Jobs 11.2.3.2 Collecting back-office data into a data cache You can use messages to keep the data cache current. Real-time jobs can receive messages from a back-office application and load them into a data cache or data warehouse. 11.2.3.3 Retrieving values, data cache, back-office applications You can create real-time jobs that use values from a data cache to determine whether or not to query the back-office application (such as an ERP system) directly. 253 2011-06-09
- 254. Real-time Jobs 11.3 Creating real-time jobs You can create real-time jobs using the same objects as batch jobs (data flows, work flows, conditionals, scripts, while loops, etc.). However, object usage must adhere to a valid real-time job model. 11.3.1 Real-time job models 11.3.1.1 Single data flow model With the single data flow model, you create a real-time job using a single data flow in its real-time processing loop. This single data flow must include a single message source and a single message target. 11.3.1.2 Multiple data flow model The multiple data flow model allows you to create a real-time job using multiple data flows in its real-time processing loop. By using multiple data flows, you can ensure that data in each message is completely processed in an initial data flow before processing for the next data flows starts. For example, if the data represents 40 254 2011-06-09
- 255. Real-time Jobs items, all 40 must pass though the first data flow to a staging or memory table before passing to a second data flow. This allows you to control and collect all the data in a message at any point in a real-time job for design and troubleshooting purposes. If you use multiple data flows in a real-time processing loop: • The first object in the loop must be a data flow. This data flow must have one and only one message source. • The last object in the loop must be a data flow. This data flow must have a message target. • Additional data flows cannot have message sources or targets. • You can add any number of additional data flows to the loop, and you can add them inside any number of work flows. • All data flows can use input and/or output memory tables to pass data sets on to the next data flow. Memory tables store data in memory while a loop runs. They improve the performance of real-time jobs with multiple data flows. 11.3.2 Using real-time job models 11.3.2.1 Single data flow model When you use a single data flow within a real-time processing loop your data flow diagram might look like this: Notice that the data flow has one message source and one message target. 255 2011-06-09
- 256. Real-time Jobs 11.3.2.2 Multiple data flow model When you use multiple data flows within a real-time processing loop your data flow diagrams might look like those in the following example scenario in which Data Services writes data to several targets according to your multiple data flow design. Example scenario requirements: Your job must do the following tasks, completing each one before moving on to the next: • • • Receive requests about the status of individual orders from a web portal and record each message to a backup flat file Perform a query join to find the status of the order and write to a customer database table. Reply to each message with the query join results Solution: First, create a real-time job and add a data flow, a work flow, and another data flow to the real-time processing loop. Second, add a data flow to the work flow. Next, set up the tasks in each data flow: • The first data flow receives the XML message (using an XML message source) and records the message to the flat file (flat file format target). Meanwhile, this same data flow writes the data into a memory table (table target). Note: You might want to create a memory table to move data to sequential data flows. For more information, see Memory datastores. • The second data flow reads the message data from the memory table (table source), performs a join with stored data (table source), and writes the results to a database table (table target) and a new memory table (table target). Notice this data flow has neither a message source nor a message target. 256 2011-06-09
- 257. Real-time Jobs • The last data flow sends the reply. It reads the result of the join in the memory table (table source) and loads the reply (XML message target). Related Topics • Designing real-time applications 11.3.3 To create a real-time job with a single dataflow 1. In the Designer, create or open an existing project. 2. From the project area, right-click the white space and select New Real-time job from the shortcut menu. New_RTJob1 appears in the project area. The workspace displays the job's structure, which consists of two markers: • RT_Process_begins • Step_ends These markers represent the beginning and end of a real-time processing loop. 3. In the project area, rename New_RTJob1. Always add a prefix to job names with their job type. In this case, use the naming convention: RTJOB_JobName. Although saved real-time jobs are grouped together under the Job tab of the object library, job names may also appear in text editors used to create adapter or Web Services calls. In these cases, a prefix saved with the job name will help you identify it. 4. If you want to create a job with a single data flow: a. Click the data flow icon in the tool palette. 257 2011-06-09
- 258. Real-time Jobs You can add data flows to either a batch or real-time job. When you place a data flow icon into a job, you are telling Data Services to validate the data flow according the requirements of the job type (batch or real-time). b. Click inside the loop. The boundaries of a loop are indicated by begin and end markers. One message source and one message target are allowed in a real-time processing loop. c. Connect the begin and end markers to the data flow. d. Build the data flow including a message source and message target. e. Add, configure, and connect initialization object(s) and clean-up object(s) as needed. 11.4 Real-time source and target objects Real-time jobs must contain a real-time source and/or target object. Those normally available are: Object Description Used as a: Software Access XML message An XML message structured in a DTD or XML Schema format Source or target Directly or through adapters Outbound message A real-time message with an application-specific format (not readable by XML parser) Target Through an adapter You can also use IDoc messages as real-time sources for SAP applications. For more information, see the Supplement for SAP. Adding sources and targets to real-time jobs is similar to adding them to batch jobs, with the following additions: For Prerequisite Object library location XML messages Import a DTD or XML Schema to define a format Formats tab Outbound message Define an adapter datastore and import object metadata. Datastores tab, under adapter datastore Related Topics • To import a DTD or XML Schema format • Adapter datastores 258 2011-06-09
- 259. Real-time Jobs 11.4.1 To view an XML message source or target schema In the workspace of a real-time job, click the name of an XML message source or XML message target to open its editor. If the XML message source or target contains nested data, the schema displays nested tables to represent the relationships among the data. 11.4.2 Secondary sources and targets Real-time jobs can also have secondary sources or targets (see Source and target objects). For example, suppose you are processing a message that contains a sales order from a Web application. The order contains the customer name, but when you apply the order against your ERP system, you need to supply more detailed customer information. Inside a data flow of a real-time job, you can supplement the message with the customer information to produce the complete document to send to the ERP system. The supplementary information might come from the ERP system itself or from a data cache containing the same information. Tables and files (including XML files) as sources can provide this supplementary information. The software reads data from secondary sources according to the way you design the data flow. The software loads data to secondary targets in parallel with a target message. Add secondary sources and targets to data flows in real-time jobs as you would to data flows in batch jobs (See Adding source or target objects to data flows). 11.4.3 Transactional loading of tables 259 2011-06-09
- 260. Real-time Jobs Target tables in real-time jobs support transactional loading, in which the data resulting from the processing of a single data flow can be loaded into multiple tables as a single transaction. No part of the transaction applies if any part fails. Note: Target tables in batch jobs also support transactional loading. However, use caution when you consider enabling this option for a batch job because it requires the use of memory, which can reduce performance when moving large amounts of data. You can specify the order in which tables in the transaction are included using the target table editor. This feature supports a scenario in which you have a set of tables with foreign keys that depend on one with primary keys. You can use transactional loading only when all the targets in a data flow are in the same datastore. If the data flow loads tables in more than one datastore, targets in each datastore load independently. While multiple targets in one datastore may be included in one transaction, the targets in another datastores must be included in another transaction. You can specify the same transaction order or distinct transaction orders for all targets to be included in the same transaction. If you specify the same transaction order for all targets in the same datastore, the tables are still included in the same transaction but are loaded together. Loading is committed after all tables in the transaction finish loading. If you specify distinct transaction orders for all targets in the same datastore, the transaction orders indicate the loading orders of the tables. The table with the smallest transaction order is loaded first, and so on, until the table with the largest transaction order is loaded last. No two tables are loaded at the same time. Loading is committed when the last table finishes loading. 11.4.4 Design tips for data flows in real-time jobs Keep in mind the following when you are designing data flows: • • In real-time jobs, do not cache data from secondary sources unless the data is static. The data will be read when the real-time job starts and will not be updated while the job is running. • If no rows are passed to the XML target, the real-time job returns an empty response to the Access Server. For example, if a request comes in for a product number that does not exist, your job might be designed in such a way that no data passes to the reply message. You might want to provide appropriate instructions to your user (exception handling in your job) to account for this type of scenario. • 260 If you include a table in a join with a real-time source, the software includes the data set from the real-time source as the outer loop of the join. If more than one supplementary source is included in the join, you can control which table is included in the next outer-most loop of the join using the join ranks for the tables. If more than one row passes to the XML target, the target reads the first row and discards the other rows. To avoid this issue, use your knowledge of the software's Nested Relational Data Model (NRDM) and structure your message source and target formats so that one "row" equals one 2011-06-09
- 261. Real-time Jobs message. With NRDM, you can structure any amount of data into a single "row" because columns in tables can contain other tables. • Recovery mechanisms are not supported in real-time jobs. Related Topics • Reference Guide: Objects, Real-time job • Nested Data 11.5 Testing real-time jobs 11.5.1 Executing a real-time job in test mode You can test real-time job designs without configuring the job as a service associated with an Access Server. In test mode, you can execute a real-time job using a sample source message from a file to determine if the software produces the expected target message. 11.5.1.1 To specify a sample XML message and target test file 1. In the XML message source and target editors, enter a file name in the XML test file box. Enter the full path name for the source file that contains your sample data. Use paths for both test files relative to the computer that runs the Job Server for the current repository. 2. Execute the job. Test mode is always enabled for real-time jobs. The software reads data from the source test file and loads it into the target test file. 11.5.2 Using View Data To ensure that your design returns the results you expect, execute your job using View Data. With View Data, you can capture a sample of your output data to ensure your design is working. 261 2011-06-09
- 262. Real-time Jobs Related Topics • Design and Debug 11.5.3 Using an XML file target You can use an "XML file target" to capture the message produced by a data flow while allowing the message to be returned to the Access Server. Just like an XML message, you define an XML file by importing a DTD or XML Schema for the file, then dragging the format into the data flow definition. Unlike XML messages, you can include XML files as sources or targets in batch and real-time jobs. 11.5.3.1 To use a file to capture output from steps in a real-time job 1. In the Formats tab of the object library, drag the DTD or XML Schema into a data flow of a real-time job. A menu prompts you for the function of the file. 2. Choose Make XML File Target. The XML file target appears in the workspace. 3. In the file editor, specify the location to which the software writes data. Enter a file name relative to the computer running the Job Server. 4. Connect the output of the step in the data flow that you want to capture to the input of the file. 262 2011-06-09
- 263. Real-time Jobs 11.6 Building blocks for real-time jobs 11.6.1 Supplementing message data The data included in messages from real-time sources might not map exactly to your requirements for processing or storing the information. If not, you can define steps in the real-time job to supplement the message information. One technique for supplementing the data in a real-time source includes these steps: 1. Include a table or file as a source. In addition to the real-time source, include the files or tables from which you require supplementary information. 2. Use a query to extract the necessary data from the table or file. 3. Use the data in the real-time source to find the necessary supplementary data. You can include a join expression in the query to extract the specific values required from the supplementary source. The Join Condition joins the two input schemas resulting in output for only the sales item document and line items included in the input from the application. 263 2011-06-09
- 264. Real-time Jobs Be careful to use data in the join that is guaranteed to return a value. If no value returns from the join, the query produces no rows and the message returns to the Access Server empty. If you cannot guarantee that a value returns, consider these alternatives: • Lookup function call — Returns a default value if no match is found • Outer join — Always returns a value, even if no match is found 11.6.1.1 To supplement message data In this example, a request message includes sales order information and its reply message returns order status. The business logic uses the customer number and priority rating to determine the level of status to return. The message includes only the customer name and the order number. A real-time job is then defined to retrieve the customer number and rating from other sources before determining the order status. 1. Include the real-time source in the real-time job. 2. Include the supplementary source in the real-time job. This source could be a table or file. In this example, the supplementary information required doesn't change very often, so it is reasonable to extract the data from a data cache rather than going to an ERP system directly. 3. Join the sources. In a query transform, construct a join on the customer name: Message.CustName = Cust_Status.CustName You can construct the output to include only the columns that the real-time job needs to determine order status. 4. Complete the real-time job to determine order status. The example shown here determines order status in one of two methods based on the customer status value. Order status for the highest ranked customers is determined directly from the ERP. Order status for other customers is determined from a data cache of sales order information. 264 2011-06-09
- 265. Real-time Jobs The logic can be arranged in a single or multiple data flows. The illustration below shows a single data flow model. Both branches return order status for each line item in the order. The data flow merges the results and constructs the response. The next section describes how to design branch paths in a data flow. 11.6.2 Branching data flow based on a data cache value One of the most powerful things you can do with a real-time job is to design logic that determines whether responses should be generated from a data cache or if they must be generated from data in a back-office application (ERP, SCM, CRM). Here is one technique for constructing this logic: 1. Determine the rule for when to access the data cache and when to access the back-office application. 2. Compare data from the real-time source with the rule. 3. Define each path that could result from the outcome. You might need to consider the case where the rule indicates back-office application access, but the system is not currently available. 4. Merge the results from each path into a single data set. 5. Route the single result to the real-time target. You might need to consider error-checking and exception-handling to make sure that a value passes to the target. If the target receives an empty set, the real-time job returns an empty response (begin and end XML tags only) to the Access Server. 265 2011-06-09
- 266. Real-time Jobs This example describes a section of a real-time job that processes a new sales order. The section is responsible for checking the inventory available of the ordered products—it answers the question, "is there enough inventory on hand to fill this order?" The rule controlling access to the back-office application indicates that the inventory (Inv) must be more than a pre-determined value (IMargin) greater than the ordered quantity (Qty) to consider the data cached inventory value acceptable. The software makes a comparison for each line item in the order they are mapped. Table 11-3: Incoming sales order LineItem OrderNo CustID Item 001 1001 Qty 7333 300 002 9999 Material 2288 1400 Table 11-4: Inventory data cache Material Inv IMargin 7333 600 100 2288 1500 200 Note: The quantity of items in the sales order is compared to inventory values in the data cache. 11.6.3 Calling application functions A real-time job can use application functions to operate on data. You can include tables as input or output parameters to the function. Application functions require input values for some parameters and some can be left unspecified. You must determine the requirements of the function to prepare the appropriate inputs. To make up the input, you can specify the top-level table, top-level columns, and any tables nested one-level down relative to the tables listed in the FROM clause of the context calling the function. If the application function includes a structure as an input parameter, you must specify the individual columns that make up the structure. 266 2011-06-09
- 267. Real-time Jobs A data flow may contain several steps that call a function, retrieve results, then shape the results into the columns and tables required for a response. 11.7 Designing real-time applications The software provides a reliable and low-impact connection between a Web application and back-office applications such as an enterprise resource planning (ERP) system. Because each implementation of an ERP system is different and because the software includes versatile decision support logic, you have many opportunities to design a system that meets your internal and external information and resource needs. 11.7.1 Reducing queries requiring back-office application access This section provides a collection of recommendations and considerations that can help reduce the time you spend experimenting in your development cycles. The information you allow your customers to access through your Web application can impact the performance that your customers see on the Web. You can maximize performance through your Web application design decisions. In particular, you can structure your application to reduce the number of queries that require direct back-office (ERP, SCM, Legacy) application access. For example, if your ERP system supports a complicated pricing structure that includes dependencies such as customer priority, product availability, or order quantity, you might not be able to depend on values from a data cache for pricing information. The alternative might be to request pricing information directly from the ERP system. ERP system access is likely to be much slower than direct database access, reducing the performance your customer experiences with your Web application. To reduce the impact of queries requiring direct ERP system access, modify your Web application. Using the pricing example, design the application to avoid displaying price information along with standard product information and instead show pricing only after the customer has chosen a specific product and quantity. These techniques are evident in the way airline reservations systems provide pricing information—a quote for a specific flight—contrasted with other retail Web sites that show pricing for every item displayed as part of product catalogs. 11.7.2 Messages from real-time jobs to adapter instances If a real-time job will send a message to an adapter instance, refer to the adapter documentation to decide if you need to create a message function call or an outbound message. 267 2011-06-09
- 268. Real-time Jobs • Message function calls allow the adapter instance to collect requests and send replies. • Outbound message objects can only send outbound messages. They cannot be used to receive messages. Related Topics • Importing metadata through an adapter datastore 11.7.3 Real-time service invoked by an adapter instance This section uses terms consistent with Java programming. (Please see your adapter SDK documentation for more information about terms such as operation instance and information resource.) When an operation instance (in an adapter) gets a message from an information resource, it translates it to XML (if necessary), then sends the XML message to a real-time service. In the real-time service, the message from the adapter is represented by a DTD or XML Schema object (stored in the Formats tab of the object library). The DTD or XML Schema represents the data schema for the information resource. The real-time service processes the message from the information resource (relayed by the adapter) and returns a response. In the example data flow below, the Query processes a message (here represented by "Employment") received from a source (an adapter instance), and returns the response to a target (again, an adapter instance). 268 2011-06-09
- 269. Embedded Data Flows Embedded Data Flows The software provides an easy-to-use option to create embedded data flows. 12.1 Overview of embedded data flows An embedded data flow is a data flow that is called from inside another data flow. Data passes into or out of the embedded data flow from the parent flow through a single source or target. The embedded data flow can contain any number of sources or targets, but only one input or one output can pass data to or from the parent data flow. You can create the following types of embedded data flows: Type Use when you want to... One input Add an embedded data flow at the end of a data flow One output Add an embedded data flow at the beginning of a data flow No input or output Replicate an existing data flow. An embedded data flow is a design aid that has no effect on job execution. When the software executes the parent data flow, it expands any embedded data flows, optimizes the parent data flow, then executes it. Use embedded data flows to: • • Reuse data flow logic. Save logical sections of a data flow so you can use the exact logic in other data flows, or provide an easy way to replicate the logic and modify it for other flows. • 269 Simplify data flow display. Group sections of a data flow in embedded data flows to allow clearer layout and documentation. Debug data flow logic. Replicate sections of a data flow as embedded data flows so you can execute them independently. 2011-06-09
- 270. Embedded Data Flows 12.2 Example of when to use embedded data flows In this example, a data flow uses a single source to load three different target systems. The Case transform sends each row from the source to different transforms that process it to get a unique target output. You can simplify the parent data flow by using embedded data flows for the three different cases. 12.3 Creating embedded data flows There are two ways to create embedded data flows. • Select objects within a data flow, right-click, and select Make Embedded Data Flow. • Drag a complete and fully validated data flow from the object library into an open data flow in the workspace. Then: • 270 Open the data flow you just added. 2011-06-09
- 271. Embedded Data Flows • Right-click one object you want to use as an input or as an output port and select Make Port for that object. The software marks the object you select as the connection point for this embedded data flow. Note: You can specify only one port, which means that the embedded data flow can appear only at the beginning or at the end of the parent data flow. 12.3.1 Using the Make Embedded Data Flow option 12.3.1.1 To create an embedded data flow 1. Select objects from an open data flow using one of the following methods: • Click the white space and drag the rectangle around the objects • CTRL-click each object Ensure that the set of objects you select are: • All connected to each other • Connected to other objects according to the type of embedded data flow you want to create such as one input, one output, or no input or output 2. Right-click and select Make Embedded Data Flow. The Create Embedded Data Flow window opens, with the embedded data flow connected to the parent by one input object. 3. Name the embedded data flow using the convention EDF_EDFName for example EDF_ERP. If you deselect the Replace objects in original data flow box, the software will not make a change in the original data flow. The software saves the new embedded data flow object to the repository and displays it in the object library under the Data Flows tab. You can use an embedded data flow created without replacement as a stand-alone data flow for troubleshooting. If Replace objects in original data flow is selected, the original data flow becomes a parent data flow, which has a call to the new embedded data flow. 4. Click OK. 271 2011-06-09
- 272. Embedded Data Flows The embedded data flow appears in the new parent data flow. 5. Click the name of the embedded data flow to open it. 6. Notice that the software created a new object, EDF_ERP_Input, which is the input port that connects this embedded data flow to the parent data flow. When you use the Make Embedded Data flow option, the software automatically creates an input or output object based on the object that is connected to the embedded data flow when it is created. For example, if an embedded data flow has an output connection, the embedded data flow will include a target XML file object labeled EDFName_Output. The naming conventions for each embedded data flow type are: Type Naming Conventions One input EDFName_Input One output EDFName_Output No input or output The software creates an embedded data flow without an input or output object 12.3.2 Creating embedded data flows from existing flows 272 2011-06-09
- 273. Embedded Data Flows To call an existing data flow from inside another data flow, put the data flow inside the parent data flow, then mark which source or target to use to pass data between the parent and the embedded data flows. 12.3.2.1 To create an embedded data flow out of an existing data flow 1. Drag an existing valid data flow from the object library into a data flow that is open in the workspace. 2. Consider renaming the flow using the EDF_EDFName naming convention. The embedded data flow appears without any arrowheads (ports) in the workspace. 3. Open the embedded data flow. 4. Right-click a source or target object (file or table) and select Make Port. Note: Ensure that you specify only one input or output port. Like a normal data flow, different types of embedded data flow ports are indicated by directional markings on the embedded data flow icon. 12.3.3 Using embedded data flows When you create and configure an embedded data flow using the Make Embedded Data Flow option, the software creates new input or output XML file and saves the schema in the repository as an XML Schema. You can reuse an embedded data flow by dragging it from the Data Flow tab of the object library into other data flows. To save mapping time, you might want to use the Update Schema option or the Match Schema option. The following example scenario uses both options: • • Select objects in data flow 1, and create embedded data flow 1 so that parent data flow 1 calls embedded data flow 1. • Create data flow 2 and data flow 3 and add embedded data flow 1 to both of them. • Go back to data flow 1. Change the schema of the object preceding embedded data flow 1 and use the Update Schema option with embedded data flow 1. It updates the schema of embedded data flow 1 in the repository. • 273 Create data flow 1. Now the schemas in data flow 2 and data flow 3 that are feeding into embedded data flow 1 will be different from the schema the embedded data flow expects. 2011-06-09
- 274. Embedded Data Flows • Use the Match Schema option for embedded data flow 1 in both data flow 2 and data flow 3 to resolve the mismatches at runtime. The Match Schema option only affects settings in the current data flow. The following sections describe the use of the Update Schema and Match Schema options in more detail. 12.3.3.1 Updating Schemas The software provides an option to update an input schema of an embedded data flow. This option updates the schema of an embedded data flow's input object with the schema of the preceding object in the parent data flow. All occurrences of the embedded data flow update when you use this option. 12.3.3.1.1 To update a schema 1. Open the embedded data flow's parent data flow. 2. Right-click the embedded data flow object and select Update Schema. 12.3.3.2 Matching data between parent and embedded data flow The schema of an embedded data flow's input object can match the schema of the preceding object in the parent data flow by name or position. A match by position is the default. 12.3.3.2.1 To specify how schemas should be matched 1. Open the embedded data flow's parent data flow. 2. Right-click the embedded data flow object and select Match SchemaBy Name or Match SchemaBy Position. The Match Schema option only affects settings for the current data flow. Data Services also allows the schema of the preceding object in the parent data flow to have more or fewer columns than the embedded data flow. The embedded data flow ignores additional columns and reads missing columns as NULL. Columns in both schemas must have identical or convertible data types. See the section on "Type conversion" in the Reference Guide for more information. 274 2011-06-09
- 275. Embedded Data Flows 12.3.3.3 Deleting embedded data flow objects You can delete embedded data flow ports, or remove entire embedded data flows. 12.3.3.3.1 To remove a port Right-click the input or output object within the embedded data flow and deselect Make Port. Data Services removes the connection to the parent object. Note: You cannot remove a port simply by deleting the connection in the parent flow. 12.3.3.3.2 To remove an embedded data flow Select it from the open parent data flow and choose Delete from the right-click menu or edit menu. If you delete embedded data flows from the object library, the embedded data flow icon appears with a red circle-slash flag in the parent data flow. Delete these defunct embedded data flow objects from the parent data flows. 12.3.4 Separately testing an embedded data flow Embedded data flows can be tested by running them separately as regular data flows. 1. Specify an XML file for the input port or output port. When you use the Make Embedded Data Flow option, an input or output XML file object is created and then (optional) connected to the preceding or succeeding object in the parent data flow. To test the XML file without a parent data flow, click the name of the XML file to open its source or target editor to specify a file name. 2. Put the embedded data flow into a job. 3. Run the job. You can also use the following features to test embedded data flows: • • 275 View Data to sample data passed into an embedded data flow. Auditing statistics about the data read from sources, transformed, and loaded into targets, and rules about the audit statistics to verify the expected data is processed. 2011-06-09
- 276. Embedded Data Flows Related Topics • Reference Guide: XML file • Design and Debug 12.3.5 Troubleshooting embedded data flows The following situations produce errors: • Both an input port and output port are specified in an embedded data flow. • Trapped defunct data flows. • Deleted connection to the parent data flow while the Make Port option, in the embedded data flow, remains selected. • Transforms with splitters (such as the Case transform) specified as the output port object because a splitter produces multiple outputs, and embedded data flows can only have one. • Variables and parameters declared in the embedded data flow that are not also declared in the parent data flow. • Embedding the same data flow at any level within itself. You can however have unlimited embedding levels. For example, DF1 data flow calls EDF1 embedded data flow which calls EDF2. Related Topics • To remove an embedded data flow • To remove a port 276 2011-06-09
- 277. Variables and Parameters Variables and Parameters This section contains information about the following: • Adding and defining local and global variables for jobs • Using environment variables • Using substitution parameters and configurations 13.1 Overview of variables and parameters You can increase the flexibility and reusability of work flows and data flows by using local and global variables when you design your jobs. Variables are symbolic placeholders for values. The data type of a variable can be any supported by the software such as an integer, decimal, date, or text string. You can use variables in expressions to facilitate decision-making or data manipulation (using arithmetic or character substitution). For example, a variable can be used in a LOOP or IF statement to check a variable's value to decide which step to perform: If $amount_owed > 0 print('$invoice.doc'); If you define variables in a job or work flow, the software typically uses them in a script, catch, or conditional process. 277 2011-06-09
- 278. Variables and Parameters You can use variables inside data flows. For example, use them in a custom function or in the WHERE clause of a query transform. In the software, local variables are restricted to the object in which they are created (job or work flow). You must use parameters to pass local variables to child objects (work flows and data flows). Global variables are restricted to the job in which they are created; however, they do not require parameters to be passed to work flows and data flows. Note: If you have workflows that are running in parallel, the global variables are not assigned. Parameters are expressions that pass to a work flow or data flow when they are called in a job. You create local variables, parameters, and global variables using the Variables and Parameters window in the Designer. You can set values for local or global variables in script objects. You can also set global variable values using external job, execution, or schedule properties. Using global variables provides you with maximum flexibility. For example, during production you can change values for default global variables at runtime from a job's schedule or “SOAP” call without having to open a job in the Designer. Variables can be used as file names for: • Flat file sources and targets • XML file sources and targets • XML message targets (executed in the Designer in test mode) • IDoc file sources and targets (in an SAP application environment) • IDoc message sources and targets (SAP application environment) Related Topics • Management Console Guide: Administrator, Support for Web Services 13.2 The Variables and Parameters window The software displays the variables and parameters defined for an object in the "Variables and Parameters" window. 13.2.1 To view the variables and parameters in each job, work flow, or data flow 278 2011-06-09
- 279. Variables and Parameters 1. In the Tools menu, select Variables. The "Variables and Parameters" window opens. 2. From the object library, double-click an object, or from the project area click an object to open it in the workspace. The Context box in the window changes to show the object you are viewing. If there is no object selected, the window does not indicate a context. The Variables and Parameters window contains two tabs. The Definitions tab allows you to create and view variables (name and data type) and parameters (name, data type, and parameter type) for an object type. Local variable and parameters can only be set at the work flow and data flow level. Global variables can only be set at the job level. The following table lists what type of variables and parameters you can create using the Variables and Parameters window when you select different objects. Object Type What you can create for the object Used by Local variables A script or conditional in the job Global variables Any object in the job Local variables This work flow or passed down to other work flows or data flows using a parameter. Job Work flow Parameters Data flow Parameters Parent objects to pass local variables. Work flows may also return variables or parameters to parent objects. A WHERE clause, column mapping, or a function in the data flow. Data flows cannot return output values. The Calls tab allows you to view the name of each parameter defined for all objects in a parent object's definition. You can also enter values for each parameter. For the input parameter type, values in the Calls tab can be constants, variables, or another parameter. For the output or input/output parameter type, values in the Calls tab can be variables or parameters. Values in the Calls tab must also use: • • 279 The same data type as the variable if they are placed inside an input or input/output parameter type, and a compatible data type if they are placed inside an output parameter type. Scripting language rules and syntax 2011-06-09
- 280. Variables and Parameters The following illustration shows the relationship between an open work flow called DeltaFacts, the Context box in the Variables and Parameters window, and the content in the Definition and Calls tabs. 13.3 Using local variables and parameters To pass a local variable to another object, define the local variable, then from the calling object, create a parameter and map the parameter to the local variable by entering a parameter value. For example, to use a local variable inside a data flow, define the variable in a parent work flow and then pass the value of the variable as a parameter of the data flow. 280 2011-06-09
- 281. Variables and Parameters 13.3.1 Parameters Parameters can be defined to: • Pass their values into and out of work flows • Pass their values into data flows Each parameter is assigned a type: input, output, or input/output. The value passed by the parameter can be used by any object called by the work flow or data flow. Note: You can also create local variables and parameters for use in custom functions. Related Topics • Reference Guide: Custom functions 13.3.2 Passing values into data flows You can use a value passed as a parameter into a data flow to control the data transformed in the data flow. For example, the data flow DF_PartFlow processes daily inventory values. It can process all of the part numbers in use or a range of part numbers based on external requirements such as the range of numbers processed most recently. If the work flow that calls DF_PartFlow records the range of numbers processed, it can pass the end value of the range $EndRange as a parameter to the data flow to indicate the start value of the range to process next. The software can calculate a new end value based on a stored number of parts to process each time, such as $SizeOfSet, and pass that value to the data flow as the end value. A query transform in the data flow uses the parameters passed in to filter the part numbers extracted from the source. 281 2011-06-09
- 282. Variables and Parameters The data flow could be used by multiple calls contained in one or more work flows to perform the same task on different part number ranges by specifying different parameters for the particular calls. 13.3.3 To define a local variable 1. Click the name of the job or work flow in the project area or workspace, or double-click one from the object library. 2. Click Tools > Variables. The "Variables and Parameters" window appears. 3. From the Definitions tab, select Variables. 4. Right-click and select Insert. A new variable appears (for example, $NewVariable0). A focus box appears around the name cell and the cursor shape changes to an arrow with a yellow pencil. 5. To edit the name of the new variable, click the name cell. The name can include alphanumeric characters or underscores (_), but cannot contain blank spaces. Always begin the name with a dollar sign ($). 6. Click the data type cell for the new variable and select the appropriate data type from the drop-down list. 7. Close the "Variables and Parameters" window. 13.3.4 Defining parameters There are two steps for setting up a parameter for a work flow or data flow: • • 282 Add the parameter definition to the flow. Set the value of the parameter in the flow call. 2011-06-09
- 283. Variables and Parameters 13.3.4.1 To add the parameter to the flow definition 1. Click the name of the work flow or data flow. 2. Click Tools > Variables. The "Variables and Parameters" window appears. 3. Go to the Definition tab. 4. Select Parameters. 5. Right-click and select Insert. A new parameter appears (for example, $NewParameter0). A focus box appears and the cursor shape changes to an arrow with a yellow pencil. 6. To edit the name of the new variable, click the name cell. The name can include alphanumeric characters or underscores (_), but cannot contain blank spaces. Always begin the name with a dollar sign ($). 7. Click the data type cell for the new parameter and select the appropriate data type from the drop-down list. If the parameter is an input or input/output parameter, it must have the same data type as the variable; if the parameter is an output parameter type, it must have a compatible data type. 8. Click the parameter type cell and select the parameter type (input, output, or input/output). 9. Close the "Variables and Parameters" window. 13.3.4.2 To set the value of the parameter in the flow call 1. Open the calling job, work flow, or data flow. 2. Click Tools > Variables to open the "Variables and Parameters" window. 3. Select the Calls tab. The Calls tab shows all the objects that are called from the open job, work flow, or data flow. 4. Click the Argument Value cell. A focus box appears and the cursor shape changes to an arrow with a yellow pencil. 5. Enter the expression the parameter will pass in the cell. If the parameter type is input, then its value can be an expression that contains a constant (for example, 0, 3, or 'string1'), a variable, or another parameter (for example, $startID or $parm1). 283 2011-06-09
- 284. Variables and Parameters If the parameter type is output or input/output, then the value must be a variable or parameter. The value cannot be a constant because, by definition, the value of an output or input/output parameter can be modified by any object within the flow. To indicate special values, use the following syntax: Value type Special syntax Variable $variable_name String 'string ' 13.4 Using global variables Global variables are global within a job. Setting parameters is not necessary when you use global variables. However, once you use a name for a global variable in a job, that name becomes reserved for the job. Global variables are exclusive within the context of the job in which they are created. 13.4.1 Creating global variables Define variables in the Variables and Parameter window. 13.4.1.1 To create a global variable 1. Click the name of a job in the project area or double-click a job from the object library. 2. Click Tools > Variables. The "Variables and Parameters" window appears. 3. From the Definitions tab, select Global Variables. 4. Right-click Global Variables and select Insert. A new global variable appears (for example, $NewJobGlobalVariable0). A focus box appears and the cursor shape changes to an arrow with a yellow pencil. 284 2011-06-09
- 285. Variables and Parameters 5. To edit the name of the new variable, click the name cell. The name can include alphanumeric characters or underscores (_), but cannot contain blank spaces. Always begin the name with a dollar sign ($). 6. Click the data type cell for the new variable and select the appropriate data type from the drop-down list. 7. Close the "Variables and Parameters" window. 13.4.2 Viewing global variables Global variables, defined in a job, are visible to those objects relative to that job. A global variable defined in one job is not available for modification or viewing from another job. You can view global variables from the Variables and Parameters window (with an open job in the work space) or from the Properties dialog of a selected job. 13.4.2.1 To view global variables in a job from the Properties dialog 1. In the object library, select the Jobs tab. 2. Right-click the job whose global variables you want to view and select Properties. 3. Click the Global Variable tab. Global variables appear on this tab. 13.4.3 Setting global variable values In addition to setting a variable inside a job using an initialization script, you can set and maintain global variable values outside a job. Values set outside a job are processed the same way as those set in an initialization script. However, if you set a value for the same variable both inside and outside a job, the internal value will override the external job value. Values for global variables can be set outside a job: • As a job property • As an execution or schedule property Global variables without defined values are also allowed. They are read as NULL. 285 2011-06-09
- 286. Variables and Parameters All values defined as job properties are shown in the Properties and the Execution Properties dialogs of the Designer and in the Execution Options and Schedule pages of the Administrator. By setting values outside a job, you can rely on these dialogs for viewing values set for global variables and easily edit values when testing or scheduling a job. Note: You cannot pass global variables as command line arguments for real-time jobs. 13.4.3.1 To set a global variable value as a job property 1. Right-click a job in the object library or project area. 2. Click Properties. 3. Click the Global Variable tab. All global variables created in the job appear. 4. Enter values for the global variables in this job. You can use any statement used in a script with this option. 5. Click OK. The software saves values in the repository as job properties. You can also view and edit these default values in the Execution Properties dialog of the Designer and in the Execution Options and Schedule pages of the Administrator. This allows you to override job property values at run-time. Related Topics • Reference Guide: Scripting Language 13.4.3.2 To set a global variable value as an execution property 1. Execute a job from the Designer, or execute or schedule a batch job from the Administrator. Note: For testing purposes, you can execute real-time jobs from the Designer in test mode. Make sure to set the execution properties for a real-time job. 2. View the global variables in the job and their default values (if available). 3. Edit values for global variables as desired. 4. If you are using the Designer, click OK. If you are using the Administrator, click Execute or Schedule. 286 2011-06-09
- 287. Variables and Parameters The job runs using the values you enter. Values entered as execution properties are not saved. Values entered as schedule properties are saved but can only be accessed from within the Administrator. 13.4.3.3 Automatic ranking of global variable values in a job Using the methods described in the previous section, if you enter different values for a single global variable, the software selects the highest ranking value for use in the job. A value entered as a job property has the lowest rank. A value defined inside a job has the highest rank. • If you set a global variable value as both a job and an execution property, the execution property value overrides the job property value and becomes the default value for the current job run. You cannot save execution property global variable values. For example, assume that a job, JOB_Test1, has three global variables declared: $YEAR, $MONTH, and $DAY. Variable $YEAR is set as a job property with a value of 2003. For your the job run, you set variables $MONTH and $DAY as execution properties to values 'JANUARY' and 31 respectively. The software executes a list of statements which includes default values for JOB_Test1: $YEAR=2003; $MONTH='JANUARY'; $DAY=31; For the second job run, if you set variables $YEAR and $MONTH as execution properties to values 2002 and 'JANUARY' respectively, then the statement $YEAR=2002 will replace $YEAR=2003. The software executes the following list of statements: $YEAR=2002; $MONTH='JANUARY'; Note: In this scenario, $DAY is not defined and the software reads it as NULL. You set $DAY to 31 during the first job run; however, execution properties for global variable values are not saved. • If you set a global variable value for both a job property and a schedule property, the schedule property value overrides the job property value and becomes the external, default value for the current job run. The software saves schedule property values in the repository. However, these values are only associated with a job schedule, not the job itself. Consequently, these values are viewed and edited from within the Administrator. • A global variable value defined inside a job always overrides any external values. However, the override does not occur until the software attempts to apply the external values to the job being processed with the internal value. Up until that point, the software processes execution, schedule, or job property values as default values. For example, suppose you have a job called JOB_Test2 that has three work flows, each containing a data flow. The second data flow is inside a work flow that is preceded by a script in which $MONTH 287 2011-06-09
- 288. Variables and Parameters is defined as 'MAY'. The first and third data flows have the same global variable with no value defined. The execution property $MONTH = 'APRIL' is the global variable value. In this scenario, 'APRIL' becomes the default value for the job. 'APRIL' remains the value for the global variable until it encounters the other value for the same variable in the second work flow. Since the value in the script is inside the job, 'MAY' overrides 'APRIL' for the variable $MONTH. The software continues the processing the job with this new value. 13.4.3.4 Advantages to setting values outside a job While you can set values inside jobs, there are advantages to defining values for global variables outside a job. For example, values defined as job properties are shown in the Properties and the Execution Properties dialogs of the Designer and in the Execution Options and Schedule pages of the Administrator. By setting values outside a job, you can rely on these dialogs for viewing all global variables and their values. You can also easily edit them for testing and scheduling. In the Administrator, you can set global variable values when creating or editing a schedule without opening the Designer. For example, use global variables as file names and start and end dates. 288 2011-06-09
- 289. Variables and Parameters 13.5 Local and global variable rules When defining local or global variables, consider rules for: • • • Naming Replicating jobs and work flows Importing and exporting 13.5.1 Naming • Local and global variables must have unique names within their job context. • Any name modification to a global variable can only be performed at the job level. 13.5.2 Replicating jobs and work flows • When you replicate all objects, the local and global variables defined in that job context are also replicated. • When you replicate a data flow or work flow, all parameters and local and global variables are also replicated. However, you must validate these local and global variables within the job context in which they were created. If you attempt to validate a data flow or work flow containing global variables without a job, Data Services reports an error. 13.5.3 Importing and exporting • • 289 When you export a job object, you also export all local and global variables defined for that job. When you export a lower-level object (such as a data flow) without the parent job, the global variable is not exported. Only the call to that global variable is exported. If you use this object in another job without defining the global variable in the new job, a validation error will occur. 2011-06-09
- 290. Variables and Parameters 13.6 Environment variables You can use system-environment variables inside jobs, work flows, or data flows. The get_env, set_env, and is_set_env functions provide access to underlying operating system variables that behave as the operating system allows. You can temporarily set the value of an environment variable inside a job, work flow or data flow. Once set, the value is visible to all objects in that job. Use the get_env, set_env, and is_set_env functions to set, retrieve, and test the values of environment variables. 13.7 Setting file names at run-time using variables You can set file names at runtime by specifying a variable as the file name. Variables can be used as file names for: • The following sources and targets: • Flat files • XML files and messages • IDoc files and messages (in an SAP environment) • The lookup_ext function (for a flat file used as a lookup table parameter) 13.7.1 To use a variable in a flat file name 1. Create a local or global variable using the Variables and Parameters window. 2. Create a script to set the value of a local or global variable, or call a system environment variable. 3. Declare the variable in the file format editor or in the Function editor as a lookup_ext parameter. • When you set a variable value for a flat file, specify both the file name and the directory name. Enter the variable in the File(s) property under Data File(s) in the File Format Editor. You cannot enter a variable in the Root directory property. • For lookups, substitute the path and file name in the Lookup table box in the lookup_ext function editor with the variable name. The following figure shows how you can set values for variables in flat file sources and targets in a script. 290 2011-06-09
- 291. Variables and Parameters When you use variables as sources and targets, you can also use multiple file names and wild cards. Neither is supported when using variables in the lookup_ext function. The figure above provides an example of how to use multiple variable names and wild cards. Notice that the $FILEINPUT variable includes two file names (separated by a comma). The two names (KNA1comma.* and KNA1c?mma.in) also make use of the wild cards (* and ?) supported by the software. Related Topics • Reference Guide: lookup_ext • Reference Guide: Data Services Scripting Language 13.8 Substitution parameters 13.8.1 Overview of substitution parameters Substitution parameters are useful when you want to export and run a job containing constant values in a specific environment. For example, if you create a job that references a unique directory on your 291 2011-06-09
- 292. Variables and Parameters local computer and you export that job to another computer, the job will look for the unique directory in the new environment. If that directory doesn’t exist, the job won’t run. Instead, by using a substitution parameter, you can easily assign a value for the original, constant value in order to run the job in the new environment. After creating a substitution parameter value for the directory in your environment, you can run the job in a different environment and all the objects that reference the original directory will automatically use the value. This means that you only need to change the constant value (the original directory name) in one place (the substitution parameter) and its value will automatically propagate to all objects in the job when it runs in the new environment. You can configure a group of substitution parameters for a particular run-time environment by associating their constant values under a substitution parameter configuration. 13.8.1.1 Substitution parameters versus global variables Substitution parameters differ from global variables in that they apply at the repository level. Global variables apply only to the job in which they are defined. You would use a global variable when you do not know the value prior to execution and it needs to be calculated in the job. You would use a substitution parameter for constants that do not change during execution. A substitution parameter defined in a given local repository is available to all the jobs in that repository. Therefore, using a substitution parameter means you do not need to define a global variable in each job to parameterize a constant value. The following table describes the main differences between global variables and substitution parameters. Global variables Substitution parameters Defined at the job level Defined at the repository level Cannot be shared across jobs Available to all jobs in a repository Data-type specific No data type (all strings) Value can change during job execution Fixed value set prior to execution of job (constants) However, you can use substitution parameters in all places where global variables are supported, for example: • • • • • • • 292 Query transform WHERE clauses Mappings SQL transform SQL statement identifiers Flat-file options User-defined transforms Address cleanse transform options Matching thresholds 2011-06-09
- 293. Variables and Parameters 13.8.1.2 Using substitution parameters You can use substitution parameters in expressions, SQL statements, option fields, and constant strings. For example, many options and expression editors include a drop-down menu that displays a list of all the available substitution parameters. The software installs some default substitution parameters that are used by some Data Quality transforms. For example, the USA Regulatory Address Cleanse transform uses the following built-in substitution parameters: • • $$RefFilesAddressCleanse defines the location of the address cleanse directories. $$ReportsAddressCleanse (set to Yes or No) enables data collection for creating reports with address cleanse statistics. This substitution parameter provides one location where you can enable or disable that option for all jobs in the repository. Other examples of where you can use substitution parameters include: • In a script, for example: Print('Data read in : [$$FilePath]'); or Print('[$$FilePath]'); • In a file format, for example with [$$FilePath]/file.txt as the file name 13.8.2 Using the Substitution Parameter Editor Open the Substitution Parameter Editor from the Designer by selecting Tools > Tools Substitution Parameter Configurations. Use the Substitution Parameter editor to do the following tasks: • • • • • • • 293 Add and define a substitution parameter by adding a new row in the editor. For each substitution parameter, use right-click menus and keyboard shortcuts to Cut, Copy, Paste, Delete, and Insert parameters. Change the order of substitution parameters by dragging rows or using the Cut, Copy, Paste, and Insert commands. Add a substitution parameter configuration by clicking the Create New Substitution Parameter Configuration icon in the toolbar. Duplicate an existing substitution parameter configuration by clicking the Create Duplicate Substitution Parameter Configuration icon. Rename a substitution parameter configuration by clicking the Rename Substitution Parameter Configuration icon. Delete a substitution parameter configuration by clicking the Delete Substitution Parameter Configuration icon. 2011-06-09
- 294. Variables and Parameters • • • Reorder the display of configurations by clicking the Sort Configuration Names in Ascending Order and Sort Configuration Names in Descending Order icons. Move the default configuration so it displays next to the list of substitution parameters by clicking the Move Default Configuration To Front icon. Change the default configuration. Related Topics • Adding and defining substitution parameters 13.8.2.1 Naming substitution parameters When you name and define substitution parameters, use the following rules: • • The name prefix is two dollar signs $$ (global variables are prefixed with one dollar sign). When adding new substitution parameters in the Substitution Parameter Editor, the editor automatically adds the prefix. When typing names in the Substitution Parameter Editor, do not use punctuation (including quotes or brackets) except underscores. The following characters are not allowed: ,: / ' " = < > + | - * % ; t [ ] ( ) r n $ ] + • • • • • You can type names directly into fields, column mappings, transform options, and so on. However, you must enclose them in square brackets, for example [$$SamplesInstall]. Names can include any alpha or numeric character or underscores but cannot contain spaces. Names are not case sensitive. The maximum length of a name is 64 characters. Names must be unique within the repository. 13.8.2.2 Adding and defining substitution parameters 1. In the Designer, open the Substitution Parameter Editor by selecting Tools > Substitution Parameter Configurations. 2. The first column lists the substitution parameters available in the repository. To create a new one, double-click in a blank cell (a pencil icon will appear in the left) and type a name. The software automatically adds a double dollar-sign prefix ($$) to the name when you navigate away from the cell. 3. The second column identifies the name of the first configuration, by default Configuration1 (you can change configuration names by double-clicking in the cell and retyping the name). Double-click in the blank cell next to the substitution parameter name and type the constant value that the parameter represents in that configuration. The software applies that value when you run the job. 294 2011-06-09
- 295. Variables and Parameters 4. To add another configuration to define a second value for the substitution parameter, click the Create New Substitution Parameter Configuration icon on the toolbar. 5. Type a unique name for the new substitution parameter configuration. 6. Enter the value the substitution parameter will use for that configuration. You can now select from one of the two substitution parameter configurations you just created. To change the default configuration that will apply when you run jobs, select it from the drop-down list box at the bottom of the window. You can also export these substitution parameter configurations for use in other environments. Example: In the following example, the substitution parameter $$NetworkDir has the value D:/Data/Staging in the configuration named Windows_Subst_Param_Conf and the value /usr/data/staging in the UNIX_Subst_Param_Conf configuration. Notice that each configuration can contain multiple substitution parameters. Related Topics • Naming substitution parameters • Exporting and importing substitution parameters 13.8.3 Associating a substitution parameter configuration with a system configuration 295 2011-06-09
- 296. Variables and Parameters A system configuration groups together a set of datastore configurations and a substitution parameter configuration. A substitution parameter configuration can be associated with one or more system configurations. For example, you might create one system configuration for your local system and a different system configuration for another system. Depending on your environment, both system configurations might point to the same substitution parameter configuration or each system configuration might require a different substitution parameter configuration. At job execution time, you can set the system configuration and the job will execute with the values for the associated substitution parameter configuration. To associate a substitution parameter configuration with a new or existing system configuration: 1. In the Designer, open the System Configuration Editor by selecting Tools > System Configurations. 2. Optionally create a new system configuration. 3. Under the desired system configuration name, select a substitution parameter configuration to associate with the system configuration. 4. Click OK. Example: The following example shows two system configurations, Americas and Europe. In this case, there are substitution parameter configurations for each region (Europe_Subst_Parm_Conf and Americas_Subst_Parm_Conf). Each substitution parameter configuration defines where the data source files are located for that region, for example D:/Data/Americas and D:/Data/Europe. Select the appropriate substitution parameter configuration and datastore configurations for each system configuration. Related Topics • Defining a system configuration 296 2011-06-09
- 297. Variables and Parameters 13.8.4 Overriding a substitution parameter in the Administrator In the Administrator, you can override the substitution parameters, or select a system configuration to specify a substitution parameter configuration, on four pages: • Execute Batch Job • Schedule Batch Job • Export Execution Command • Real-Time Service Configuration For example, the Execute Batch Job page displays the name of the selected system configuration, the substitution parameter configuration, and the name of each substitution parameter and its value. To override a substitution parameter: 1. Select the appropriate system configuration. 2. Under Substitution Parameters, click Add Overridden Parameter, which displays the available substitution parameters. 3. From the drop-down list, select the substitution parameter to override. 4. In the second column, type the override value. Enter the value as a string without quotes (in contrast with Global Variables). 5. Execute the job. 13.8.5 Executing a job with substitution parameters To see the details of how substitution parameters are being used in the job during execution in the Designer trace log: 1. 2. 3. 4. Right-click the job name and click Properties. Click the Trace tab. For the Trace Assemblers option, set the value to Yes. Click OK. When you execute a job from the Designer, the Execution Properties window displays. You have the following options: • On the Execution Options tab from the System configuration drop-down menu, optionally select the system configuration with which you want to run the job. If you do not select a system configuration, the software applies the default substitution parameter configuration as defined in the Substitution Parameter Editor. You can click Browse to view the "Select System Configuration" window in order to see the substitution parameter configuration associated with each system configuration. The "Select System 297 2011-06-09
- 298. Variables and Parameters Configuration" is read-only. If you want to change a system configuration, click Tools > System Configurations. • You can override the value of specific substitution parameters at run time. Click the Substitution Parameter tab, select a substitution parameter from the Name column, and enter a value by double-clicking in the Value cell. To override substitution parameter values when you start a job via a Web service, see the Integrator's Guide. Related Topics • Associating a substitution parameter configuration with a system configuration • Overriding a substitution parameter in the Administrator 13.8.6 Exporting and importing substitution parameters Substitution parameters are stored in a local repository along with their configured values. The software does not include substitution parameters as part of a regular export.You can, however, export substitution parameters and configurations to other repositories by exporting them to a file and then importing the file to another repository. 13.8.6.1 Exporting substitution parameters 1. Right-click in the local object library and select Repository > Export Substitution Parameter Configurations. 2. Select the check box in the Export column for the substitution parameter configurations to export. 3. Save the file. The software saves it as a text file with an .atl extension. 13.8.6.2 Importing substitution parameters The substitution parameters must have first been exported to an ATL file. Be aware of the following behaviors when importing substitution parameters: • The software adds any new substitution parameters and configurations to the destination local repository. 298 2011-06-09
- 299. Variables and Parameters • If the repository has a substitution parameter with the same name as in the exported file, importing will overwrite the parameter's value. Similarly, if the repository has a substitution parameter configuration with the same name as the exported configuration, importing will overwrite all the parameter values for that configuration. 1. In the Designer, right-click in the object library and select Repository > Import from file. 2. Browse to the file to import. 3. Click OK. Related Topics • Exporting substitution parameters 299 2011-06-09
- 301. Executing Jobs Executing Jobs This section contains an overview of the software job execution, steps to execute jobs, debug errors, and change job server options. 14.1 Overview of job execution You can run jobs in three different ways. Depending on your needs, you can configure: • Immediate jobs The software initiates both batch and real-time jobs and runs them immediately from within the De signer. For these jobs, both the Designer and designated Job Server (where the job executes, usually many times on the same machine) must be running. You will most likely run immediate jobs only during the development cycle. • Scheduled jobs Batch jobs are scheduled. To schedule a job, use the Administrator or use a third-party scheduler. When jobs are scheduled by third-party software: • The job initiates outside of the software. • The job operates on a batch job (or shell script for UNIX) that has been exported from the software. When a job is invoked by a third-party scheduler: • • • The corresponding Job Server must be running. The Designer does not need to be running. Services Real-time jobs are set up as services that continuously listen for requests from an Access Server and process requests on-demand as they are received. Use the Administrator to create a service from a real-time job. 14.2 Preparing for job execution 301 2011-06-09
- 302. Executing Jobs 14.2.1 Validating jobs and job components You can also explicitly validate jobs and their components as you create them by: Clicking the Validate All button from the toolbar (or choosing ValidateAll Objects in View from the Debug menu). This command checks the syntax of the object definition for the active workspace and for all objects that are called from the active workspace view recursively. Clicking the Validate Current View button from the toolbar (or choosing ValidateCurrent View from the Debug menu). This command checks the syntax of the object definition for the active workspace. You can set the Designer options (Tools > Options > Designer > General) to validate jobs started in Designer before job execution. The default is not to validate. The software also validates jobs before exporting them. If during validation the software discovers an error in an object definition, it opens a dialog box indicating that an error exists, then opens the Output window to display the error. If there are errors, double-click the error in the Output window to open the editor of the object containing the error. If you are unable to read the complete error text in the window, you can access additional information by right-clicking the error listing and selecting View from the context menu. Error messages have these levels of severity: Severity Description Information Informative message only—does not prevent the job from running. No action is required. Warning 302 The error is not severe enough to stop job execution, but you might get unexpected results. For example, if the data type of a source column in a transform within a data flow does not match the data type of the target column in the transform, the software alerts you with a warning message. 2011-06-09
- 303. Executing Jobs Severity Description Error The error is severe enough to stop job execution. You must fix the error before the job will execute. 14.2.2 Ensuring that the Job Server is running Before you execute a job (either as an immediate or scheduled task), ensure that the Job Server is associated with the repository where the client is running. When the Designer starts, it displays the status of the Job Server for the repository to which you are connected. Icon Description Job Server is running Job Server is inactive The name of the active Job Server and port number appears in the status bar when the cursor is over the icon. 14.2.3 Setting job execution options Options for jobs include Debug and Trace. Although these are object options—they affect the function of the object—they are located in either the Property or the Execution window associated with the job. Execution options for jobs can either be set for a single instance or as a default value. • • 303 The right-click Execute menu sets the options for a single execution only and overrides the default settings The right-click Properties menu sets the default settings 2011-06-09
- 304. Executing Jobs 14.2.3.1 To set execution options for every execution of the job 1. From the Project area, right-click the job name and choose Properties. 2. Select options on the Properties window: Related Topics • Viewing and changing object properties • Reference Guide: Parameters • Reference Guide: Trace properties • Setting global variable values 14.3 Executing jobs as immediate tasks Immediate or "on demand" tasks are initiated from the Designer. Both the Designer and Job Server must be running for the job to execute. 14.3.1 To execute a job as an immediate task 1. In the project area, select the job name. 2. Right-click and choose Execute. The software prompts you to save any objects that have changes that have not been saved. 3. The next step depends on whether you selected the Perform complete validation before job execution check box in the Designer Options: • If you have not selected this check box, a window opens showing execution properties (debug and trace) for the job. Proceed to the next step. • If you have selected this check box, the software validates the job before it runs. You must correct any serious errors before the job will run. There might also be warning messages—for example, messages indicating that date values will be converted to datetime values. Correct them if you want (they will not prevent job execution) or click OK to continue. After the job validates, a window opens showing the execution properties (debug and trace) for the job. 4. Set the execution properties. 304 2011-06-09
- 305. Executing Jobs You can choose the Job Server that you want to process this job, datastore profiles for sources and targets if applicable, enable automatic recovery, override the default trace properties, or select global variables at runtime. For more information, see: Note: Setting execution properties here affects a temporary change for the current execution only. 5. Click OK. As the software begins execution, the execution window opens with the trace log button active. Use the buttons at the top of the log window to display the trace log, monitor log, and error log (if there are any errors). After the job is complete, use an RDBMS query tool to check the contents of the target table or file. Related Topics • Designer — General • Reference Guide: Parameters • Reference Guide: Trace properties • Setting global variable values • Debugging execution errors • Examining target data 14.3.2 Monitor tab The Monitor tab lists the trace logs of all current or most recent executions of a job. The traffic-light icons in the Monitor tab have the following meanings: • A green light indicates that the job is running You can right-click and select Kill Job to stop a job that is still running. • A red light indicates that the job has stopped You can right-click and select Properties to add a description for a specific trace log. This description is saved with the log which can be accessed later from the Log tab. • 305 A red cross indicates that the job encountered an error 2011-06-09
- 306. Executing Jobs 14.3.3 Log tab You can also select the Log tab to view a job's trace log history. Click a trace log to open it in the workspace. Use the trace, monitor, and error log icons (left to right at the top of the job execution window in the workspace) to view each type of available log for the date and time that the job was run. 14.4 Debugging execution errors The following tables lists tools that can help you understand execution errors: Tool Definition Trace log Itemizes the steps executed in the job and the time execution began and ended. Monitor log Displays each step of each data flow in the job, the number of rows streamed through each step, and the duration of each step. Error log Displays the name of the object being executed when an error occurred and the text of the resulting error message. If the job ran against SAP data, some of the ABAP errors are also available in the error log. Target data Always examine your target data to see if your job produced the results you expected. Related Topics • Using logs • Examining trace logs • Examining monitor logs • Examining error logs • Examining target data 306 2011-06-09
- 307. Executing Jobs 14.4.1 Using logs This section describes how to use logs in the Designer. • To open the trace log on job execution, select Tools > Options > Designer > General > Open monitor on job execution. • To copy log content from an open log, select one or multiple lines and use the key commands [Ctrl+C]. 14.4.1.1 To access a log during job execution If your Designer is running when job execution begins, the execution window opens automatically, displaying the trace log information. Use the monitor and error log icons (middle and right icons at the top of the execution window) to view these logs. The execution window stays open until you close it. 14.4.1.2 To access a log after the execution window has been closed 1. In the project area, click the Log tab. 2. Click a job name to view all trace, monitor, and error log files in the workspace. Or expand the job you are interested in to view the list of trace log files and click one. Log indicators signify the following: Job Log Indicator N_ Description Indicates that the job executed successfully on this explicitly selected Job Server. Indicates that the was job executed successfully by a server group. The Job Server listed executed the job. 307 2011-06-09
- 308. Executing Jobs Job Log Indicator Description Indicates that the job encountered an error on this explicitly selected Job Server. Indicates that the job encountered an error while being executed by a server group. The Job Server listed executed the job. 3. Click the log icon for the execution of the job you are interested in. (Identify the execution from the position in sequence or datetime stamp.) 4. Use the list box to switch between log types or to view No logs or All logs. 14.4.1.3 To delete a log You can set how long to keep logs in Administrator. If want to delete logs from the Designer manually: 1. In the project area, click the Log tab. 2. Right-click the log you want to delete and select Delete Log. Related Topics • Administrator Guide: Setting the log retention period 14.4.1.4 Examining trace logs Use the trace logs to determine where an execution failed, whether the execution steps occur in the order you expect, and which parts of the execution are the most time consuming. 14.4.1.5 Examining monitor logs The monitor log quantifies the activities of the components of the job. It lists the time spent in a given component of a job and the number of data rows that streamed through the component. The following screen shows an example of a monitor log. 308 2011-06-09
- 309. Executing Jobs 14.4.1.6 Examining error logs The software produces an error log for every job execution. Use the error logs to determine how an execution failed. If the execution completed without error, the error log is blank. 14.4.2 Examining target data The best measure of the success of a job is the state of the target data. Always examine your data to make sure the data movement operation produced the results you expect. Be sure that: • Data was not converted to incompatible types or truncated. • Data was not duplicated in the target. • Data was not lost between updates of the target. • Generated keys have been properly incremented. • Updated values were handled properly. 14.5 Changing Job Server options Familiarize yourself with the more technical aspects of how the software handles data (using the Reference Guide) and some of its interfaces like those for adapters and SAP application. There are many options available in the software for troubleshooting and tuning a job. 309 2011-06-09
- 310. Executing Jobs Option Option Description Default Value Adapter Data Exchange Time-out (For adapters) Defines the time a function call or outbound message will wait for the response from the adapter operation. 10800000 Adapter Start Time-out (For adapters) Defines the time that the Administrator or Designer will wait for a response from the Job Server that manages adapters (start/stop/status). 90000 (90 seconds) AL_JobServerLoadBal anceDebug Enables a Job Server to log server group information if the value is set to TRUE. Information is saved in: $LINK_DIR/log/<JobServerName>/serv er_eventlog.txt FALSE AL_JobServerLoad OSPolling Sets the polling interval (in seconds) that the software uses to get status information used to calculate the load balancing index. This index is used by server groups. 60 (3 hours) Displays the software's internal datastore CD_DS_d0cafae2 and its related jobs in the object library. The CD_DS_d0cafae2 datastore supports two internal jobs. The first calculates usage dependencies on repository tables and the second updates server group configurations. Display DI Internal Jobs If you change your repository password, user name, or other connection information, change the default value of this option to TRUE, close and reopen the Designer, then update the CD_DS_d0cafae2 datastore configuration to match your new repository configuration. This enables the calculate usage dependency job (CD_JOBd0cafae2) and the server group job (di_job_al_mach_info) to run without a connection error. FALSE FTP Number of Retry Sets the number of retries for an FTP connection that initially fails. 0 FTP Retry Interval Sets the FTP connection retry interval in milliseconds. 1000 310 2011-06-09
- 311. Executing Jobs Option Option Description Default Value Global_DOP Sets the Degree of Parallelism for all data flows run by a given Job Server. You can also set the Degree of parallelism for individual data flows from each data flow's Properties window. If a data flow's Degree of parallelism value is 0, then the Job Server will use the Global_DOP value. The Job Server will use the data flow's Degree of parallelism value if it is set to any value except zero because it overrides the Global_DOP value. 1 Ignore Reduced Msg Type (For SAP applications) Disables IDoc reduced message type processing for all message types if the value is set to TRUE. FALSE Ignore Reduced Msg Type_foo (For SAP application) Disables IDoc reduced message type processing for a specific message type (su ch as foo ) if the value is set to TRUE. FALSE OCI Server Attach Retry The engine calls the Oracle OCIServerAttach function each time it makes a connection to Oracle. If the engine calls this function too fast (processing parallel data flows for example), the function may fail. To correct this, increase the retry value to 5. 3 Splitter Optimization The software might hang if you create a job in which a file source feeds into two queries. If this option is set to TRUE, the engine internally creates two source files that feed the two queries instead of a splitter that feeds the two queries. FALSE Use Explicit Database Links Jobs with imported database links normally will show improved performance because the software uses these links to push down processing to a database. If you set this option to FALSE, all data flows will not use linked datastores. TRUE The use of linked datastores can also be disabled from any data flow properties dialog. The data flow level option takes precedence over this Job Server level option. 311 2011-06-09
- 312. Executing Jobs Option Option Description Default Value Use Domain Name Adds a domain name to a Job Server name in the repository. This creates a fully qualified server name and allows the Designer to locate a Job Server on a different domain. TRUE Related Topics • Performance Optimization Guide: Using parallel Execution, Degree of parallelism • Performance Optimization Guide: Maximizing Push-Down Operations, Database link support for push-down operations across datastores 14.5.1 To change option values for an individual Job Server 1. Select the Job Server you want to work with by making it your default Job Server. a. Select Tools > Options > Designer > Environment. b. Select a Job Server from the Default Job Server section. c. Click OK. 2. Select Tools > Options > Job Server > General. 312 2011-06-09
- 313. Executing Jobs 3. Enter the section and key you want to use from the following list of value pairs: Section Key int AdapterDataExchangeTimeout int AdapterStartTimeout AL_JobServer AL_JobServerLoadBalanceDebug AL_JobServer AL_JobServerLoadOSPolling string DisplayDIInternalJobs AL_Engine FTPNumberOfRetry AL_Engine FTPRetryInterval AL_Engine Global_DOP AL_Engine IgnoreReducedMsgType AL_Engine IgnoreReducedMsgType_foo AL_Engine OCIServerAttach_Retry AL_Engine SPLITTER_OPTIMIZATION AL_Engine Repository 313 UseExplicitDatabaseLinks UseDomainName 2011-06-09
- 314. Executing Jobs 4. Enter a value. For example, enter the following to change the default value for the number of times a Job Server will retry to make an FTP connection if it initially fails: Option Sample value Section AL_Engine Key FTPNumberOfRetry Value 2 These settings will change the default value for the FTPNumberOfRetry option from zero to two. 5. To save the settings and close the Options window, click OK. 6. Re-select a default Job Server by repeating step 1, as needed. 14.5.2 To use mapped drive names in a path The software supports only UNC (Universal Naming Convention) paths to directories. If you set up a path to a mapped drive, the software will convert that mapped drive to its UNC equivalent. To make sure that your mapped drive is not converted back to the UNC path, you need to add your drive names in the "Options "window in the Designer. 1. Choose Tools > Options. 2. In the "Options" window, expand Job Server and then select General. 3. In the Section edit box, enter MappedNetworkDrives. 4. In the Key edit box, enter LocalDrive1 to map to a local drive or RemoteDrive1 to map to a remote drive. 5. In the Value edit box, enter a drive letter, such as M: for a local drive or <ma chine_name><share_name> for a remote drive. 6. Click OK to close the window. If you want to add another mapped drive, you need to close the "Options" window and re-enter. Be sure that each entry in the Key edit box is a unique name. 314 2011-06-09
- 315. Data Assessment Data Assessment With operational systems frequently changing, data quality control becomes critical in your extract, transform and load (ETL) jobs. The Designer provides data quality controls that act as a firewall to identify and fix errors in your data. These features can help ensure that you have trusted information. The Designer provides the following features that you can use to determine and improve the quality and structure of your source data: • Use the Data Profiler to determine: • • The distribution, relationship, and structure of your source data to better design your jobs and data flows, as well as your target data warehouse. • • The quality of your source data before you extract it. The Data Profiler can identify anomalies in your source data to help you better define corrective actions in the Validation transform, data quality, or other transforms. The content of your source and target data so that you can verify that your data extraction job returns the results you expect. Use the View Data feature to: • • • View your source data before you execute a job to help you create higher quality job designs. Compare sample data from different steps of your job to verify that your data extraction job returns the results you expect. Use the Validation transform to: • • • Verify that your source data meets your business rules. Take appropriate actions when the data does not meet your business rules. Use the auditing data flow feature to: • Define rules that determine if a source, transform, or target object processes correct data. • Define the actions to take when an audit rule fails. • Use data quality transforms to improve the quality of your data. • Use Data Validation dashboards in the Metadata Reporting tool to evaluate the reliability of your target data based on the validation rules you created in your batch jobs. This feedback allows business users to quickly review, assess, and identify potential inconsistencies or errors in source data. Related Topics • Using the Data Profiler 315 2011-06-09
- 316. Data Assessment • Using View Data to determine data quality • Using the Validation transform • Using Auditing • Overview of data quality • Management Console Guide: Data Validation Dashboard Reports 15.1 Using the Data Profiler The Data Profiler executes on a profiler server to provide the following data profiler information that multiple users can view: • Column analysis—The Data Profiler provides two types of column profiles: • • • Basic profiling—This information includes minimum value, maximum value, average value, minimum string length, and maximum string length. Detailed profiling—Detailed column analysis includes distinct count, distinct percent, median, median string length, pattern count, and pattern percent. Relationship analysis—This information identifies data mismatches between any two columns for which you define a relationship, including columns that have an existing primary key and foreign key relationship. You can save two levels of data: • Save the data only in the columns that you select for the relationship. • Save the values in all columns in each row. 15.1.1 Data sources that you can profile You can execute the Data Profiler on data contained in the following sources. See the Release Notes for the complete list of sources that the Data Profiler supports. • Databases, which include: • • DB2 • Oracle • SQL Server • Sybase IQ • 316 Attunity Connector for mainframe databases Teradata 2011-06-09
- 317. Data Assessment • Applications, which include: • • JDE World • Oracle Applications • PeopleSoft • SAP Applications • • JDE One World Siebel Flat files 15.1.2 Connecting to the profiler server You must install and configure the profiler server before you can use the Data Profiler. The Designer must connect to the profiler server to run the Data Profiler and view the profiler results. You provide this connection information on the Profiler Server Login window. 1. Use one of the following methods to invoke the Profiler Server Login window: • From the tool bar menu, select Tools > Profiler Server Login. • On the bottom status bar, double-click the Profiler Server icon which is to the right of the Job Server icon. 2. Enter your user credentials for the CMS. • System Specify the server name and optionally the port for the CMS. • User name Specify the user name to use to log into CMS. • Password Specify the password to use to log into the CMS. • Authentication Specify the authentication type used by the CMS. 3. Click Log on. The software attempts to connect to the CMS using the specified information. When you log in successfully, the list of profiler repositories that are available to you is displayed. 4. Select the repository you want to use. 5. Click OK to connect using the selected repository. 317 2011-06-09
- 318. Data Assessment When you successfully connect to the profiler server, the Profiler Server icon on the bottom status bar no longer has the red X on it. In addition, when you move the pointer over this icon, the status bar displays the location of the profiler server. Related Topics • Management Console Guide: Profile Server Management • Management Console Guide: Defining profiler users 15.1.3 Profiler statistics 15.1.3.1 Column profile You can generate statistics for one or more columns. The columns can all belong to one data source or from multiple data sources. If you generate statistics for multiple sources in one profile task, all sources must be in the same datastore. Basic profiling By default, the Data Profiler generates the following basic profiler attributes for each column that you select. Basic Attribute Min Of all values, the lowest value in this column. Min count Number of rows that contain this lowest value in this column. Max Of all values, the highest value in this column. Max count Number of rows that contain this highest value in this column. Average For numeric columns, the average value in this column. Min string length For character columns, the length of the shortest string value in this column. Max string length For character columns, the length of the longest string value in this column. Average string length For character columns, the average length of the string values in this column. Nulls Number of NULL values in this column. Nulls % 318 Description Percentage of rows that contain a NULL value in this column. 2011-06-09
- 319. Data Assessment Basic Attribute Description Zeros Number of 0 values in this column. Zeros % Percentage of rows that contain a 0 value in this column. Blanks For character columns, the number of rows that contain a blank in this column. Blanks % Percentage of rows that contain a blank in this column. Detailed profiling You can generate more detailed attributes in addition to the above attributes, but detailed attributes generation consumes more time and computer resources. Therefore, it is recommended that you do not select the detailed profile unless you need the following attributes: Detailed Attribute Description Median The value that is in the middle row of the source table. Median string length For character columns, the value that is in the middle row of the source table. Distincts Number of distinct values in this column. Distinct % Percentage of rows that contain each distinct value in this column. Patterns Number of different patterns in this column. Pattern % Percentage of rows that contain each pattern in this column. Examples of using column profile statistics to improve data quality You can use the column profile attributes to assist you in different tasks, including the following tasks: • • Identify variations of the same content. For example, part number might be an integer data type in one data source and a varchar data type in another data source. You might then decide which data type you want to use in your target data warehouse. • Discover data patterns and formats. For example, the profile statistics might show that phone number has several different formats. With this profile information, you might decide to define a validation transform to convert them all to use the same target format. • 319 Obtain basic statistics, frequencies, ranges, and outliers. For example, these profile statistics might show that a column value is markedly higher than the other values in a data source. You might then decide to define a validation transform to set a flag in a different table when you load this outlier into the target table. Analyze the numeric range. For example, customer number might have one range of numbers in one source, and a different range in another source. Your target will need to have a data type that can accommodate the maximum range. 2011-06-09
- 320. Data Assessment • Identify missing information, nulls, and blanks in the source system. For example, the profile statistics might show that nulls occur for fax number. You might then decide to define a validation transform to replace the null value with a phrase such as "Unknown" in the target table. Related Topics • To view the column attributes generated by the Data Profiler • Submitting column profiler tasks 15.1.3.2 Relationship profile A relationship profile shows the percentage of non matching values in columns of two sources. The sources can be: • Tables • Flat files • A combination of a table and a flat file The key columns can have a primary key and foreign key relationship defined or they can be unrelated (as when one comes from a datastore and the other from a file format). You can choose between two levels of relationship profiles to save: • Save key columns data only By default, the Data Profiler saves the data only in the columns that you select for the relationship. Note: The Save key columns data only level is not available when using Oracle datastores. • Save all columns data You can save the values in the other columns in each row, but this processing will take longer and consume more computer resources to complete. When you view the relationship profile results, you can drill down to see the actual data that does not match. You can use the relationship profile to assist you in different tasks, including the following tasks: • • 320 Identify missing data in the source system. For example, one data source might include region, but another source might not. Identify redundant data across data sources. For example, duplicate names and addresses might exist between two sources or no name might exist for an address in one source. 2011-06-09
- 321. Data Assessment • Validate relationships across data sources. For example, two different problem tracking systems might include a subset of common customer-reported problems, but some problems only exist in one system or the other. Related Topics • Submitting relationship profiler tasks • Viewing the profiler results 15.1.4 Executing a profiler task The Data Profiler allows you to calculate profiler statistics for any set of columns you choose. Note: This optional feature is not available for columns with nested schemas, LONG or TEXT data type. You cannot execute a column profile task with a relationship profile task. 15.1.4.1 Submitting column profiler tasks 1. In the Object Library of the Designer, you can select either a table or flat file. For a table, go to the "Datastores" tab and select a table. If you want to profile all tables within a datastore, select the datastore name. To select a subset of tables in the "Ddatastore" tab, hold down the Ctrl key as you select each table. For a flat file, go to the "Formats" tab and select a file. 2. After you select your data source, you can generate column profile statistics in one of the following ways: • Right-click and select Submit Column Profile Request. Some of the profile statistics can take a long time to calculate. Select this method so the profile task runs asynchronously and you can perform other Designer tasks while the profile task executes. This method also allows you to profile multiple sources in one profile task. • 321 Right-click, select View Data, click the "Profile" tab, and click Update. This option submits a synchronous profile task, and you must wait for the task to complete before you can perform other tasks in the Designer. 2011-06-09
- 322. Data Assessment You might want to use this option if you are already in the "View Data" window and you notice that either the profile statistics have not yet been generated, or the date that the profile statistics were generated is older than you want. 3. (Optional) Edit the profiler task name. The Data Profiler generates a default name for each profiler task. You can edit the task name to create a more meaningful name, a unique name, or to remove dashes which are allowed in column names but not in task names. If you select a single source, the default name has the following format: username_t_sourcename If you select multiple sources, the default name has the following format: username_t_firstsourcename_lastsourcename Column Description username Name of the user that the software uses to access system services. t Type of profile. The value is C for column profile that obtains attributes (such as low value and high value) for each selected column. firstsourcename Name of first source in alphabetic order. lastsourcename Name of last source in alphabetic order if you select multiple sources. 4. If you select one source, the "Submit Column Profile Request" window lists the columns and data types. Keep the check in front of each column that you want to profile and remove the check in front of each column that you do not want to profile. Alternatively, you can click the check box at the top in front of Name to deselect all columns and then select the check boxes. 5. If you selected multiple sources, the "Submit Column Profiler Request" window lists the sources on the left. a. Select a data source to display its columns on the right side. b. On the right side of the "Submit Column Profile Request" window, keep the check in front of each column that you want to profile, and remove the check in front of each column that you do not want to profile. Alternatively, you can click the check box at the top in front of Name to deselect all columns and then select the individual check box for the columns you want to profile. c. Repeat steps 1 and 2 for each data source. 6. (Optional) Select Detailed profiling for a column. 322 2011-06-09
- 323. Data Assessment Note: The Data Profiler consumes a large amount of resources when it generates detailed profile statistics. Choose Detailed profiling only if you want these attributes: distinct count, distinct percent, median value, median string length, pattern, pattern count. If you choose Detailed profiling, ensure that you specify a pageable cache directory that contains enough disk space for the amount of data you profile. If you want detailed attributes for all columns in all sources listed, click "Detailed profiling" and select Apply to all columns of all sources. If you want to remove Detailed profiling for all columns, click "Detailed profiling "and select Remove from all columns of all sources. 7. Click Submit to execute the profile task. Note: If the table metadata changed since you imported it (for example, a new column was added), you must re-import the source table before you execute the profile task. If you clicked the Submit Column Profile Request option to reach this "Submit Column Profiler Request" window, the Profiler monitor pane appears automatically when you click Submit. If you clicked Update on the "Profile" tab of the "View Data" window, the "Profiler" monitor window does not appear when you click Submit. Instead, a profile task is submitted asynchronously and you must wait for it to complete before you can do other tasks in the Designer. You can also monitor your profiler task by name in the Administrator. 8. When the profiler task has completed, you can view the profile results in the View Data option. Related Topics • Column profile • Monitoring profiler tasks using the Designer • Viewing the profiler results • Administrator Guide: To configure run-time resources • Management Console Guide: Monitoring profiler tasks using the Administrator 15.1.4.2 Submitting relationship profiler tasks A relationship profile shows the percentage of non matching values in columns of two sources. The sources can be any of the following: • • Flat files • 323 Tables A combination of a table and a flat file 2011-06-09
- 324. Data Assessment The columns can have a primary key and foreign key relationship defined or they can be unrelated (as when one comes from a datastore and the other from a file format). The two columns do not need to be the same data type, but they must be convertible. For example, if you run a relationship profile task on an integer column and a varchar column, the Data Profiler converts the integer value to a varchar value to make the comparison. Note: The Data Profiler consumes a large amount of resources when it generates relationship values. If you plan to use Relationship profiling, ensure that you specify a pageable cache directory that contains enough disk space for the amount of data you profile. Related Topics • Data sources that you can profile • Administrator Guide: To configure run-time resources 15.1.4.2.1 To generate a relationship profile for columns in two sources 1. In the Object Library of the Designer, select two sources. To select two sources in the same datastore or file format: a. Go to the "Datastore" or "Format" tab in the Object Library. b. Hold the Ctrl key down as you select the second table. c. Right-click and select Submit Relationship Profile Request . To select two sources from different datastores or files: a. Go to the "Datastore" or "Format" tab in the Object Library. b. Right-click on the first source, select Submit > Relationship Profile Request > Relationship with. c. Change to a different Datastore or Format in the Object Library d. Click on the second source. The "Submit Relationship Profile Request" window appears. Note: You cannot create a relationship profile for the same column in the same source or for columns with a LONG or TEXT data type. 2. (Optional) Edit the profiler task name. You can edit the task name to create a more meaningful name, a unique name, or to remove dashes, which are allowed in column names but not in task names. The default name that the Data Profiler generates for multiple sources has the following format: username_t_firstsourcename_lastsourcename Column username 324 Description Name of the user that the software uses to access system services. 2011-06-09
- 325. Data Assessment Column Description t Type of profile. The value is R for Relationship profile that obtains non matching values in the two selected columns. firstsourcename Name first selected source. lastsourcename Name last selected source. 3. By default, the upper pane of the "Submit Relationship Profile Request" window shows a line between the primary key column and foreign key column of the two sources, if the relationship exists. You can change the columns to profile. The bottom half of the "Submit Relationship Profile Request "window shows that the profile task will use the equal (=) operation to compare the two columns. The Data Profiler will determine which values are not equal and calculate the percentage of non matching values. 4. To delete an existing relationship between two columns, select the line, right-click, and select Delete Selected Relation. To delete all existing relationships between the two sources, do one of the following actions: • Right-click in the upper pane and click Delete All Relations. • Click Delete All Relations near the bottom of the "Submit Relationship Profile Request" window. 5. If a primary key and foreign key relationship does not exist between the two data sources, specify the columns to profile. You can resize each data source to show all columns. To specify or change the columns for which you want to see relationship values: a. Move the cursor to the first column to select. Hold down the cursor and draw a line to the other column that you want to select. b. If you deleted all relations and you want the Data Profiler to select an existing primary-key and foreign-key relationship, either right-click in the upper pane and click Propose Relation, or click Propose Relation near the bottom of the "Submit Relationship Profile Request" window. 6. By default, the is selected. This option indicates that the Data Profiler saves the data only in the columns that you select for the relationship, and you will not see any sample data in the other columns when you view the relationship profile. If you want to see values in the other columns in the relationship profile, select Save all columns data. 7. Click Submit to execute the profiler task. Note: If the table metadata changed since you imported it (for example, a new column was added), you must re-import the source table before you execute the profile task. 8. The Profiler monitor pane appears automatically when you click Submit. You can also monitor your profiler task by name in the Administrator. 325 2011-06-09
- 326. Data Assessment 9. When the profiler task has completed, you can view the profile results in the View Data option when you right click on a table in the Object Library. Related Topics • To view the relationship profile data generated by the Data Profiler • Monitoring profiler tasks using the Designer • Management Console Guide: Monitoring profiler tasks using the Administrator • Viewing the profiler results 15.1.5 Monitoring profiler tasks using the Designer The "Profiler" monitor window appears automatically when you submit a profiler task if you clicked the menu bar to view the "Profiler" monitor window. You can dock this profiler monitor pane in the Designer or keep it separate. The Profiler monitor pane displays the currently running task and all of the profiler tasks that have executed within a configured number of days. You can click on the icons in the upper-left corner of the Profiler monitor to display the following information: Refreshes the Profiler monitor pane to display the latest status of profiler tasks Sources that the selected task is profiling. If the task failed, the "Information" window also displays the error message. The Profiler monitor shows the following columns: 326 2011-06-09
- 327. Data Assessment Column Description Name of the profiler task that was submitted from the Designer. If the profiler task is for a single source, the default name has the following format: Name username_t_sourcename If the profiler task is for multiple sources, the default name has the following format: username_t_firstsourcename_lastsourcename Type The type of profiler task can be: • Column • Relationship The status of a profiler task can be: • Done— The task completed successfully. • Pending— The task is on the wait queue because the maximum number of concurrent tasks has been reached or another task is profiling the same table. • Running— The task is currently executing. • Error — The task terminated with an error. Double-click on the value in this Status column to display the error message. Status Timestamp Date and time that the profiler task executed. Sources Names of the tables for which the profiler task executes. Related Topics • Executing a profiler task • Management Console Guide: Configuring profiler task parameters 15.1.6 Viewing the profiler results 327 2011-06-09
- 328. Data Assessment The Data Profiler calculates and saves the profiler attributes into a profiler repository that multiple users can view. Related Topics • To view the column attributes generated by the Data Profiler • To view the relationship profile data generated by the Data Profiler 15.1.6.1 To view the column attributes generated by the Data Profiler 1. In the Object Library, select the table for which you want to view profiler attributes. 2. Right-click and select View Data. 3. Click the "Profile" tab (second) to view the column profile attributes. a. The "Profile" tab shows the number of physical records that the Data Profiler processed to generate the values in the profile grid. b. The profile grid contains the column names in the current source and profile attributes for each column. To populate the profile grid, execute a profiler task or select names from this column and click Update. c. You can sort the values in each attribute column by clicking the column heading. The value n/a in the profile grid indicates an attribute does not apply to a data type, Relevant data type Basic Profile attribute Description Min Character Numeric Datetime Of all values, the lowest value in this column. Yes Yes Yes Min count Number of rows that contain this lowest value in this column. Yes Yes Yes Max Of all values, the highest value in this column. Yes Yes Yes Max count Number of rows that contain this highest value in this column. Yes Yes Yes 328 2011-06-09
- 329. Data Assessment Relevant data type Basic Profile attribute Description Character Numeric Datetime Average For numeric columns, the average value in this column. n/a Yes n/a Min string length For character columns, the length of the shortest string value in this column. Yes No No Max string length For character columns, the length of the longest string value in this column. Yes No No Average string length For character columns, the average length of the string values in this column. Yes No No Nulls Number of NULL values in this column. Yes Yes Yes Nulls % Percentage of rows that contain a NULL value in this column. Yes Yes Yes Zeros Number of 0 values in this column. No Yes No Zeros % Percentage of rows that contain a 0 value in this column. No Yes No Blanks For character columns, the number of rows that contain a blank in this column. Yes No No 329 2011-06-09
- 330. Data Assessment Relevant data type Basic Profile attribute Description Blanks % Percentage of rows that contain a blank in this column. Character Numeric Datetime Yes No No d. If you selected the Detailed profiling option on the "Submit Column Profile Request" window, the "Profile" tab also displays the following detailed attribute columns. Detailed Profile attribute Description Relevant data type Character Numeric Datetime Distincts Number of distinct values in this column. Yes Yes Yes Distinct % Percentage of rows that contain each distinct value in this column. Yes Yes Yes Median The value that is in the middle row of the source table. Yes Yes Yes Median string length For character columns, the value that is in the middle row of the source table. Yes No No Pattern % Percentage of rows that contain each distinct value in this column. The format of each unique pattern in this column. Yes No No Patterns Number of different patterns in this column. Yes No No 4. Click an attribute value to view the entire row in the source table. The bottom half of the "View Data" window displays the rows that contain the attribute value that you clicked. You can hide columns that you do not want to view by clicking the Show/Hide Columns icon. 330 2011-06-09
- 331. Data Assessment For example, your target ADDRESS column might only be 45 characters, but the Profiling data for this Customer source table shows that the maximum string length is 46. Click the value 46 to view the actual data. You can resize the width of the column to display the entire string. 5. (Optional) Click Update if you want to update the profile attributes. Reasons to update at this point include: • The profile attributes have not yet been generated • The date that the profile attributes were generated is older than you want. The Last updated value in the bottom left corner of the Profile tab is the timestamp when the profile attributes were last generated. Note: The Update option submits a synchronous profile task, and you must wait for the task to complete before you can perform other tasks in the Designer. The "Submit column Profile Request" window appears. Select only the column names you need for this profiling operation because Update calculations impact performance. You can also click the check box at the top in front of Name to deselect all columns and then select each check box in front of each column you want to profile. 6. Click a statistic in either Distincts or Patterns to display the percentage of each distinct value or pattern value in a column. The pattern values, number of records for each pattern value, and percentages appear on the right side of the Profile tab. For example, the following "Profile" tab for table CUSTOMERS shows the profile attributes for column REGION. The Distincts attribute for the REGION column shows the statistic 19 which means 19 distinct values for REGION exist. 331 2011-06-09
- 332. Data Assessment 7. Click the statistic in the Distincts column to display each of the 19 values and the percentage of rows in table CUSTOMERS that have that value for column REGION. In addition, the bars in the right-most column show the relative size of each percentage. 8. The Profiling data on the right side shows that a very large percentage of values for REGION is Null. Click either Null under Value or 60 under Records to display the other columns in the rows that have a Null value in the REGION column. 9. Your business rules might dictate that REGION should not contain Null values in your target data warehouse. Therefore, decide what value you want to substitute for Null values when you define a validation transform. Related Topics • Executing a profiler task • Defining a validation rule based on a column profile 15.1.6.2 To view the relationship profile data generated by the Data Profiler Relationship profile data shows the percentage of non matching values in columns of two sources. The sources can be tables, flat files, or a combination of a table and a flat file. The columns can have a primary key and foreign key relationship defined or they can be unrelated (as when one comes from a datastore and the other from a file format). 1. In the Object Library, select the table or file for which you want to view relationship profile data. 2. Right-click and select View Data. 3. Click the "Relationship" tab (third) to view the relationship profile results. Note: The "Relationship" tab is visible only if you executed a relationship profile task. 4. Click the nonzero percentage in the diagram to view the key values that are not contained within the other table. For example, the following View Data Relationship tab shows the percentage (16.67) of customers that do not have a sales order. The relationship profile was defined on the CUST_ID column in table ODS_CUSTOMER and CUST_ID column in table ODS_SALESORDER. The value in the left oval indicates that 16.67% of rows in table ODS_CUSTOMER have CUST_ID values that do not exist in table ODS_SALESORDER. 332 2011-06-09
- 333. Data Assessment Click the 16.67 percentage in the ODS_CUSTOMER oval to display the CUST_ID values that do not exist in the ODS_SALESORDER table. The non matching values KT03 and SA03 display on the right side of the Relationship tab. Each row displays a non matching CUST_ID value, the number of records with that CUST_ID value, and the percentage of total customers with this CUST_ID value. 5. Click one of the values on the right side to display the other columns in the rows that contain that value. The bottom half of the" Relationship Profile" tab displays the values in the other columns of the row that has the value KT03 in the column CUST_ID. Note: If you did not select Save all column data on the "Submit Relationship Profile Request "window, you cannot view the data in the other columns. Related Topics • Submitting relationship profiler tasks 15.2 Using View Data to determine data quality Use View Data to help you determine the quality of your source and target data. View Data provides the capability to: • 333 View sample source data before you execute a job to create higher quality job designs. 2011-06-09
- 334. Data Assessment • Compare sample data from different steps of your job to verify that your data extraction job returns the results you expect. Related Topics • Defining a validation rule based on a column profile • Using View Data 15.2.1 Data tab The "Data" tab is always available and displays the data contents of sample rows. You can display a subset of columns in each row and define filters to display a subset of rows. For example, your business rules might dictate that all phone and fax numbers be in one format for each country. The following "Data" tab shows a subset of rows for the customers that are in France. Notice that the PHONE and FAX columns displays values with two different formats. You can now decide which format you want to use in your target data warehouse and define a validation transform accordingly. Related Topics • View Data Properties • Defining a validation rule based on a column profile • Data tab 334 2011-06-09
- 335. Data Assessment 15.2.2 Profile tab Two displays are available on the "Profile" tab: • Without the Data Profiler, the "Profile" tab displays the following column attributes: distinct values, NULLs, minimum value, and maximum value. • If you configured and use the Data Profiler, the "Profile" tab displays the same above column attributes plus many more calculated statistics, such as average value, minimum string length, and maximum string length, distinct count, distinct percent, median, median string length, pattern count, and pattern percent. Related Topics • Profile tab • To view the column attributes generated by the Data Profiler 15.2.3 Relationship Profile or Column Profile tab The third tab that displays depends on whether or not you configured and use the Data Profiler. • If you do not use the Data Profiler, the "Column Profile" tab allows you to calculate statistical information for a single column. • If you use the Data Profiler, the "Relationship" tab displays the data mismatches between two columns from which you can determine the integrity of your data between two sources. Related Topics • Column Profile tab • To view the relationship profile data generated by the Data Profiler 15.3 Using the Validation transform The Data Profiler and View Data features can identify anomalies in incoming data. You can then use a Validation transform to define the rules that sort good data from bad. You can write the bad data to a table or file for subsequent review. 335 2011-06-09
- 336. Data Assessment For details on the Validation transform including how to implement reusable validation functions, see the SAP BusinessObjects Data Services Reference Guide. Related Topics • Reference Guide: Transforms, Validation 15.3.1 Analyzing the column profile You can obtain column profile information by submitting column profiler tasks. For example, suppose you want to analyze the data in the Customers table in the Microsoft SQL Server Northwinds sample database. Related Topics • Submitting column profiler tasks 15.3.1.1 To analyze column profile attributes 1. In the object library, right-click the profiled Customers table and select View Data. 2. Select the Profile tab in the "View Data" window. The Profile tab displays the column-profile attributes shown in the following graphic. 336 2011-06-09
- 337. Data Assessment The Patterns attribute for the PHONE column shows the value 20, which means 20 different patterns exist. 3. Click the value 20 in the "Patterns" attribute column. The "Profiling data" pane displays the individual patterns for the column PHONE and the percentage of rows for each pattern. 4. Suppose that your business rules dictate that all phone numbers in France should have the format 99.99.99.99. However, the profiling data shows that two records have the format (9) 99.99.99.99. To display the columns for these two records in the bottom pane, click either (9) 99.99.99.99 under Value or click 2 under Records. You can see that some phone numbers in France have a prefix of (1). You can use a Validation transform to identify rows containing the unwanted prefix. Then you can correct the data to conform to your busness rules then reload it. The next section describes how to configure the Validation transform to identify the errant rows. Related Topics • Defining a validation rule based on a column profile 15.3.2 Defining a validation rule based on a column profile 337 2011-06-09
- 338. Data Assessment This section takes the Data Profiler results and defines the Validation transform according to the sample business rules. Based on the preceding example of the phone prefix (1) for phone numbers in France, the following procedure describes how to define a data flow and validation rule that identifies that pattern. You can then review the failed data, make corrections, and reload the data. 15.3.2.1 To define the validation rule that identifies a pattern This procedure describes how to define a data flow and validation rule that identifies rows containing the (1) prefix described in the previous section. 1. Create a data flow with the Customers table as a source, add a Validation transform and a target, and connect the objects. 2. Open the Validation transform by clicking its name. 3. In the transform editor, click Add. The Rule Editor dialog box displays. 4. Type a Name and optionally a Description for the rule. 5. Verify the Enabled check box is selected. 6. For "Action on Fail", select Send to Fail. 7. Select the Column Validation radio button. a. Select the "Column" CUSTOMERS.PHONE from the drop-down list. b. For "Condition", from the drop-down list select Match pattern. c. For the value, type the expression '99.99.99.99'. 8. Click OK. The rule appears in the Rules list. After running the job, the incorrectly formatted rows appear in the Fail output. You can now review the failed data, make corrections as necessary upstream, and reload the data. Related Topics • Analyzing the column profile 15.4 Using Auditing Auditing provides a way to ensure that a data flow loads correct data into the warehouse. Use auditing to perform the following tasks: • 338 Define audit points to collect run time statistics about the data that flows out of objects. Auditing stores these statistics in the repository. 2011-06-09
- 339. Data Assessment • Define rules with these audit statistics to ensure that the data at the following points in a data flow is what you expect: • Extracted from sources • Processed by transforms • Loaded into targets • Generate a run time notification that includes the audit rule that failed and the values of the audit statistics at the time of failure. • Display the audit statistics after the job execution to help identify the object in the data flow that might have produced incorrect data. Note: If you add an audit point prior to an operation that is usually pushed down to the database server, performance might degrade because pushdown operations cannot occur after an audit point. 15.4.1 Auditing objects in a data flow You can collect audit statistics on the data that flows out of any object, such as a source, transform, or target. If a transform has multiple distinct or different outputs (such as Validation or Case), you can audit each output independently. To use auditing, you define the following objects in the "Audit" window: Object name Audit point 339 Description The object in a data flow where you collect audit statistics. You can audit a source, a transform, or a target. You identify the object to audit when you define an audit function on it. 2011-06-09
- 340. Data Assessment Object name Description The audit statistic that the software collects for a table, output schema, or column. The following table shows the audit functions that you can define. Data object Audit function Description This function collects two statistics: • Table or output schema Good count for rows that were successfully processed. • Error count for rows that generated some type of error if you enabled error handling. Count Sum Column Average Average of the numeric values in the column. Applicable data types include decimal, double, integer, and real. This function only includes the Good rows. Column Audit function Sum of the numeric values in the column. Applicable data types include decimal, double, integer, and real. This function only includes the Good rows. Checksum Checksum of the values in the column. Column Audit label Audit rule A Boolean expression in which you use audit labels to verify the job. If you define multiple rules in a data flow, all rules must succeed or the audit fails. Actions on audit failure 340 The unique name in the data flow that the software generates for the audit statistics collected for each audit function that you define. You use these labels to define audit rules for the data flow. One or more of three ways to generate notification of an audit rule (or rules) failure: email, custom script, raise exception. 2011-06-09
- 341. Data Assessment 15.4.1.1 Audit function This section describes the data types for the audit functions and the error count statistics. Data types The following table shows the default data type for each audit function and the permissible data types. You can change the data type in the "Properties" window for each audit function in the Designer. Audit Functions Default Data Type Allowed Data Types Count INTEGER INTEGER Sum Type of audited column INTEGER, DECIMAL, DOUBLE, REAL Average Type of audited column INTEGER, DECIMAL, DOUBLE, REAL Checksum VARCHAR(128) VARCHAR(128) Error count statistic When you enable a Count audit function, the software collects two types of statistics: • Good row count for rows processed without any error. • Error row count for rows that the job could not process but ignores those rows to continue processing. One way that error rows can result is when you specify the Use overflow file option in the Source Editor or Target Editor. 15.4.1.2 Audit label The software generates a unique name for each audit function that you define on an audit point. You can edit the label names. You might want to edit a label name to create a shorter meaningful name or to remove dashes, which are allowed in column names but not in label names. Generating label names If the audit point is on a table or output schema, the software generates the following two labels for the audit function Count: $Count_objectname $CountError_objectname If the audit point is on a column, the software generates an audit label with the following format: $ auditfunction_objectname 341 2011-06-09
- 342. Data Assessment If the audit point is in an embedded data flow, the labels have the following formats: $Count_objectname_embeddedDFname $CountError_objectname_embeddedDFname $auditfunction_objectname_embeddedDFname Editing label names You can edit the audit label name when you create the audit function and before you create an audit rule that uses the label. If you edit the label name after you use it in an audit rule, the audit rule does not automatically use the new name. You must redefine the rule with the new name. 15.4.1.3 Audit rule An audit rule is a Boolean expression which consists of a Left-Hand-Side (LHS), a Boolean operator, and a Right-Hand-Side (RHS). • The LHS can be a single audit label, multiple audit labels that form an expression with one or more mathematical operators, or a function with audit labels as parameters. • The RHS can be a single audit label, multiple audit labels that form an expression with one or more mathematical operators, a function with audit labels as parameters, or a constant. The following Boolean expressions are examples of audit rules: $Count_CUSTOMER = $Count_CUSTDW $Sum_ORDER_US + $Sum_ORDER_EUROPE = $Sum_ORDER_DW round($Avg_ORDER_TOTAL) >= 10000 15.4.1.4 Audit notification You can choose any combination of the following actions for notification of an audit failure. If you choose all three actions, the software executes them in this order: • Email to list — the software sends a notification of which audit rule failed to the email addresses that you list in this option. Use a comma to separate the list of email addresses. You can specify a variable for the email list. This option uses the smtp_to function to send email. Therefore, you must define the server and sender for the Simple Mail Transfer Protocol (SMTP) in the Server Manager. • • 342 Script — the software executes the custom script that you create in this option. Raise exception — The job fails if an audit rule fails, and the error log shows which audit rule failed. The job stops at the first audit rule that fails. This action is the default. 2011-06-09
- 343. Data Assessment You can use this audit exception in a try/catch block. You can continue the job execution in a try/catch block. If you clear this action and an audit rule fails, the job completes successfully and the audit does not write messages to the job log. You can view which rule failed in the Auditing Details report in the Metadata Reporting tool. For more information, see Viewing audit results . 15.4.2 Accessing the Audit window Access the Audit window from one of the following places in the Designer: • From the Data Flows tab of the object library, right-click on a data flow name and select the Auditing option. • In the workspace, right-click on a data flow icon and select the Auditing option. • When a data flow is open in the workspace, click the Audit icon in the toolbar. When you first access the Audit window, the Label tab displays the sources and targets in the data flow. If your data flow contains multiple consecutive query transforms, the Audit window shows the first query. Click the icons on the upper left corner of the Label tab to change the display. Icon Description Collapse All Collapses the expansion of the source, transform, and target objects. Show All Objects Displays all the objects within the data flow. Show Source, Target and first-level Query Default display which shows the source, target, and first-level query objects in the data flow. If the data flow contains multiple consecutive query transforms, only the first-level query displays. Show Labelled Objects 343 Tool tip Displays the objects that have audit labels defined. 2011-06-09
- 344. Data Assessment 15.4.3 Defining audit points, rules, and action on failure 1. Access the "Audit" window. 2. Define audit points. On the Label tab, right-click on an object that you want to audit and choose an audit function or Properties. When you define an audit point, the software generates the following: • • An audit icon on the object in the data flow in the workspace An audit label that you use to define audit rules. In addition to choosing an audit function, the Properties window allows you to edit the audit label and change the data type of the audit function. For example, the data flow Case_DF has the following objects and you want to verify that all of the source rows are processed by the Case transform. • • Source table ODS_CUSTOMER Four target tables: R1 contains rows where ODS_CUSTOMER.REGION_ID = 1 R2 contains rows where ODS_CUSTOMER.REGION_ID = 2 R3 contains rows where ODS_CUSTOMER.REGION_ID = 3 R123 contains rows where ODS_CUSTOMER.REGION_ID IN (1, 2 or 3) a. Right-click on source table ODS_CUSTOMER and choose Count. The software creates the audit labels $Count_ODS_CUSTOMER and $CountError_ODS_CUSTOMER, and an audit icon appears on the source object in the workspace. 344 2011-06-09
- 345. Data Assessment b. Similarly, right-click on each of the target tables and choose Count. The Audit window shows the following audit labels. Target table Audit Function Audit Label ODS_CUSTOMER Count $Count_ODS_CUSTOMER R1 Count $Count_ R1 R2 Count $Count_ R2 R3 Count $Count_ R3 R123 Count $Count_ R123 c. If you want to remove an audit label, right-click on the label, and the audit function that you previously defined displays with a check mark in front of it. Click the function to remove the check mark and delete the associated audit label. When you right-click on the label, you can also select Properties, and select the value (No Audit) in the Audit function drop-down list. 3. Define audit rules. On the Rule tab in the "Audit" window, click Add which activates the expression editor of the Auditing Rules section. If you want to compare audit statistics for one object against one other object, use the expression editor, which consists of three text boxes with drop-down lists: a. Select the label of the first audit point in the first drop-down list. b. Choose a Boolean operator from the second drop-down list. The options in the editor provide common Boolean operators. If you require a Boolean operator that is not in this list, use the Custom expression box with its function and smart editors to type in the operator. c. Select the label for the second audit point from the third drop-down list. If you want to compare the first audit value to a constant instead of a second audit value, use the Customer expression box. For example, to verify that the count of rows from the source table is equal to the rows in the target table, select audit labels and the Boolean operation in the expression editor as follows: If you want to compare audit statistics for one or more objects against statistics for multiple other objects or a constant, select the Custom expression box. a. b. c. d. e. Click the ellipsis button to open the full-size smart editor window. Click the Variables tab on the left and expand the Labels node. Drag the first audit label of the object to the editor pane. Type a Boolean operator Drag the audit labels of the other objects to which you want to compare the audit statistics of the first object and place appropriate mathematical operators between them. f. Click OK to close the smart editor. g. The audit rule displays in the Custom editor. To update the rule in the top Auditing Rule box, click on the title "Auditing Rule" or on another option. 345 2011-06-09
- 346. Data Assessment h. Click Close in the Audit window. For example, to verify that the count of rows from the source table is equal to the sum of rows in the first three target tables, drag the audit labels, type in the Boolean operation and plus signs in the smart editor as follows: Count_ODS_CUSTOMER = $Count_R1 + $Count_R2 + $Count_R3 4. Define the action to take if the audit fails. You can choose one or more of the following actions: • Raise exception: The job fails if an audit rule fails and the error log shows which audit rule failed. This action is the default. If you clear this option and an audit rule fails, the job completes successfully and the audit does not write messages to the job log. You can view which rule failed in the Auditing Details report in the Metadata Reporting tool. • Email to list: The software sends a notification of which audit rule failed to the email addresses that you list in this option. Use a comma to separate the list of email addresses. You can specify a variable for the email list. • Script: The software executes the script that you create in this option. 5. Execute the job. The "Execution Properties" window has the Enable auditing option checked by default. Clear this box if you do not want to collect audit statistics for this specific job execution. 6. Look at the audit results. You can view passed and failed audit rules in the metadata reports. If you turn on the audit trace on the Trace tab in the "Execution Properties" window, you can view all audit results on the Job Monitor Log. Related Topics • Auditing objects in a data flow • Viewing audit results 15.4.4 Guidelines to choose audit points The following are guidelines to choose audit points: • When you audit the output data of an object, the optimizer cannot pushdown operations after the audit point. Therefore, if the performance of a query that is pushed to the database server is more important than gathering audit statistics from the source, define the first audit point on the query or later in the data flow. For example, suppose your data flow has a source, query, and target objects, and the query has a WHERE clause that is pushed to the database server that significantly reduces the amount of data 346 2011-06-09
- 347. Data Assessment that returns to the software. Define the first audit point on the query, rather than on the source, to obtain audit statistics on the query results. • If a pushdown_sql function is after an audit point, the software cannot execute it. • You can only audit a bulkload that uses the Oracle API method. For the other bulk loading methods, the number of rows loaded is not available to the software. • Auditing is disabled when you run a job with the debugger. • You cannot audit NRDM schemas or real-time jobs. • You cannot audit within an ABAP Dataflow, but you can audit the output of an ABAP Dataflow. • If you use the CHECKSUM audit function in a job that normally executes in parallel, the software disables the DOP for the whole data flow. The order of rows is important for the result of CHECKSUM, and DOP processes the rows in a different order than in the source. 15.4.5 Auditing embedded data flows You can define audit labels and audit rules in an embedded data flow. This section describes the following considerations when you audit embedded data flows: • Enabling auditing in an embedded data flow • Audit points not visible outside of the embedded data flow 15.4.5.1 Enabling auditing in an embedded data flow If you want to collect audit statistics on an embedded data flow when you execute the parent data flow, you must enable the audit label of the embedded data flow. 15.4.5.1.1 To enable auditing in an embedded data flow 1. Open the parent data flow in the Designer workspace. 2. Click on the Audit icon in the toolbar to open the Audit window 3. On the Label tab, expand the objects to display any audit functions defined within the embedded data flow. If a data flow is embedded at the beginning or at the end of the parent data flow, an audit function might exist on the output port or on the input port. The following Audit window shows an example of an embedded audit function that does not have an audit label defined in the parent data flow. 347 2011-06-09
- 348. Data Assessment 4. Right-click the Audit function and choose Enable. You can also choose Properties to change the label name and enable the label. 5. You can define audit rules with the enabled label. 15.4.5.2 Audit points not visible outside of the embedded data flow When you embed a data flow at the beginning of another data flow, data passes from the embedded data flow to the parent data flow through a single source. When you embed a data flow at the end of another data flow, data passes into the embedded data flow from the parent through a single target. In either case, some of the objects are not visible in the parent data flow. Because some of the objects are not visible in the parent data flow, the audit points on these objects are also not visible in the parent data flow. For example, the following embedded data flow has an audit function defined on the source SQL transform and an audit function defined on the target table. The following Audit window shows these two audit points. 348 2011-06-09
- 349. Data Assessment When you embed this data flow, the target Output becomes a source for the parent data flow and the SQL transform is no longer visible. An audit point still exists for the entire embedded data flow, but the label is no longer applicable. The following Audit window for the parent data flow shows the audit function defined in the embedded data flow, but does not show an Audit Label. If you want to audit the embedded data flow, right-click on the audit function in the Audit window and select Enable. 349 2011-06-09
- 350. Data Assessment 15.4.6 Resolving invalid audit labels An audit label can become invalid in the following situations: • If you delete the audit label in an embedded data flow that the parent data flow has enabled. • If you delete or rename an object that had an audit point defined on it 15.4.6.1 To resolve invalid audit labels 1. 2. 3. 4. Open the Audit window. Expand the Invalid Labels node to display the individual labels. Note any labels that you would like to define on any new objects in the data flow. After you define a corresponding audit label on a new object, right-click on the invalid label and choose Delete. 5. If you want to delete all of the invalid labels at once, right click on the Invalid Labels node and click on Delete All. 15.4.7 Viewing audit results You can see the audit status in one of the following places: • Job Monitor Log • If the audit rule fails, the places that display audit information depends on the Action on failure option that you chose: Action on failure Raise exception Job Error Log, Metadata Reports Email to list Email message, Metadata Reports Script 350 Places where you can view audit information Wherever the custom script sends the audit messages, Metadata Reports 2011-06-09
- 351. Data Assessment Related Topics • Job Monitor Log • Job Error Log • Metadata Reports 15.4.7.1 Job Monitor Log If you set Audit Trace to Yes on the Trace tab in the Execution Properties window, audit messages appear in the Job Monitor Log. You can see messages for audit rules that passed and failed. The following sample audit success messages appear in the Job Monitor Log when Audit Trace is set to Yes: Audit Label $Count_R2 = 4. Data flow <Case_DF>. Audit Label $CountError_R2 = 0. Data flow <Case_DF>. Audit Label $Count_R3 = 3. Data flow <Case_DF>. Audit Label $CountError_R3 = 0. Data flow <Case_DF>. Audit Label $Count_R123 = 12. Data flow <Case_DF>. Audit Label $CountError_R123 = 0. Data flow <Case_DF>. Audit Label $Count_R1 = 5. Data flow <Case_DF>. Audit Label $CountError_R1 = 0. Data flow <Case_DF>. Audit Label $Count_ODS_CUSTOMER = 12. Data flow <Case_DF>. Audit Label $CountError_ODS_CUSTOMER = 0. Data flow <Case_DF>. Audit Rule passed ($Count_ODS_CUSTOMER = (($CountR1 + $CountR2 + $Count_R3)): LHS=12, RHS=12. Data flow <Case_DF>. Audit Rule passed ($Count_ODS_CUSTOMER = $CountR123): LHS=12, RHS=12. Data flow <Case_DF>. 15.4.7.2 Job Error Log When you choose the Raise exception option and an audit rule fails, the Job Error Log shows the rule that failed. The following sample message appears in the Job Error Log: Audit rule failed <($Count_ODS_CUSTOMER = $CountR1)> for <Data flow Case_DF>. 15.4.7.3 Metadata Reports You can look at the Audit Status column in the Data Flow Execution Statistics reports of the Metadata Report tool. This Audit Status column has the following values: • 351 Not Audited 2011-06-09
- 352. Data Assessment • • • Passed — All audit rules succeeded. This value is a link to the Auditing Details report which shows the audit rules and values of the audit labels. Information Collected — This status occurs when you define audit labels to collect statistics but do not define audit rules. This value is a link to the Auditing Details report which shows the values of the audit labels. Failed — Audit rule failed. This value is a link to the Auditing Details report which shows the rule that failed and values of the audit labels. Related Topics • Management Console Guide: Operational Dashboard Reports 352 2011-06-09
- 353. Data Quality Data Quality 16.1 Overview of data quality Data quality is a term that refers to the set of transforms that work together to improve the quality of your data by cleansing, enhancing, matching and consolidating data elements. Data quality is primarily accomplished in the software using four transforms: • • • • Address Cleanse. Parses, standardizes, corrects, and enhances address data. Data Cleanse. Parses, standardizes, corrects, and enhances customer and operational data. Geocoding. Uses geographic coordinates, addresses, and point-of-interest (POI) data to append address, latitude and longitude, census, and other information to your records. Match. Identifies duplicate records at multiple levels within a single pass for individuals, households, or corporations within multiple tables or databases and consolidates them into a single source. Related Topics • Address Cleanse • Geocoding • Matching strategies 16.2 Data Cleanse 16.2.1 About cleansing data Data cleansing is the process of parsing and standardizing data. The parsing rules and other information that define how to parse and standardize are stored in a cleansing package. The Cleansing Package Builder in SAP BusinessObjects Information Steward provides a graphical user interface to create and refine cleansing packages. You can create a cleansing 353 2011-06-09
- 354. Data Quality package from scratch based on sample data or adapt an existing cleansing package or SAP-supplied cleansing package to meet your specific data cleansing requirements and standards. A cleansing package is created and published within Cleansing Package Builder and then referenced by the Data Cleanse transform within SAP BusinessObjects Data Services for testing and production deployment. Within a Data Services work flow, the Data Cleanse transform identifies and isolates specific parts of mixed data, and then parses and formats the data based on the referenced cleansing package as well as options set directly in the transform. The following diagram shows how SAP BusinessObjects Data Services and SAP BusinessObjects Information Steward work together to allow you to develop a cleansing package specific to your data requirements and then apply it when you cleanse your data. 16.2.2 Cleansing package lifecycle: develop, deploy and maintain The process of developing, deploying, and maintaining a cleansing package is the result of action and communication between the Data Services administrator, Data Services tester, and Cleansing Package Builder data steward. The exact roles, responsibilities, and titles vary by organization, but often include the following: 354 2011-06-09
- 355. Data Quality Role Responsibility Cleansing Package Builder data steward Uses Cleansing Package Builder and has domain knowledge to develop and refine a cleansing package for a specific data domain. Data Services tester In a Data Services test environment, uses the Data Cleanse transform to cleanse data and verify the results. Works with the Cleansing Package Builder data steward to refine a cleansing package. Data Services administrator In a Data Services production environment, uses the Data Cleanse transform to cleanse data based on the rules and standards defined in the selected cleansing package. There are typically three iterative phases in a cleansing package workflow: develop (create and test), deploy, and maintain. In the create and test phase, the data steward creates a cleansing package based on sample data provided by the Data Services administrator and then works with the Data Services tester to refine the cleansing package. When everyone is satisfied with the results, the cleansing package is deployed to production. In the deployment phase the Data Services administrator, tester, and data steward work together to further refine the cleansing package so that production data is cleansed within the established acceptable range. Finally, the cleansing package is moved to the maintenance phase and updated only when the results of regularly scheduled jobs fall out of range or when new data is introduced. A typical workflow is shown in the diagram below: 355 2011-06-09
- 356. Data Quality 16.2.3 Configuring the Data Cleanse transform Prerequisites for configuring the Data Cleanse transform include: • Access to the necessary cleansing package. • Access to the ATL file transferred from Cleansing Package Builder. • Input field and attribute (output field) mapping information for user-defined pattern matching rules defined in the Reference Data tab of Cleansing Package Builder. 356 2011-06-09
- 357. Data Quality To configure the Data Cleanse transform: 1. Import the ATL file transferred from Cleansing Package Builder. Importing the ATL file brings the required information and automatically sets the following options: • Cleansing Package • Engine • Filter Output Fields • Input Word Breaker • Parser Configuration Note: You can install and use SAP-supplied cleansing packages without modifications directly in Data Services. To do so, skip step 1 and manually set any required options in the Data Cleanse transform. 2. In the input schema, select the input fields that you want to map and drag them to the appropriate fields in the Input tab. • • • Name and firm data can be mapped either to discrete fields or multiline fields. Custom data must be mapped to multiline fields. Phone, date, email, Social Security number, and user-defined pattern data can be mapped either to discrete fields or multiline fields. The corresponding parser must be enabled. 3. In the Options tab, select the appropriate option values to determine how Data Cleanse will process your data. If you change an option value from its default value, a green triangle appears next to the option name to indicate that the value has been changed. The ATL file that you imported in step 1 sets certain options based on information in the cleansing package. 4. In the Output tab, select the fields that you want to output from the transform. In Cleansing Package Builder, output fields are referred to as attributes. Ensure that you map any attributes (output fields) defined in user-defined patterns in Cleansing Package Builder reference data. Related Topics • Transform configurations • Data Quality transform editors • To add a Data Quality transform to a data flow 16.2.4 Ranking and prioritizing parsing engines When dealing with multiline input, you can configure the Data Cleanse transform to use only specific parsers and to specify the order the parsers are run. Carefully selecting which parsers to use and in what order can be beneficial. Turning off parsers that you do not need significantly improves parsing speed and reduces the chances that your data will be parsed incorrectly. 357 2011-06-09
- 358. Data Quality You can change the parser order for a specific multiline input by modifying the corresponding parser sequence option in the Parser_Configuration options group of the Data Cleanse transform. For example, to change the order of parsers for the Multiline1 input field, modify the Parser_Sequence_Multiline1 option. To change the selected parsers or the parser order: select a parser sequence, click OK at the message and then use the "Ordered Options" window to make your changes. Note: In the "Ordered Options" window, parsers that are not valid are displayed in red. Related Topics • Ordered options editor 16.2.5 About parsing data The Data Cleanse transform can identify and isolate a wide variety of data. Within the Data Cleanse transform, you map the input fields in your data to the appropriate input fields in the transform. Custom data containing operational or product data is always mapped to multiline fields. Person and firm data, phone, date, email, and Social Security number data can be mapped to either discrete input fields or multiline input fields. The example below shows how Data Cleanse parses product data from a multiline input field and displays it in discrete output fields. The data also can be displayed in composite fields, such as “Standard Description”, which can be customized in Cleansing Package Builder to meet your needs. 358 2011-06-09
- 359. Data Quality Input data Parsed data Glove ultra grip profit 2.3 large black synthetic leather elastic with Velcro Mechanix Wear Product Category Glove Size Large Material Synthetic Leather Trademark Pro-Fit 2.3 Series Cuff Style Elastic Velcro Palm Type Ultra-Grip Color Black Vendor Mechanix Wear Standard Description Glove - Synthetic Leather, Black, size: Large, Cuff Style: Elastic Velcro, Ultra-Grip, Mechanix Wear The examples below show how Data Cleanse parses name and firm data and displays it in discrete output fields. The data also can be displayed in composite fields which can be customized in Cleansing Package Builder to meet your needs. Input data Parsed data Prename Given Name 1 Dan Given Name 2 R. Smith Maturity Postname Jr. Honorary Postname CPA Title Account Mgr. Firm Jones, Inc. Extra PO Box 567 Extra 359 Mr. Family Name 1 Mr. Dan R. Smith, Jr., CPA Account Mgr. Jones Inc. PO Box 567 Wisconsin Rapids, WI 54495 Wisconsin Rapids, WI 54495 2011-06-09
- 360. Data Quality Input data Parsed data Given Name 1 James Family Name 1 Witt Social Security 421-55-2424 E-mail address [email protected] Phone 507.555.3423 Date James Witt 421-55-2424 [email protected] 507-555-3423 Aug 20, 2003 August 20, 2003 The Data Cleanse transform parses up to six names per record, two per input field. For all six names found, it parses components such as prename, given names, family name, and postname. Then it sends the data to individual fields. The Data Cleanse transform also parses up to six job titles per record. The Data Cleanse transform parses up to six firm names per record, one per input field. 16.2.5.1 About parsing phone numbers Data Cleanse can parse both North American Numbering Plan (NANP) and international phone numbers. When Data Cleanse parses a phone number, it outputs the individual components of the number into the appropriate fields. Phone numbering systems differ around the world. Data Cleanse recognizes phone numbers by their pattern and (for non-NANP numbers) by their country code, too. Data Cleanse searches for North American phone numbers by commonly used patterns such as: (234) 567-8901, 234-567-8901, and 2345678901. Data Cleanse gives you the option for some reformatting on output (such as your choice of delimiters). Data Cleanse searches for non-North American numbers by pattern. The patterns used are specified in Cleansing Package Builder in the Reference Data tab. The country code must appear at the beginning of the number. Data Cleanse does not offer any options for reformatting international phone numbers. Also, Data Cleanse does not cross-compare to the address to see whether the country and city codes in the phone number match the address. 16.2.5.2 About parsing dates Data Cleanse recognizes dates in a variety of formats and breaks those dates into components. 360 2011-06-09
- 361. Data Quality Data Cleanse can parse up to six dates from your defined record. That is, Data Cleanse identifies up to six dates in the input, breaks those dates into components, and makes dates available as output in either the original format or a user-selected standard format. 16.2.5.3 About parsing Social Security numbers Data Cleanse parses U.S. Social Security numbers (SSNs) that are either by themselves or on an input line surrounded by other text. Fields used Data Cleanse outputs the individual components of a parsed Social Security number—that is, the entire SSN, the area, the group, and the serial. How Data Cleanse parses Social Security numbers Data Cleanse parses Social Security numbers in two steps: 1. Identifies a potential SSN by looking for the following patterns: Pattern Digits per grouping Delimited by nnnnnnnnn 9 consecutive digits not applicable nnn nn nnnn 3, 2, and 4 (for area, group, and serial) spaces nnn-nn-nnnn 3, 2, and 4 (for area, group, and serial) all supported delimiters 2. Performs a validity check on the first five digits only. The possible outcomes of this validity check are: Outcome Description Pass Data Cleanse successfully parses the data—and the Social Security number is output to a SSN output field. Fail Data Cleanse does not parse the data because it is not a valid Social Security number as defined by the U.S. government. The data is output as Extra, unparsed data. Check validity When performing a validity check, Data Cleanse does not verify that a particular 9-digit Social Security number has been issued, or that it is the correct number for any named person. Instead, it validates only the first 5 digits (area and group). Data Cleanse does not validate the last 4 digits (serial)—except to confirm that they are digits. 361 2011-06-09
- 362. Data Quality SSA data Data Cleanse validates the first five digits based on a table from the Social Security Administration (http://www.ssa.gov/employer/highgroup.txt). That table is updated monthly as the SSA opens new groups. The rules and data that guide this check are available at http://www.ssa.gov/history/ssn/geo card.html. The Social Security number information that Data Cleanse references is included in the cleansing package. The data steward responsible for the cleansing package can ensure that it contains the most recent information. Outputs valid SSNs Data Cleanse outputs only Social Security numbers that pass its validation. If an apparent SSN fails validation, Data Cleanse does not pass on the number as a parsed, but invalid, Social Security number. Related Topics • Reference Guide: Transforms, Data Cleanse output fields 16.2.5.4 About parsing email addresses When Data Cleanse parses input data that it determines is an email address, it places the components of that data into specific fields for output. Below is an example of a simple email address: [email protected] By identifying the various data components (user name, host, and so on) by their relationships to each other, Data Cleanse can assign the data to specific attributes (output fields). Output fields Data Cleanse uses Data Cleanse outputs the individual components of a parsed email address—that is, the email user name, complete domain name, top domain, second domain, third domain, fourth domain, fifth domain, and host name. What Data Cleanse does Data Cleanse can take the following actions: • • • • Parse an email address located either in a discrete field or combined with other data in a multiline field. Break down the domain name down into sub-elements. Verify that an email address is properly formatted. Flag that the address includes an internet service provider (ISP) or email domain name listed in the email type of Reference Data in Cleansing Package Builder. This flag is shown in the Email_is_ISP output field. What Data Cleanse does not verify Several aspects of an email address are not verified by Data Cleanse. Data Cleanse does not verify: 362 2011-06-09
- 363. Data Quality • • • • whether the domain name (the portion to the right of the @ sign) is registered. whether an email server is active at that address. whether the user name (the portion to the left of the @ sign) is registered on that email server (if any). whether the personal name in the record can be reached at this email address. Email components The output field where Data Cleanse places the data depends on the position of the data in the record. Data Cleanse follows the Domain Name System (DNS) in determining the correct output field. For example, if [email protected] were input data, Data Cleanse would output the elements in the following fields: Output field Output value Email [email protected] Email_User expat Email_Domain_All london.home.office.city.co.uk Email_Domain_Top uk Email_Domain_Second co Email_Domain_Third city Email_Domain_Fourth office Email_Domain_Fifth home Email_Domain_Host london Related Topics • Data Services Reference Guide: Transforms, Data Cleanse output fields 16.2.5.5 About parsing user-defined patterns Data Cleanse can parse patterns found in a wide variety of data such as: • • • • • 363 account numbers part numbers purchase orders invoice numbers VINs (vehicle identification numbers) 2011-06-09
- 364. Data Quality • driver license numbers In other words, Data Cleanse can parse any alphanumeric sequence for which you can define a pattern. The user-defined pattern matching (UDPM) parser looks for the pattern across each entire field. Patterns are defined using regular expressions in the Reference Data tab of Cleansing Package Builder. Check with the cleansing package owner to determine any required mappings for input fields and output fields (attributes). 16.2.5.6 About parsing street addresses Data Cleanse does not identify and parse individual address components. To parse data that contains address information, process it using a Global Address Cleanse or U.S. Regulatory Address Cleanse transform prior to Data Cleanse. If address data is processed by the Data Cleanse transform, it is usually output to the "Extra" fields. Related Topics • How address cleanse works 16.2.6 About standardizing data Standard forms for individual variations are defined within a cleansing package using Cleansing Package Builder. Additionally, the Data Cleanse transform can standardize data to make its format more consistent. Data characteristics that the transform can standardize include case, punctuation, and abbreviations. 16.2.7 About assigning gender descriptions and prenames Each variation in a cleansing package has a gender associated with it. By default, the gender is “unassigned”. You can assign a gender to a variation in the Advanced mode of Cleansing Package Builder. Gender descriptions are: strong male, strong female, weak male, weak female, and ambiguous. Variations in SAP-supplied name and firm cleansing packages have been assigned genders. You can use the Data Cleanse transform to output the gender associated with a variation to the GENDER output field. 364 2011-06-09
- 365. Data Quality The Prename output field always includes prenames that are part of the name input data. Additionally, when the Assign Prenames option is set to Yes, Data Cleanse populates the PRENAME output field when a strong male or strong female gender is assigned to a variation. When dual names are parsed, Data Cleanse offers four additional gender descriptions: female multi-name, male multi-name, mixed multi-name, and ambiguous multi-name. These genders are generated within Data Cleanse based on the assigned genders of the two names. The table below shows how the multi-name genders are assigned: Dual name Gender of first name Gender of second name Assigned gender for dual name Bob and Sue Jones strong male strong female mixed multi-name Bob and Tom Jones strong male strong male male multi-name Sue and Sara Jones strong female strong female female multi-name Bob and Pat Jones strong male ambiguous ambiguous multi-name Related Topics • Reference Guide: Transforms, Data Cleanse, Data Cleanse options, Gender Standardization options 16.2.8 Prepare records for matching If you are planning a data flow that includes matching, it is recommended that you first use Data Cleanse to standardize the data to enhance the accuracy of your matches. The Data Cleanse transform should be upstream from the Match transform. The Data Cleanse transform can generate match standards or alternates for many name and firm fields as well as all custom output fields. For example, Data Cleanse can tell you that Patrick and Patricia are potential matches for the name Pat. Match standards can help you overcome two types of matching problems: alternate spellings (Catherine and Katherine) and nicknames (Pat and Patrick). This example shows how Data Cleanse can prepare records for matching. 365 2011-06-09
- 366. Data Quality Table 16-8: Data source 1 Input record Cleansed record Intl Marketing, Inc. Given Name 1 Pat Pat Smith, Accounting Mgr. Match Standards Patrick, Patricia 328 Bluebird Ln Given Name 2 Wisconsin Rapids, WI 54494 Family Name 1 Smith Title Accounting Mgr. Firm Intl. Mktg, Inc. Extra 328 Bluebird Ln Extra Wisconsin Rapids Extra WI Extra 54494 Table 16-9: Data source 2 Input record Cleansed record Smith, Patricia R. Given Name 1 International Marketing, Incorp. Match Standards 328 Bluebird Ln Given Name 2 R Wisconsin Rapids, Wisconsin Family Name 1 Smith Patricia Title Firm Intl. Mktg, Inc. Extra 328 Bluebird Ln Extra Wisconsin Rapids Extra WI When a cleansing package does not include an alternate, the match standard output field for that term will be empty. In the case of a multi-word output such as a firm name, when none of the variations in the firm name have an alternate, then the match standard output will be empty. However, if at least one variation has an alternate associated with it, the match standard is generated using the variation alternate where available and the variations for words that do not have an alternate. 366 2011-06-09
- 367. Data Quality 16.2.9 Region-specific data 16.2.9.1 Cleansing packages and transforms SAP offers SAP-supplied person and firm cleansing packages for a variety of regions. Each cleansing package is designed to enhance the ability of Data Cleanse to appropriately cleanse the data according to the cultural standards of the region. The table below illustrates how name parsing may vary by culture: Parsed Output Culture Name Given_Name1 Given_Name2 Family_Name1 C. Sánchez Spanish Juan C. Sánchez Juan Portuguese João A. Lopes João A. Lopes French Jean Christophe Rousseau Jean Christophe Rousseau German Hans Joachim Müller Hans Joachim Müller American James Andrew Smith James Andrew Smith Because the cleansing packages are based on the standard Data Cleanse transform, you can use the sample transforms in your projects in the same way you would use the base Data Cleanse transform and gain the advantage of the enhanced regional accuracy. 16.2.9.2 Customize prenames per country When the input name does not include a prename, Data Cleanse generates the English prenames Mr. and Ms. To modify these terms, add a Query transform following the Data Cleanse transform and use the search_replace function to replace the terms with region-appropriate prenames. 367 2011-06-09
- 368. Data Quality 16.2.9.3 Personal identification numbers Data Cleanse can identify U.S. Social Security numbers and separate them into discrete components. If your data includes personal identification numbers other than U.S. Social Security numbers, you can create user-defined pattern rules to identify the numbers. User-defined pattern rules are part of the cleansing package and are defined in the Edit Reference Data tab of Cleansing Package Builder. User-defined pattern rules are parsed in Data Cleanse with the UDPM parser. U.S. Social Security numbers are parsed in Data Cleanse with the SSN parser. Related Topics • Information Steward User Guide: Cleansing Package Builder, Parse common types of data (reference data), 16.2.10 Japanese data 16.2.10.1 About Japanese data Data Cleanse can identify and parse Japanese data or mixed data that contains both Japanese and Latin characters. To ensure that Data Cleanse parses the data correctly, you must use the Japanese engine. In general, Data Cleanse uses a word breaker to break an input string into individual parsed values and then attempts to recombine adjacent parsed values into variations. Each variation is assigned one or more classifications based on how the variation is defined in the cleansing package. The input is then parsed according to the parser and parsing rules defined in the cleansing package. Due to its structure, Japanese data cannot be accurately broken and parsed using the same algorithm as other data. When the Data Cleanse Japanese engine is used, Data Cleanse first identifies the script in each input field as kanji, kana, or Latin and assigns it to the appropriate script classification. Input fields containing data classified as kana or kanji script are then processed using a special Japanese lexer and parser. Input fields containing data classified as Latin script are processed using the regular Data Cleanse methodology. Note: Only data in Latin script is parsed based on the value set for the Parse data on whitespace only transform option. All kana and kanji input is broken by the Japanese word breaker. 368 2011-06-09
- 369. Data Quality 16.2.10.2 Text width in output fields Many Japanese characters are represented in both fullwidth and halfwidth forms. Latin characters can be encoded in either a proportional or fullwidth form. In either case, the fullwidth form requires more space than the halfwidth or proportional form. To standardize your data, you can use the Character Width Style option to set the character width for all output fields to either fullwidth or halfwidth. The normal width value reflects the normalized character width based on script type. Thus some output fields contain halfwidth characters and other fields contain fullwidth characters. For example, all fullwidth Latin characters are standardized to their halfwidth forms and all halfwidth katakana characters are standardized to their fullwidth forms. NORMAL_WIDTH does not require special processing and thus is the most efficient setting. Note: Because the output width is based on the normalized width for the character type, the output data may be larger than the input data. You may need to increase the column width in the target table. For template tables, selecting the Use NVARCHAR for VARCHAR columns in supported databases box changes the VARCHAR column type to NVARCHAR and allows for increased data size. Related Topics • Reference Guide: Locales and Multi-byte Functionality, Multi-byte support, Column Sizing 16.3 Geocoding This section describes how the Geocoder transform works, different ways that you can use the transform, and how to understand your output. Note: GeoCensus functionality in the USA Regulatory Address Cleanse transform will be deprecated in a future version. It is recommended that you upgrade any data flows that currently use the GeoCensus functionality to use the Geocoder transform. For instructions on upgrading from GeoCensus to the Geocoder transform, see the Upgrade Guide. How the Geocoder transform works The Geocoder transform uses geographic coordinates expressed as latitude and longitude, addresses, and point-of-interest (POI) data to append data to your records. Using the transform, you can append address, latitude and longitude, census data, and other information. For census data, you can use census data from two census periods to compare data, when available. Based on mapped input fields, the Geocoder transform has two modes of geocode processing: 369 2011-06-09
- 370. Data Quality • • point-of-interest and address geocoding point-of-interest and address reverse geocoding In general, the transform uses geocoding directories to calculate latitude and longitude values for a house by interpolating between a beginning and ending point of a line segment where the line segment represents a range of houses. The latitude and longitude values may be slightly offset from the exact location from where the house actually exists. The Geocoder transform also supports geocoding parcel directories, which contain the most precise and accurate latitude and longitude values available for addresses, depending on the available country data. Geocoding parcel data is stored as points, so rather than getting you near the house, it takes you to the exact door. Typically, the Geocoder transform is used in conjunction with the Global Address Cleanse or USA Regulatory Address Cleanse transform. Related Topics • Reference Guide: Transforms, Geocoder • Reference Guide: Data Quality Fields, Geocoder fields • GeoCensus (USA Regulatory Address Cleanse) 16.3.1 POI and address geocoding In address geocoding mode, the Geocoder transform assigns geographic data. Based on the completeness of the input address data, the Geocoder transform can return multiple levels of latitude and longitude data. Including latitude and longitude information in your data may help your organization to target certain population sizes and other regional geographical data. If you have a complete address as input data, including the primary number, the Geocoder transform returns the latitude and longitude coordinates to the exact location. If you have an address that has only a locality or Postcode, you receive coordinates in the locality or Postcode area, respectively. Point-of-interest geocoding lets you provide an address or geographical coordinates to return a list of locations that match your criteria within a geographical area. A point of interest, or POI, is the name of a location that is useful or interesting, such as a gas station or historical monument. Prepare records for geocoding The Geocoder transform works best when it has standardized and corrected address data, so to obtain the most accurate information you may want to place an address cleanse transform before the Geocoder transform in the workflow. 370 2011-06-09
- 371. Data Quality 16.3.1.1 Geocoding scenarios Scenario 1 Scenario: Use an address or an address and a point of interest to assign latitude and longitude information. Number of output results: Single record The following sections describe the required and optional input fields and available output fields to obtain results for this scenario. We also provide an example with sample data. Required input fields For required input fields, the Country field must be mapped. The more input data you can provide, the better results you will obtain. Category Input field name Address Country (required) Locality1–4 Postcode1–2 Primary_Name1–4 Primary_Number Primary_Postfix1 Primary_Prefix1 Primary_Type1–4 Region1–2 Optional input fields Category Input field name Address POI POI_Name POI_Type Available output fields All output fields are optional. 371 2011-06-09
- 372. Data Quality Category Output field name Assignment Level Assignment_Level Assignment_Level_Locality Assignment_Level_Postcode Census Data Census_Tract_Block Census_Tract_Block_Prev Census_Tract_Block_Group Census_Tract_Block_Group_Prev Gov_County_Code Gov_Locality1_Code Gov_Region1_Code Metro_Stat_Area_Code Metro_Stat_Area_Code_Prev Minor_Div_Code Minor_Div_Code_Prev Stat_Area_Code Stat_Area_Code_Prev Info Code Info_Code Latitude/Longitude Latitude Latitude_Locality Latitude_Postcode Latitude_Primary_Number Longitude Longitude_Locality Longitude_Postcode Longitude_Primary_Number Other Population_Class_Locality1 Side_Of_Primary_Address Example 372 2011-06-09
- 373. Data Quality Input: You map input fields that contain the following data: Input field name Input value Country US Postcode1 54601 Postcode2 4023 Primary_Name1 Front Primary_Number 332 Primary_Type1 St. Output: The mapped output fields contain the following results: Output field name Output value Assignment_Level PRE Latitude 43.811616 Longitude -91.256695 Scenario 2 Scenario: Use an address and point-of-interest information to identify a list of potential point-of-interest matches. Number of output results: Multiple records. The number of records is determined by the Max_Records input field (if populated), or the Default Max Records option. The following sections describe the required input fields and available output fields to obtain results for this scenario. We also provide an example with sample data. Required input fields For required input fields, at least one input field in each category must be mapped. The Country field must be mapped. The more input data you can provide, the better results you will obtain. 373 2011-06-09
- 374. Data Quality Category Input field name Address Country (required) Locality1–4 Postcode1–2 Primary_Number Primary_Name1–4 Primary_Postfix1 Primary_Prefix1 Primary_Type1–4 Region1–2 Address POI POI_Name POI_Type Max Records Max_Records Optional input fields Not applicable. Available output fields All output fields are optional. Category Output field name Info Code Info_Code Result Result_List Result_List_Count Example The following example illustrates a scenario using an address and point-of-interest information to identify a list of potential point-of-interest matches. Input: You map input fields that contain the following data: 374 2011-06-09
- 375. Data Quality Input field name Input value Country US Postcode1 54601 Postcode2 4023 Primary_Number 332 Primary_Name1 Front Primary_Type1 St. POI_Name ABC Company POI_Type 5800 Max_Records 10 Output: The mapped output fields contain the following results with one record: Output field name Output value Result_List Output as XML; example shown below Result_List_Count 2 Result_List XML: The XML result for this example has one record. <RESULT_LIST> <RECORD> <ASSIGNMENT_LEVEL>PRE</ASSIGNMENT_LEVEL> <COUNTRY_CODE>US</COUNTRY_CODE> <DISTANCE>0.3340</DISTANCE> <LATITUDE>43.811616</LATITUDE> <LOCALITY1>LA CROSSE</LOCALITY1> <LONGITUDE>-91.256695</LONGITUDE> <POI_NAME>ABC COMPANY</POI_NAME> <POI_TYPE>5800</POI_TYPE> <POSTCODE1>56001</POSTCODE1> <PRIMARY_NAME1>FRONT</PRIMARY_NAME1> <PRIMARY_NUMBER>332</PRIMARY_NUMBER> <PRIMARY_TYPE1>ST</PRIMARY_TYPE1> <RANKING>1</RANKING> <REGION1>WI</REGION1> </RECORD> </RESULT_LIST> Related Topics • Understanding your output • Reference Guide: Data Quality fields, Geocoder fields, Input fields • Reference Guide: Data Quality fields, Geocoder fields, Output fields 375 2011-06-09
- 376. Data Quality 16.3.2 POI and address reverse geocoding Reverse geocoding lets you identify the closest address or point of interest based on an input reference location, which can be one of the following: • latitude and longitude • address • point of interest Mapping the optional radius input field lets you define the distance from the specified reference point and identify an area in which matching records are located. With reverse geocoding, you can find one or more locations that can be points of interest, addresses, or both by setting the Search_Filter_Name or Search_Filter_Type input field. This limits the output matches to your search criteria. To return an address only, enter ADDR in the Search_Filter_Type input field. To return a point of interest only, enter the point-of-interest name or type. If you don't set a search filter, the transform returns both addresses and points of interest. 16.3.2.1 Reverse geocoding scenarios Scenario 3 Scenario: Use latitude and longitude to find one or more addresses or points of interest. The following sections describe the required and optional input fields and available output fields to obtain either single-record or multiple-record results for this scenario. We also provide an example with sample data. Required input fields For a single-record result, both Latitude and Longitude input fields must be mapped. For multiple-record results, the Latitude, Longitude, and Max_Records input fields must all be mapped. 376 2011-06-09
- 377. Data Quality Single-record results Category Multiple-record results Input field name Input field name Latitude/Longitude Latitude Latitude Longitude Max Records Longitude n/a Max_Records Optional input fields Single-record results Input field name Radius Radius Search_Filter_Name Search_Filter_Type Search Filter Input field name Search_Filter_Name Category Multiple-record results Search_Filter_Type Available output fields All output fields are optional. 377 2011-06-09
- 378. Data Quality Single-record results Category Address Multiple-record results Input field name Input field name Country_Code n/a Locality1–4 POI_Name POI_Type Postcode1–2 Primary_Name1–4 Primary_Number Primary_Postfix1 Primary_Prefix1 Primary_Range_High Primary_Range_Low Primary_Type1–4 Region1–2 Assignment Level Assignment_Level n/a Assignment_Level_Locality Assignment_Level_Postcode Census Data 378 n/a 2011-06-09
- 379. Data Quality Single-record results Category Multiple-record results Input field name Input field name Census_Tract_Block Census_Tract_Block_Prev Census_Tract_Block_Group Census_Tract_Block_Group_Prev Gov_County_Code Gov_Locality1_Code Gov_Region1_Code Metro_Stat_Area_Code Metro_Stat_Area_Code_Prev Minor_Div_Code Minor_Div_Code_Prev Stat_Area_Code Stat_Area_Code_Prev Distance n/a Info Code 379 Distance Info_Code Info_Code 2011-06-09
- 380. Data Quality Single-record results Category Multiple-record results Input field name Input field name Latitude/Lon- Latitude gitude Latitude_Locality n/a Latitude_Postcode Latitude_Primary_Number Longitude Longitude_Locality Longitude_Postcode Longitude_Primary_Number Other Population_Class_Locality1 n/a Side_Of_Primary_Address Result n/a Result_List Result_List_Count Example The following example illustrates a scenario using latitude and longitude and a search filter to output a single point of interest closest to the input latitude and longitude. Input: You map input fields that contain the following data: Input field name Input value Latitude 43.811616 Longitude -91.256695 Search_Filter_Name ABC Company Output: The mapped output fields contain the following results: Output field name Assignment_Level PRE Country US Distance 1.3452 Locality1 LA CROSSE Postcode1 380 Output value 54601 2011-06-09
- 381. Data Quality Output field name Output value Postcode2 4023 Primary_Number 332 Primary_Name1 FRONT Primary_Type1 ST POI_Name ABC COMPANY POI_Type 5800 Region1 WI Scenario 4 Scenario: Use an address or point of interest to find one or more closest addresses or points of interest. In addition, the Geocoder transform outputs latitude and longitude information for both the input reference point and the matching output results. The following sections describe the required and optional input fields and available output fields to obtain either single-record or multiple-record results for this scenario. We also provide examples with sample data. Required input fields For required input fields, at least one input field in each category must be mapped. 381 2011-06-09
- 382. Data Quality Single-record results Multiple-record results Input field name Input field name Country Country Locality1–4 Locality1–4 Postcode1–2 Postcode1–2 Primary_Number Primary_Number Primary_Name1–4 Primary_Name1–4 Primary_Postfix1 Primary_Postfix1 Primary_Prefix1 Primary_Prefix1 Primary_Type1–4 Primary_Type1–4 Region1–2 Region1–2 Max Records n/a Max_Records Search Filter Radius Radius Search_Filter_Name Search_Filter_Name Search_Filter_Type Search_Filter_Type Category Address Optional input fields Single-record results Address POI Input field name Input field name POI_Name POI_Name POI_Type Category Multiple-record results POI_Type Available output fields All output fields are optional. For a single-record result, the output fields are the results for the spatial search. For multiple-record results, the output fields in the Assignment Level and Latitude/Longitude categories are the results for the reference address assignment. Output fields in the Results category are the results for the spatial search. For multiple-record results, the number of output records is determined by the Max_Records input field (if populated), or the Default Max Records option 382 2011-06-09
- 383. Data Quality Single-record results Category Address Multiple-record results Input field name Input field name Country_Code n/a Locality1–4 POI_Name POI_Type Postcode1–2 Primary_Name1–4 Primary_Number Primary_Postfix1 Primary_Prefix1 Primary_Range_High Primary_Range_Low Primary_Type1–4 Region1–2 AssignAssignment_Level ment LevAssignment_Level_Locality el Assignment_Level_Postcode Census Data 383 Assignment_Level n/a Assignment_Level_Locality Assignment_Level_Postcode 2011-06-09
- 384. Data Quality Single-record results Category Multiple-record results Input field name Input field name Census_Tract_Block Census_Tract_Block_Prev Census_Tract_Block_Group Census_Tract_Block_Group_Prev Gov_County_Code Gov_Locality1_Code Gov_Region1_Code Metro_Stat_Area_Code Metro_Stat_Area_Code_Prev Minor_Div_Code Minor_Div_Code_Prev Stat_Area_Code Stat_Area_Code_Prev Distance Distance Info Code Info_Code 384 n/a Info_Code 2011-06-09
- 385. Data Quality Single-record results Category Multiple-record results Input field name Input field name LatiLatitude tude/LonLatitude_Locality gitude Latitude_Postcode Latitude Latitude_Locality Latitude_Postcode Latitude_Primary_Number Longitude Longitude Longitude_Locality Longitude_Locality Longitude_Postcode Longitude_Postcode Longitude_Primary_Number Other Latitude_Primary_Number Longitude_Primary_Number Population_Class_Locality1 n/a Side_Of_Primary_Address Result n/a Result_List Result_List_Count Example 1 The following example illustrates a scenario using an address and a search filter to output a single point of interest closest to the input address. The transform also outputs latitude and longitude information for the output result. Input: You map input fields that contain the following data: Input field name Input value Country US Locality1 La Crosse Search_Filter_Name ABC Company Region1 WI Output: The mapped output fields contain the following results: 385 2011-06-09
- 386. Data Quality Output field name Output value Assignment_Level PRE Country US Distance 1.3046 Latitude 43.811616 Locality1 LA CROSSE Longitude -91.256695 POI_Name ABC Company POI_Type 5800 Postcode1 54601 Postcode2 4023 Primary_Name1 FRONT Primary_Number 332 Primary_Type1 ST Region1 WI Example 2 The following example illustrates a scenario using a point of address and a search filter to output a single address closest to the point of interest. The transform also outputs latitude and longitude information for the output result. Input: You map input fields that contain the following data: Input field name Input value Country US Locality1 La Crosse POI_Name ABC Company Region1 WI Search_Filter_Name ADDR Output: The mapped output fields contain the following results: 386 2011-06-09
- 387. Data Quality Output field name Output value Assignment_Level PRE Country US Distance 1.3023 Latitude 43.811616 Locality1 LA CROSSE Longitude -91.256695 Postcode1 54601 Postcode2 4023 Primary_Name1 FRONT Primary_Number 332 Primary_Type1 ST Region1 WI Related Topics • Understanding your output • Reference Guide: Data Quality fields, Geocoder fields, Input fields • Reference Guide: Data Quality fields, Geocoder fields, Output fields 16.3.3 Understanding your output Latitude and longitude On output from the Geocoder transform, you will have latitude and longitude data. Latitude and longitude are denoted on output by decimal degrees, for example, 12.12345. Latitude (0-90 degrees north or south of the equator) shows a negative sign in front of the output number when the location is south of the equator. Longitude (0-180 degrees east or west of Greenwich Meridian in London, England) shows a negative sign in front of the output number when the location is within 180 degrees west of Greenwich. Assignment level You can understand the accuracy of the assignment based on the Assignment_Level output field. The return code of PRE means that you have the finest depth of assignment available to the exact location. The second finest depth of assignment is a return code of PRI, which is the primary address range, or house number. The most general output level is either P1 (Postcode level) or L1 (Locality level), depending on the option you chose in the Best Assignment Level option. 387 2011-06-09
- 388. Data Quality Multiple results For multiple-record results, the Result_List output field is output as XML which can contain the following output, depending on the available data. Category Address Output field name Country_Code Locality1–4 POI_Name POI_Type Postcode1–2 Primary_Name1–4 Primary_Number Primary_Postfix1 Primary_Prefix1 Primary_Type1–4 Region1–2 Latitude/Longitude Latitude Latitude_Primary_Number Longitude Longitude_Primary_Number Ranking Ranking Standardize address information The geocoding data provided by vendors is not standardized. To standardize the address data that is output by the Geocoder transform, you can insert a Global Address Cleanse or USA Regulatory Address Cleanse transform in the data flow after the Geocoder transform. If you have set up the Geocoder transform to output multiple records, the address information in the XML output string must first be unnested before it can be cleansed. Related Topics • Reference Guide: Transforms, Geocoder options 388 2011-06-09
- 389. Data Quality 16.4 Match 16.4.1 Matching strategies Here are a few examples of strategies to help you think about how you want to approach the setup of your matching data flow. • Simple match. Use this strategy when your matching business rules consist of a single match criteria for identifying relationships in consumer, business, or product data. • Consumer Householding. Use this strategy when your matching business rules consist of multiple levels of consumer relationships, such as residential matches, family matches, and individual matches. • Corporate Householding. Use this strategy when your matching business rules consist of multiple levels of corporate relationships, such as corporate matches, subsidiary matches, and contact matches. • Multinational consumer match. Use this match strategy when your data consists of multiple countries and your matching business rules are different for different countries. • Identify a person multiple ways. Use this strategy when your matching business rules consist of multiple match criteria for identifying relationships, and you want to find the overlap between all of those definitions. See Association matching for more information. Think about the answers to these questions before deciding on a match strategy: • • What does my data consist of? (Customer data, international data, and so on) What fields do I want to compare? (last name, firm, and so on.) • What are the relative strengths and weaknesses of the data in those fields? Tip: You will get better results if you cleanse your data before matching. Also, data profiling can help you answer this question. • What end result do I want when the match job is complete? (One record per family, per firm, and so on.) 16.4.2 Match components 389 2011-06-09
- 390. Data Quality The basic components of matching are: • Match sets • Match levels • Match criteria Match sets A match set is represented by a Match transform on your workspace. Each match set can have its own break groups, match criteria, and prioritization. Match sets let you control how the Match transform matches certain records, segregate records, and match on records independently. For example, you could choose to match U.S. records differently than records containing international data. A match set has three purposes: • To allow only select data into a given set of match criteria for possible comparison (for example, exclude blank SSNs, international addresses, and so on). • To allow for related match scenarios to be stacked to create a multi-level match set. • To allow for multiple match sets to be considered for association in an Associate match set. Match levels A match level is an indicator to what type of matching will occur, such as on individual, family, resident, firm, and so on. A match level refers not to a specific criteria, but to the broad category of matching. You can have as many match levels as you want. However, the Match wizard restricts you to three levels during setup (more can be added later). You can define each match level in a match set in a way that is increasingly more strict. Multi-level matching feeds only the records that match from match level to match level (for example, resident, family, individual) for comparison. Match component Description Family The purpose of the family match type is to determine whether two people should be considered members of the same family, as reflected by their record data. The Match transform compares the last name and the address data. A match means that the two records represent members of the same family. The result of the match is one record per family. Individual The purpose of the individual match type is to determine whether two records are for the same person, as reflected by their record data. The Match transform compares the first name, last name, and address data. A match means that the two records represent the same person. The result of the match is one record per individual. 390 2011-06-09
- 391. Data Quality Match component Description Resident The purpose of the resident match type is to determine whether two records should be considered members of the same residence, as reflected by their record data. The Match transform compares the address data. A match means that the two records represent members of the same household. Contrast this match type with the family match type, which also compares last-name data. The result of the match is one record per residence. Firm The purpose of the firm match type is to determine whether two records reflect the same firm. This match type involves comparisons of firm and address data. A match means that the two records represent the same firm. The result of the match is one record per firm. Firm-Individual The purpose of the firm-individual match type is to determine whether two records are for the same person at the same firm, as reflected by their record data. With this match type, we compare the first name, last name, firm name, and address data. A match means that the two records reflect the same person at the same firm. The result of the match is one record per individual per firm. Match criteria Match criteria refers to the field you want to match on. You can use criteria options to specify business rules for matching on each of these fields. They allow you to control how close to exact the data needs to be for that data to be considered a match. For example, you may require first names to be at least 85% similar, but also allow a first name initial to match a spelled out first name, and allow a first name to match a middle name. • • • Family level match criteria may include family (last) name and address, or family (last) name and telephone number. Individual level match criteria may include full name and address, full name and SSN, or full name and e-mail address. Firm level match criteria may include firm name and address, firm name and Standard Industrial Classification (SIC) Code, or firm name and Data Universal Numbering System (DUNS) number. 16.4.3 Match Wizard 391 2011-06-09
- 392. Data Quality 16.4.3.1 Match wizard The Match wizard can quickly set up match data flows, without requiring you to manually create each individual transform it takes to complete the task. What the Match wizard does The Match wizard: • Builds all the necessary transforms to perform the match strategy you choose. • Applies default values to your match criteria based on the strategy you choose. • Places the resulting transforms on the workspace, connected to the upstream transform you choose. • Detects the appropriate upstream fields and maps to them automatically. What the Match wizard does not do The Match wizard provides you with a basic match setup that in some cases, will require customization to meet your business rules. The Match wizard: • Does not alter any data that flows through it. To correct non-standard entries or missing data, place one of the address cleansing transforms and a Data Cleanse transform upstream from the matching process. • Does not connect the generated match transforms to any downstream transform, such as a Loader. You are responsible for connecting these transforms. • Does not allow you to set rule-based or weighted scoring values for matching. The Match wizard incorporates a "best practices" standard that set these values for you. You may want to edit option values to conform to your business rules. Related Topics • Combination method 16.4.3.2 Before you begin Prepare a data flow for the Match wizard To maximize its usefulness, be sure to include the following in your data flow before you launch the Match wizard: 392 2011-06-09
- 393. Data Quality • Include a Reader in your data flow. You may want to match on a particular input field that our data cleansing transforms do not handle. • Include one of the address cleansing transforms and the Data Cleanse transform. The Match wizard works best if the data you're matching has already been cleansed and parsed into discrete fields upstream in the data flow. • If you want to match on any address fields, be sure that you pass them through the Data Cleanse transform. Otherwise, they will not be available to the Match transform (and Match Wizard). This rule is also true if you have the Data Cleanse transform before an address cleanse transform. 16.4.3.3 Use the Match Wizard 16.4.3.3.1 Select match strategy The Match wizard begins by prompting you to choose a match strategy, based on your business rule requirements. The path through the Match wizard depends on the strategy you select here. Use these descriptions to help you decide which strategy is best for you: • Simple match. Use this strategy when your matching business rules consist of a single match criteria for identifying relationships in consumer, business, or product data. • Consumer Householding. Use this strategy when your matching business rules consist of multiple levels of consumer relationships, such as residential matches, family matches, and individual matches. • Corporate Householding. Use this strategy when your matching business rules consist of multiple levels of corporate relationships, such as corporate matches, subsidiary matches, and contact matches. • Multinational consumer match. Use this match strategy when your data consists of multiple countries and your matching business rules are different for different countries. Note: The multinational consumer match strategy sets up a data flow that expects Latin1 data. If you want to use Unicode matching, you must edit your data flow after it has been created. • Identify a person multiple ways. Use this strategy when your matching business rules consist of multiple match criteria for identifying relationships, and you want to find the overlap between all of those definitions. Source statistics If you want to generate source statistics for reports, make sure a field that houses the physical source value exists in all of the data sources. To generate source statistics for your match reports, select the Generate statistics for your sources checkbox, and then select a field that contains your physical source value. 393 2011-06-09
- 394. Data Quality Related Topics • Unicode matching • Association matching 16.4.3.3.2 Identify matching criteria Criteria represent the data that you want to use to help determine matches. In this window, you will define these criteria for each match set that you are using. Match sets compare data to find similar records, working independently within each break group that you designate (later in the Match wizard). The records in one break group are not compared against those in any other break group. To find the data that matches all the fields, use a single match set with multiple fields. To find the data that matches only in a specific combination of fields, use multiple match sets with two fields. When working on student or snowbird data, an individual may use the same name but have multiple valid addresses. Select a combination of fields that best shows which information overlaps, such as the family name and the SSN. Data1 Data2 Data3 Data4 R. Carson 1239 Whistle Lane Columbus, Ohio 555-23-4333 Robert T. Carson 52 Sunbird Suites Tampa, Florida 555-23-4333 1. Enter the number of ways you have to identify an individual. This produces the corresponding number of match sets (transforms) in the data flow. 2. The default match set name appears in the Name field. Select a match set in the Match sets list, and enter a more descriptive name if necessary. 3. For each match set, choose the criteria you want to match on. Later, you will assign fields from upstream transforms to these criteria. 4. Select the option you want to use for comparison in the Compare using column. The options vary depending on the criteria chosen. The compare options are: • Field similarity • Word similarity • Numeric difference • Numeric percent difference • Geo proximity 5. Optional: If you choose to match on Custom, enter a name for the custom criteria in the Custom name column. 394 2011-06-09
- 395. Data Quality 6. Optional: If you choose to match on Custom, specify how close the data must be for that criteria in two records to be considered a match. The values that result determine how similar you expect the data to be during the comparison process for this criteria only. After selecting a strategy, you may change the values for any of the comparison rules options in order to meet your specific matching requirements.Select one of the following from the list in the Custom exactness column: • Exact: Data in this criteria must be exactly the same; no variation in the data is allowed. • Tight: Data in this criteria must have a high level of similarity; a small amount of variation in the data is allowed. • Medium: Data in this criteria may have a medium level of similarity; a medium amount of variation in the data is allowed. • Loose: Data in this criteria may have a lower level of similarity; a greater amount of variation in the data is allowed. 16.4.3.3.3 Define match levels Match levels allow matching processes to be defined at distinct levels that are logically related. Match levels refer to the broad category of matching not the specific rules of matching. For instance, a residencelevel match would match on only address elements, a family-level would match on only Last Name and then the individual-level would match on First Name. Multi-level matching can contain up to 3 levels within a single match set defined in a way that is increasingly more strict. Multi-level matching feeds only the records that match from match level to match level (that is, resident, family, individual) for comparison. To define match levels: 1. Click the top level match, and enter a name for the level, if you don't want to keep the default name. The default criteria is already selected. If you do not want to use the default criteria, click to remove the check mark from the box. The default criteria selection is a good place to start when choosing criteria. You can add criteria for each level to help make finer or more precise matches. 2. Select any additional criteria for this level. 3. If you want to use criteria other than those offered, click Custom and then select the desired criteria. 4. Continue until you have populated all the levels that you require. 16.4.3.3.4 Select countries Select the countries whose postal standards may be required to effectively compare the incoming data. The left panel shows a list of all available countries. The right panel shows the countries you already selected. 1. Select the country name in the All Countries list. 2. Click Add to move it into the Selected Countries list. 3. Repeat steps 1 and 2 for each country that you want to include. You can also select multiple countries and add them all by clicking the Add button. The countries that you select are remembered for the next Match wizard session. 395 2011-06-09
- 396. Data Quality 16.4.3.3.5 Group countries into tracks Create tracks to group countries into logical combinations based on your business rules (for example Asia, Europe, South America). Each track creates up to six match sets (Match transforms). 1. Select the number of tracks that you want to create. The Tracks list reflects the number of tracks you choose and assigns a track number for each. 2. To create each track, select a track title, such as Track1. 3. Select the countries that you want in that track. 4. Click Add to move the selected countries to the selected track. Use the COUNTRY UNKNOWN (__) listing for data where the country of origin has not been identified. Use the COUNTRY OTHER (--) listing for data whose country of origin has been identified, but the country does not exist in the list of selected countries. 5. From Match engines, select one of the following engines for each track: Note: All match transforms generated for the track will use the selected Match engine. • • • • • • LATIN1 (Default) CHINESE JAPANESE KOREAN TAIWANESE OTHER_NON_LATIN1 The Next button is only enabled when all tracks have an entry and all countries are assigned to a track. 16.4.3.3.6 Select criteria fields Select and deselect criteria fields for each match set and match level you create in your data flow. These selections determine which fields are compared for each record. Some criteria may be selected by default, based on the data input. If there is only one field of the appropriate content type, you will not be able to change the field for that criteria within the Match Wizard. To enable the Next button, you must select at least one non-match-standard field. 1. For each of the criteria fields you want to include, select an available field from the drop-down list, which contains fields from upstream source(s). The available fields are limited to the appropriate content types for that criteria. If no fields of the appropriate type are available, all upstream fields display in the menu. 2. Optional: Deselect any criteria fields you do not want to include. 396 2011-06-09
- 397. Data Quality 16.4.3.3.7 Create break keys Use break keys to create manageable groups of data to compare. The match set compares the data in the records within each break group only, not across the groups. Making the correct selections can save valuable processing time by preventing widely divergent data from being compared. Break keys are especially important when you deal with large amounts of data, because the size of the break groups can affect processing time. Even if your data is not extensive, break groups will help to speed up processing. Create break keys that group similar data that would most likely contain matches. Keep in mind that records in one break group will not be compared against records in any other break group. For example, when you match to find duplicate addresses, base the break key on the postcode, city, or state to create groups with the most likely matches. When you match to find duplicate individuals, base the break key on the postcode and a portion of the name as the most likely point of match. To create break keys: 1. In the How many fields column, select the number of fields to include in the break key. 2. For each break key, select the following: • the field(s) in the break key • the starting point for each field • the number of positions to read from each field 3. After you define the break keys, do one of the following: • Click Finish. This completes the match transform. • If you are performing multi-national matching, click Next to go to the Matching Criteria page. 16.4.3.4 After setup Although the Match wizard does a lot of the work, there are some things that you must do to have a runnable match job. There are also some things you want to do to refine your matching process. Connect to downstream transforms When the Match wizard is complete, it places the generated transforms on the workspace, connected to the upstream transform you selected to start the Match wizard. For your job to run, you must connect each port from the last transform to a downstream transform. To do this, click a port and drag to connect to the desired object. View and edit the new match transform To see what is incorporated in the transform(s) the Match Wizard produces, right-click the transform and choose Match Editor. 397 2011-06-09
- 398. Data Quality View and edit Associate transforms To see what is incorporated in the Associate transform(s) the Match Wizard produces, right-click the transform and choose Associate Editor. Multinational matching For the Multinational consumer match strategy, the wizard builds as many Match transforms as you specify in the Define Sets window of the wizard for each track you create. Caution: If you delete any tracks from the workspace after the wizard builds them, you must open the Case transform and delete any unwanted rules. Related Topics • Unicode matching 16.4.4 Transforms for match data flows The Match and Associate transforms are the primary transforms involved in setting up matching in a data flow. These transforms perform the basic matching functions. There are also other transforms that can be used for specific purposes to optimize matching. Trans form Case Usage Routes data to a particular Match transform (match set). A common usage for this transform is to send USA-specific and international-specific data to different transforms. You can also use this transform to route blank records around a Match transform. Merge Performs the following functions: • Brings together data from Match transforms for Association matching. • Brings together matching records and blank records after being split by a Case transform. Query Creates fields, performs functions to help prepare data for matching, orders data, and so on. Example: Any time you need to bypass records from a particular match process (usually in Associative data flows and any time you want to have records with blank data to bypass a match process) you will use the Case, Query, and Merge transforms. 398 2011-06-09
- 399. Data Quality • • • The Case transform has two routes: one route sends all records that meet the criteria to the Match transform, and one that sends all other records to the bypass match route. The Query transform adds the fields that the Match transform generates and you output. (The output schema in the Match transform and the output schema in the Query transform must be identical for them to be merged.) The contents of the newly added fields in the Query transform may be populated with an empty string. The Merge transform merges the two routes into a single route. 16.4.4.1 To remove matching from the Match transform You may want to place a transform that employs some of the functionality of a Match transform in your data flow, but does not include the actual matching features. For example, you may want to do candidate selection or prioritization in a data flow or a location in a data flow. that doesn't do matching at all. 1. Right-click the Match transform in the object library, and choose New. 2. In the Format name field, enter a meaningful name for your transform. It's helpful to indicate which type of function this transform will be performing. 3. Click OK. 4. Drag and drop your new Match transform configuration onto the workspace and connect it to your data flow. 5. Right-click the new transform, and choose Match Editor. 6. Deselect the Perform matching option in the upper left corner of the Match editor. Now you can add any available operation to this transform. 16.4.5 Working in the Match and Associate editors Editors The Match and Associate transform editors allow you to set up your input and output schemas. You can access these editors by double-clicking the appropriate transform icon on your workspace. 399 2011-06-09
- 400. Data Quality The Match and Associate editors allow you to configure your transform's options. You can access these editors by right-clicking the appropriate transform and choosing Match Editor (or Associate Editor). Order of setup Remember: The order that you set up your Match transform is important! First, it is best to map your input fields. If you don't, and you add an operation in the Match editor, you may not see a particular field you want to use for that operation. Secondly, you should configure your options in the Match editor before you map your output fields. Adding operations to the Match transform (such as Unique ID and Group Statistics) can provide you with useful Match transform-generated fields that you may want to use later in the data flow or add to your database. Example: 1. Map your input fields. 2. Configure the options for the transform. 3. Map your output fields. 16.4.6 Physical and logical sources Tracking your input data sources and other sources, whether based on an input source or based on some data element in the rows being read, throughout the data flow is essential for producing informative match reports. Depending on what you are tracking, you must create the appropriate fields in your data flow to ensure that the software generates the statistics you want, if you don't already have them in your database. • • Physical source: The filename or value attributed to the source of the input data. Logical source: A group of records spanning multiple input sources or a subset of records from a single input source. Physical input sources You track your input data source by assigning that physical source a value in a field. Then you will use this field in the transforms where report statistics are generated. To assign this value, add a Query transform after the source and add a column with a constant containing the name you want to assign to this source. Note: If your source is a flat file, you can use the Include file name option to automatically generate a column containing the file name. 400 2011-06-09
- 401. Data Quality Logical input sources If you want to count source statistics in the Match transform (for the Match Source Statistics Summary report, for example), you must create a field using a Query transform or a User-Defined transform, if you don't already have one in your input data sources. This field tracks the various sources within a Reader for reporting purposes, and is used in the Group Statistics operation of the Match transform to generate the source statistics. It is also used in compare tables, so that you can specify which sources to compare. 16.4.6.1 Using sources A source is the grouping of records on the basis of some data characteristic that you can identify. A source might be all records from one input file, or all records that contain a particular value in a particular field. Sources are abstract and arbitrary—there is no physical boundary line between sources. Source membership can cut across input files or database records as well as distinguish among records within a file or database, based on how you define the source. If you are willing to treat all your input records as normal, eligible records with equal priority, then you do not need to include sources in your job. Typically, a match user expects some characteristic or combination of characteristics to be significant, either for selecting the best matching record, or for deciding which records to include or exclude from a mailing list, for example. Sources enable you to attach those characteristics to a record, by virtue of that record’s membership in its particular source. Before getting to the details about how to set up and use sources, here are some of the many reasons you might want to include sources in your job: • • • • • • 401 To give one set of records priority over others. For example, you might want to give the records of your house database or a suppression source priority over the records from an update file. To identify a set of records that match suppression sources, such as the DMA. To set up a set of records that should not be counted toward multi-source status. For example, some mailers use a seed source of potential buyers who report back to the mailer when they receive a mail piece so that the mailer can measure delivery. These are special-type records. To save processing time, by canceling the comparison within a set of records that you know contains no matching records. In this case, you must know that there are no matching records within the source, but there may be matches among sources. To save processing time, you could set up sources and cancel comparing within each source. To get separate report statistics for a set of records within an source, or to get report statistics for groups of sources. To protect a source from having its data overwritten by a best record or unique ID operation. You can choose to protect data based on membership in a source. 2011-06-09
- 402. Data Quality 16.4.6.2 Source types You can identify each source as one of three different types: Normal, Suppression, or Special. The software can process your records differently depending on their source type. Source Description Normal A Normal source is a group of records considered to be good, eligible records. Suppress A Suppress source contains records that would often disqualify a record from use. For example, if you’re using Match to refine a mailing source, a suppress source can help remove records from the mailing. Examples: • • • • Special DMA Mail Preference File American Correctional Association prisons/jails sources No pandering or non-responder sources Credit card or bad-check suppression sources A Special source is treated like a Normal source, with one exception. A Special source is not counted in when determining whether a match group is singlesource or multi-source. A Special source can contribute records, but it’s not counted toward multi-source status. For example, some companies use a source of seed names. These are names of people who report when they receive advertising mail, so that the mailer can measure mail delivery. Appearance on the seed source is not counted toward multi-source status. The reason for identifying the source type is to set that identity for each of the records that are members of the source. Source type plays an important role in controling priority (order) of records in break group, how the software processes matching records (the members of match groups), and how the software produces output (that is, whether it includes or excludes a record from its output). 16.4.6.2.1 To manually define input sources Once you have mapped in an input field that contains the source values, you can create your sources in the Match Editor. 1. In the Match Editor, select Transform Options in the explorer pane on the left, click the Add button, and select Input Sources. The new Input Sources operation appears under Transform Options in the explorer pane. Select it to view Input Source options. 2. In the Value field drop-down list, choose the field that contains the input source value. 402 2011-06-09
- 403. Data Quality 3. In the Define sources table, create a source name, type a source value that exists in the Value field for that source, and choose a source type. 4. Choose value from the Default source name option. This name will be used for any record whose source field value is blank. Be sure to click the Apply button to save any changes you have made, before you move to another operation in the Match Editor. 16.4.6.2.2 To automatically define input sources To avoid manually defining your input sources, you can choose to do it automatically by choosing the Auto generate sources option in the Input Sources operation. 1. In the Match Editor, select Transform Options in the explorer pane on the left, click the Add button, and select Input Sources. The new Input Sources operation appears under Transform Options in the explorer pane. Select it to view Input Source options. 2. In the Value field drop-down list, choose the field that contains the input source value. 3. Choose value from the Default source name option. This name will be used for any record whose source field value is blank. 4. Select the Auto generate sources option. 5. Choose a value in the Default type option The default type will be assigned to to any source that does not already have the type defined in the Type field. 6. Select a field from the drop-down list in the Type field option. Auto generating sources will create a source for each unique value in the Value field. Any records that do not have a value field defined will be assigned to the default source name. 16.4.6.3 Source groups The source group capability adds a higher level of source management. For example, suppose you rented several files from two brokers. You define five sources to be used in ranking the records. In addition, you would like to see your job’s statistics broken down by broker as well as by file. To do this, you can define groups of sources for each broker. Source groups primarily affect reports. However, you can also use source groups to select multi-source records based on the number of source groups in which a name occurs. Remember that you cannot use source groups in the same way you use sources. For example, you cannot give one source group priority over another. 403 2011-06-09
- 404. Data Quality 16.4.6.3.1 To create source groups You must have input sources in an Input Source operation defined to be able to add this operation or define your source groups. 1. Select a Match transform in your data flow, and choose Tools > Match Editor. 2. In the Match Editor, select Transform Options in the explorer pane on the left, click the Add button, and select Source Groups. The new Source Groups operation appears under Input Sources operation in the explorer pane. Select it to view Source Group options. 3. Confirm that the input sources you need are in the Sources column on the right. 4. Double-click the first row in the Source Groups column on the left, and enter a name for your first source group, and press Enter. 5. Select a source in the Sources column and click the Add button. 6. Choose a value for the Undefined action option. This option specifies the action to take if an input source does not appear in a source group. 7. If you chose Default as the undefined action in the previous step, you must choose a value in the Default source group option. This option is populated with source groups you have already defined. If an input source is not assigned to a source group, it will be assigned to this default source group. 8. If you want, select a field in the Source group field option drop-down list that contains the value for your source groups. 16.4.7 Match preparation 16.4.7.1 Prepare data for matching Data correction and standardization Accurate matches depend on good data coming into the Match transform. For batch matching, we always recommend that you include one of the address cleansing transforms and a Data Cleanse transform in your data flow before you attempt matching. Filter out empty records You should filter out empty records before matching. This should help performance. Use a Case transform to route records to a different path or a Query transform to filter or block records. 404 2011-06-09
- 405. Data Quality Noise words You can perform a search and replace on words that are meaningless to the matching process. For matching on firm data, words such as Inc., Corp., and Ltd. can be removed. You can use the search and replace function in the Query transform to accomplish this. Break groups Break groups organize records into collections that are potential matches, thus reducing the number of comparisons that the Match transform must perform. Include a Break Group operation in your Match transform to improve performance. Match standards You may want to include variations of name or firm data in the matching process to help ensure a match. For example, a variation of Bill might be William. When making comparisons, you may want to use the original data and one or more variations. You can add anywhere from one to five variations or match standards, depending on the type of data. For example, If the first names are compared but don't match, the variations are then compared. If the variations match, the two records still have a chance of matching rather than failing, because the original first names were not considered a match. Custom Match Standards You can match on custom Data Cleanse output fields and associated aliases. Map the custom output fields from Data Cleanse and the custom fields will appear in the Match Editor's Criteria Fields tab. 16.4.7.1.1 Fields to include for matching To take advantage of the wide range of features in the Match transform, you will need to map a number of input fields, other than the ones that you want to use as match criteria. Example: Here are some of the other fields that you might want to include. The names of the fields are not important, as long as you remember which field contains the appropriate data. Field contents Contains... Logical source A value that specifies which logical source a record originated. This field is used in the Group Statistics operation, compare tables, and also the Associate transform. Physical source A value that specifies which physical source a record originated. (For example, a source object, or a group of candidate-selected records) This field is used in the Match transform options, Candidate Selection operation, and the Associate transform. Break keys A field that contains the break key value for creating break groups. Including a field that already contains the break key value could help improve the performance of break group creation, because it will save the Match transform from doing the parsing of multiple fields to create the break key. 405 2011-06-09
- 406. Data Quality Field contents Contains... Criteria fields The fields that contain the data you want to match on. Count flags A Yes or No value to specify whether a logical source should be counted in a Group Statistics operation. Record priority A value that is used to signify a record as having priority over another when ordering records. This field is used in Group Prioritization operations. Apply blank penalty A Yes or No value to specify whether Match should apply a blank penalty to a record. This field is used in Group Prioritization operations. Starting unique ID value A starting ID value that will then increment by 1 every time a unique ID is assigned. This field is used in the Unique ID operation. This is not a complete list. Depending on the features you want to use, you may want to include many other fields that will be used in the Match transform. 16.4.7.2 Control record comparisons Controlling the number of record comparisons in the matching process is important for a couple of reasons: • Speed. By controlling the actual number of comparisons, you can save processing time. • Match quality. By grouping together only those records that have a potential to match, you are assured of better results in your matching process. Controlling the number of comparisons is primarily done in the Group Forming section of the Match editor with the following operations: • • Break group: Break up your records into smaller groups of records that are more likely to match. Candidate selection: Select only match candidates from a database table. This is primarily used for real-time jobs. You can also use compare tables to include or exclude records for comparison by logical source. Related Topics • Break groups • Candidate selection • Compare tables 406 2011-06-09
- 407. Data Quality 16.4.7.2.1 Break groups When you create break groups, you place records into groups that are likely to match. For example, a common scenario is to create break groups based on a postcode. This ensures that records from different postcodes will never be compared, because the chances of finding a matching record with a different postcode are very small. Break keys You form break groups by creating a break key: a field that consists of parts of other fields or a single field, which is then used to group together records based on similar data. Here is an example of a typical break key created by combining the five digits of the Postcode1 field and the first three characters of the Address_Primary_Name field. Field (Start pos:length) Data in field Postcode1 (1:5) 10101 Address_Primary_Name (1:3) Generated break key Main 10101Mai All records that match the generated break key in this example are placed in the same break group and compared against one another. Sorting of records in the break group Records are sorted on the break key field. You can add a Group Prioritization operation after the Break Groups operation to specify which records you want to be the drivers. Remember: Order is important! If you are creating break groups using records from a Suppress-type source, be sure that the suppression records are the drivers in the break group. Break group anatomy Break groups consist of driver and passenger records. The driver record is the first record in the break group, and all other records are passengers. The driver record is the record that drives the comparison process in matching. The driver is compared to all of the passengers first. This example is based on a break key that uses the first three digits of the Postcode. 407 2011-06-09
- 408. Data Quality Phonetic break keys You can also use the Soundex and Double Metaphone functions to create fields containing phonetic codes, which can then be used to form break groups for matching. Related Topics • Phonetic matching • Management Console Guide: Data Quality Reports, Match Contribution report To create break groups We recommend that you standardize your data before you create your break keys. Data can be treated differently that is inconsistently cased, for example. 1. Add a Break Groups operation to the Group Forming option group. 2. in the Break key table, add a row by clicking the Add button. 3. Select a field in the field column that you want to use as a break key. Postcode is a common break key to use. 4. Choose the start position and length (number of characters) you want used in your break key. You can use negative integers to signify that you want to start at the end of the actual string length, not the specified length of the field. For example, Field(-3,3) takes the last 3 characters of the string, whether the string has length of 10 or a length of 5. 5. Add more rows and fields as necessary. 6. Order your rows by selecting a row and clicking the Move Up and Move Down buttons. Ordering your rows ensures that the fields are used in the right order in the break key. Your break key is now created. 408 2011-06-09
- 409. Data Quality 16.4.7.2.2 Candidate selection To speed processing in a match job, use the Candidate Selection operaton (Group forming option group) in the Match transform to append records from a relational database to an existing data collection before matching. When the records are appended, they are not logically grouped in any way. They are simply appended to the end of the data collection on a record-by-record basis until the collection reaches the specified size. For example, suppose you have a new source of records that you want to compare against your data warehouse in a batch job. From this warehouse, you can select records that match the break keys of the new source. This helps narrow down the number of comparisons the Match transform has to make. For example, here is a simplified illustration: Suppose your job is comparing a new source database—a smaller, regional file—with a large, national database that includes 15 records in each of 43,000 or so postcodes. Further assume that you want to form break groups based only on the postcode. Notes Regional National Total Without candidate selection, the Match transform reads all of the records of both databases. 1,500 750,000 751,500 With candidate selection, only those records that would be included in a break group are read. 1,500 About 600 (40 x 15) 2,100 Datastores and candidate selection To use candidate selection, you must connect to a valid datastore. You can connect to any SQL-based or persistent cache datastore. There are advantages for using one over the other, depending on whether your secondary source is static (it isn't updated often) or dynamic (the source is updated often). Persistent cache datastores Persistent cache is like any other datastore from which you can load your candidate set. If the secondary source from which you do candidate selection is fairly static (that is, it will not change often), then you 409 2011-06-09
- 410. Data Quality might want consider building a persistent cache, rather than using your secondary source directly, to use as your secondary table. You may improve performance. You may also encounter performance gains by using a flat file (a more easily searchable format than a RDBMS) for your persistent cache. If the secondary source is not an RDBMS, such as a flat file, you cannot use it as a "datastore". In this case, you can create a persistent cache out of that flat file source and then use that for candidate selection. Note: A persistent cache used in candidate selection must be created by a dataflow in double-byte mode. To do this, you will need to change the locale setting in the Data Services Locale Selector (set the code page to utf-8). Run the job to generate the persistent cache, and then you can change the code page back to its original setting if you want. Cache size Performance gains using persistent cache also depend on the size of the secondary source data. As the size of the data loaded in the persistent cache increases, the performance gains may decrease. Also note that if the original secondary source table is properly indexed and optimized for speed then there may be no benefit in creating a persistent cache (or even pre-load cache) out of it. Related Topics • Persistent cache datastores Auto-generation vs. custom SQL There are cases where the Match transform can generate SQL for you, and there are times where you must create your own SQL. This is determined by the options you and how your secondary table (the table you are selecting match candidates from) is set up. Use this table to help you determine whether you can use auto-generated SQL or if you must create your own. Note: In the following scenarios, “input data” refers to break key fields coming from a transform upstream from the Match transform (such as a Query transform) or a break key fields coming from the Break Group operation within the Match transform itself. Scenario You have a single break key field in your input data, and you have the same field in your secondary table. Auto-generate You have multiple break key fields in your input data, and you have the same fields in your secondary table. Auto-generate You have multiple break key fields in your input data, and you have one break key field in your secondary table. 410 Auto-generate or Custom? Auto-generate 2011-06-09
- 411. Data Quality Scenario Auto-generate or Custom? You have a single break key field in your input data, and you have multiple break key fields in your secondary table. Custom You have multiple break key fields in your input data, but you have a different format or number of fields in your secondary table. Custom You want to select from multiple input sources. Custom Break keys and candidate selection We recommend that you create a break key column in your secondary table (the table that contains the records you want to compare with the input data in your data flow) that matches the break key you create your break groups with in the Match transform. This makes setup of the Candidate Selection operation much easier. Also, each of these columns should be indexed. We also recommend that you create and populate the database you are selecting from with a single break key field, rather than pulling substrings from database fields to create your break key. This can help improve the performance of candidate selection. Note: Records extracted by candidate selection are appended to the end of an existing break group (if you are using break groups). So, if you do not reorder the records using a Group Prioritization operation after the Candidate Selection operation, records from the original source will always be the driver records in the break groups. If you are using candidate selection on a Suppress source, you will need to reorder the records so that the records from the Suppress source are the drivers. To set up candidate selection If you are using Candidate selection for a real-time job, be sure to deselect the Split records into break groups option in the Break Group operation of the Match transform. To speed processing in a real-time match job, use the Candidate Selection operaton (Group forming option group) in the Match transform to append records from a relational database to an existing data collection before matching. When the records are appended, they are not logically grouped in any way. They are simply appended to the end of the data collection on a record-by-record basis until the collection reaches the specified size. 1. In the Candidate Selection operation, select a valid datastore from the Datastore drop-down list. 2. In the Cache type drop-down list, choose from the following values: 411 2011-06-09
- 412. Data Quality Option Description No_Cache Captures data at a point in time. The data doesn't change until the job restarts. Pre-load Cache Use this option for static data. 3. Depending on how your input data and secondary table are structured, do one of the following: • Select Auto-generate SQL. Then select the Use break column from database option, if you have one, and choose a column from the Break key field drop-down list. Note: If you choose the Auto-generate SQL option, we recommend that you have a break key column in your secondary table and select the Use break column from database option. If you don't, the SQL that is created could be incorrect. • Select Create custom SQL, and either click the Launch SQL Editor button or type your SQL in the SQL edit box. 4. If you want to track your records from the input source, select Use constant source value. 5. Enter a value that represents your source in the Physical source value option, and then choose a field that holds this value in the Physical source field drop-down list. 6. In the Column mapping table, add as many rows as you want. Each row is a field that will be added to the collection. a. Choose a field in the Mapped name column. b. Choose a column from your secondary table (or from a custom query) in the Column name option that contains the same type of data as specified in the Mapped name column. If you have already defined your break keys in the Break Group option group, the fields used to create the break key are posted here, with the Break Group column set to YES. Writing custom SQL Use placeholders To avoid complicated SQL statements, you should use placeholders (which are replaced with real input data) in your WHERE clause. For example, let's say the customer database contains a field called MatchKey, and the record that goes through the cleansing process gets a field generated called MATCH_KEY. This field has a placeholder of [MATCHKEY]. The records that are selected from the customer database and appended to the existing data collection are those that contain the same value in MatchKey as in the transaction's MATCH_KEY. For this example, let's say the actual value is a 10-digit phone number. The following is an example of what your SQL would look like with an actual phone number instead of the [MATCHKEY] placeholder. SELECT ContactGivenName1, ContactGivenName2, ContactFamilyName, Address1, Address2, City, Region, Postcode, Country, AddrStreet, AddrStreetNumber, AddrUnitNumber FROM TblCustomer 412 2011-06-09
- 413. Data Quality WHERE MatchKey = '123-555-9876'; Caution: You must make sure that the SQL statement is optimized for best performance and will generate valid results. The Candidate Selection operation does not do this for you. Replace placeholder with actual values After testing the SQL with actual values, you must replace the actual values with placeholders ([MATCHKEY], for example). Your SQL should now look similar to the following. SELECT ContactGivenName1, ContactGivenName2, ContactFamilyName, Address1, Address2, City, Region, Postcode, Country, AddrStreet, AddrStreetNumber, AddrUnitNumber FROM TblCustomer WHERE MatchKey = [MATCHKEY]; Note: Placeholders cannot be used for list values, for example in an IN clause: WHERE status IN ([status]) If [status] is a list of values, this SQL statement will fail. 16.4.7.2.3 Compare tables Compare tables are sets of rules that define which records to compare, sort of an additional way to create break groups. You use your logical source values to determine which records are compared or are not compared. By using compare tables, you can compare records within sources, or you can compare records across sources, or a combination of both. To set up a compare table Be sure to include a field that contains a logical source value before you add a Compare table operation to the Match transform (in the Match level option group). Here is an example of how to set up your compare table. Suppose you have two IDs (A and B), and you only want to compare across sources, not within the sources. 1. If no Compare Table is present in the Matching section, right-click Matching > <Level Name>, and select Add > Compare. 2. Set the Default action option to No_Match, and type None in the Default logical source value option. This tells the Match transform to not compare everything, but follow the comparison rules set by the table entries. 413 2011-06-09
- 414. Data Quality Note: Use care when choosing logical source names. Typing “None” in the Default logical source value option will not work if you have a source ID called “None.” 3. In the Compare actions table, add a row, and then set the Driver value to A, and set the Passenger value to B. 4. Set Action to Compare. Note: Account for all logical source values. The example values entered above assumes that A will always be the driver ID. If you expect that a driver record has a value other than A, set up a table entry to account for that value and the passenger ID value. Remember that the driver record is the first record read in a collection. If you leave the Driver value or Passenger value options blank in the compare table, then it will mean that you want to compare all sources. So a Driver value of A and a blank passenger record with an action of compare will make a record from A compare against all other passenger records. Sometimes data in collections can be ordered (or not ordered, as the case may be) differently than your compare table is expecting. This can cause the matching process to miss duplicate records. In the example, the way you set up your Compare action table row means that you are expecting that the driver record should have a driver value of A, but if the driver record comes in with a value of B, and the passenger comes in with a value of A, it won't be compared. To account for situations where a driver record might have a value of B and the passenger a value of A, for example, include another row in the table that does the opposite. This will make sure that any record with a value of A or B is compared, no matter which is the Driver or Passenger. Note: In general, if you use a suppress source, you should compare within the other sources.This ensures that all of the matches of those sources are suppressed when any are found to duplicate a record on the suppress source, regardless of which record is the driver record. 16.4.7.3 Order and prioritize records You may have data sources, such as your own data warehouse, that you might trust more than records from another source, such as a rented source, for example. You may also prefer newer records over older records, or more complete records over those with blank fields. Whatever your preference, the way to express this preference in the matching process is using priorities. There are other times where you might want to ensure that your records move to a given operation, such as matching or best record, for example, in a particular order. For example, you might want your match groups to be ordered so that the first record in is the newest record of the group. In this case, you would want to order your records based on a date field. 414 2011-06-09
- 415. Data Quality Whatever the reason, there are a two ways to order your records, either before or after the comparison process: • • Sorting records in break groups or match groups using a value in a field Using penalty scores. These can be defined per field, per record, or based on input source membership. Match editor You can define your priorities and order your records in the Group Prioritization operation, available in Group Forming and in the Post-match processing operations of each match level in the Match editor. Types of priorities There are a couple of different types of priorities to consider: Priority Brief description Record priority Prefers records from one input source over another. Blank penalty Assigns a lower priority to records in which a particular field is blank. Pre-match ordering When you create break groups, you can set up your Group Forming > Group Prioritization operation to order (or sort) on a field, besides ordering on the break key. This will ensure that the highest priority record is the first record (driver) in the break group. You will also want to have Suppress-type input sources to be the driver records in a break group. Post-match ordering After the Match transform has created all of the match groups, and if order is important, you can use a Group Prioritization operation before a Group Statistics, Best Record, and Unique ID operations to ensure that the master record is the first in the match group. Tip: If you are not using a blank penalty, order may not be as important to you, and you may not want to include a Group Prioritization operation before your post-match operations. However, you may get better performance out of a Best Record operation by prioritizing records and then setting the Post only once per destination option to Yes. Blank penalty Given two records, you may prefer to keep the record that contains the most complete data. You can use blank penalty to penalize records that contain blank fields. Incorporating a blank penalty is appropriate if you feel that a blank field shouldn't disqualify one record from matching another, and you want to keep the more complete record. For example, suppose you are willing to accept a record as a match even if the Prename, Given_Name1, Given_Name2, 415 2011-06-09
- 416. Data Quality Primary_Postfix and/or Secondary Number is blank. Even though you accept these records into your match groups, you can assign them a lower priority for each blank field. 16.4.7.3.1 To order records by sorting on a field Be sure you have mapped the input fields into the Match transform that you want to order on, or they won't show up in the field drop-down list. Use this method of ordering your records if you do not consider completeness of data important. 1. Enter a Prioritization name, and select the Priority Order tab. 2. In the Priority fields table, choose a field from the drop-down list in the Input Fields column. 3. In the Field Order column, choose Ascending or Descending to specify the type of ordering. For example, if you are comparing a Normal source to a Suppress source and you are using a source ID field to order your records, you will want to ensure that records from the Suppress source are first in the break group. 4. Repeat step 2 for each row you added. 5. Order your rows in the Priority fields table by using the Move Up and Move Down buttons. The first row will be the primary order, and the rest will be secondary orders. 16.4.7.3.2 Penalty scoring system The blank penalty is a penalty-scoring system. For each blank field, you can assess a penalty of any non-negative integer. You can assess the same penalty for each blank field, or assess a higher penalty for fields you consider more important. For example, if you were targeting a mailing to college students, who primarily live in apartments or dormitories, you might assess a higher penalty for a blank Given_Name1 or apartment number. Field Prename 5 Given_Name1 20 Given_Name2 5 Primary Postfix 5 Secondary Number 416 Blank penalty 20 2011-06-09
- 417. Data Quality As a result, the records below would be ranked in the order shown (assume they are from the same source, so record priority is not a factor). Even though the first record has blank prename, Given_Name2, and street postfix fields, we want it as the master record because it does contain the data we consider more important: Given_Name1 and Secondary Number. Prename (5) Given Name1 (20) Given Name2 (5) Prim Postfix (5) Maria Ms. A 100 Main Ramirez 100 Main 100 Main St Blankfield penalty 5+5+5 = 15 St Ramirez Ms. Prim Name Sec Number (20) 6 Prim Range Ramirez Maria Family Name 20 20 + 5 = 25 6 16.4.7.3.3 Blank penalty interacts with record priority The record priority and blank penalty scores are added together and considered as one score. For example, suppose you want records from your house database to have high priority, but you also want records with blank fields to have low priority. Is source membership more important, even if some fields are blank? Or is it more important to have as complete a record as possible, even if it is not from the house database? Most want their house records to have priority, and would not want blank fields to override that priority. To make this happen, set a high penalty for membership in a rented source, and lower penalties for blank fields: Source Record priority (penalty points) Field Blank penalty House Source 100 Given Name1 20 Rented Source A 200 Given_Name2 5 Rented Source B 300 Primary Postfix 5 Rented Source C 400 Secondary Number 20 417 2011-06-09
- 418. Data Quality With this scoring system, a record from the house source always receives priority over a record from a rented source, even if the house record has blank fields. For example, suppose the records below were in the same match group. Even though the house record contains five blank fields, it receives only 155 penalty points (100 + 5 + 20 + 5 + 5 + 20), while the record from source A receives 200 penalty points. The house record, therefore, has the lower penalty and the higher priority. Source Given Name1 Given Name2 Source A Rita Source B Rita Prim Name 100 Smith 100 Bren 100 Bren Post code Rec priority Blank Penalty Total 55343 100 55 155 12A 55343 200 0 200 12 55343 300 10 310 Bren Smith A Prim Range Smith House Family Sec Num You can manipulate the scores to set priority exactly as you'd like. In the example above, suppose you prefer a rented record containing first-name data over a house record without first-name data. You could set the first-name blank penalty score to 500 so that a blank first-name field would weigh more heavily than any source membership. 16.4.7.3.4 To define priority and penalty using field values Be sure to map in any input fields that carry priority or blank penalty values. This task tells Match which fields hold your record priority and blank penalty values for your records, and whether to apply these per record. 1. Add a Group Prioritization operation to the Group Forming or Post Match Processing section in the Match Editor. 2. Enter a Prioritization name (if necessary) and select the Record Completeness tab. 3. Select the Order records based on completeness of data option. 4. Select the Define priority and penalty fields option. • • Define only field penalties: This option allows you to select a default record priority and blank penalties per field to generate your priority score. Define priority and penalty based on input source: This allows you to define priority and blank penalty based on membership in an input source. 5. Choose a field that contains the record priority value from the Record priority field option. 6. In the Apply blank penalty field option, choose a field that contains the Y or N indicator for whether to apply a blank penalty to a record. 418 2011-06-09
- 419. Data Quality 7. In the Default record priority option, enter a default record priority to use if a record priority field is blank or if you do not specify a record priority field. 8. Choose a Default apply blank penalty value (Yes or No). This determines whether the Match transform will apply blank penalty to a record if you didn't choose an apply blank penalty field or if the field is blank for a particular record. 9. In the Blank penalty score table, choose a field from the Input Field column to which you want to assign blank penalty values. 10. In the Blank Penalty column, type a blank penalty value to attribute to any record containing a blank in the field you indicated in Input Field column. 16.4.7.3.5 To define penalty values by field This task lets you define your default priority score for every record and blank penalties per field to generate your penalty score. 1. Add a Group Prioritization operation to the Group Forming or Post Match Processing section in the Match Editor. 2. Enter a Prioritization name (if necessary) and select the Record Completeness tab. 3. Select the Order records based on completeness of data option. 4. Select the Define only field penalties option. 5. In the Default record priority option, enter a default record priority that will be used in the penalty score for every record. 6. Choose a Default apply blank penalty value (Yes or No). This determines whether the Match transform will apply blank penalty to a record if you didn't choose an apply blank penalty field or if the field is blank for a particular record. 7. In the Blank penalty score table, choose a field from the Input Field column to which you want to assign blank penalty values. 8. In the Blank Penalty column, type a blank penalty value to attribute to any record containing a blank in the field you indicated in Input Field column. 16.4.7.4 Prioritize records based on source membership However you prefer to prioritize your sources (by sorting a break group or by using penalty scores), you will want to ensure that your suppress-type source records are the drivers in the break group and comparison process. For example, suppose you are a charitable foundation mailing a solicitation to your current donors and to names from two rented sources. If a name appears on your house source and a rented source, you prefer to use the name from your house source. For one of the rented sources, Source B, suppose also that you can negotiate a rebate for any records you do not use. You want to use as few records as possible from Source B so that you can get the largest possible rebate. Therefore, you want records from Source B to have the lowest preference, or priority, from among the three sources. 419 2011-06-09
- 420. Data Quality Source Priority House source Highest Rented source A Medium Rented source B Lowest Suppress-type sources and record completeness In cases where you want to use penalty scores, you will want your Suppress-type sources to have a low priority score. This makes it likely that normal records that match a suppress record will be subordinate matches in a match group, and will therefore be suppressed, as well. Within each match group, any record with a lower priority than a suppression source record is considered a suppress match. For example, suppose you are running your files against the DMA Mail Preference File (a list of people who do not want to receive advertising mailings). You would identify the DMA source as a suppression source and assign a priority of zero. Source Priority DMA Suppression source 0 House source 100 Rented source A 200 Rentd source B 300 Suppose Match found four matching records among the input records. Matching record (name fields only) House 100 Ramirez Ms. Priority Ramirez Maria Source Source B 300 Ms. Maria A Ramirez Source A 200 Ms. Maria A Ramirez DMA 0 The following match group would be established. Based on their priority, Match would rank the records as shown. As a result, the record from the suppression file (the DMA source) would be the master record, and the others would be subordinate suppress matches, and thus suppressed, as well. 420 2011-06-09
- 421. Data Quality Source Priority DMA 0 (Master record) House 100 Source A 200 Source B 300 16.4.7.4.1 To define penalties based on source membership In this task, you can attribute priority scores and blank penalties to an input source, and thus apply these scores to any record belonging to that source. Just be sure you have your input sources defined before you attempt to complete this task. 1. Add a Group Prioritization operation to the Group Forming or Post Match Processing section in the Match Editor. 2. Enter a Prioritization name (if necessary) and select the Record Completeness tab. 3. Select the Order records based on completeness of data option. 4. Select the Define priority and penalty based on input source option. 5. In the Source Attributes table, select a source from the drop-down list. 6. Type a value in the Priority column to assign a record priority to that source. Remember that the lower the score, the higher the priority. For example, you would want to assign a very low score (such as 0) to a suppress-type source. 7. In the Apply Blank Penalty column, choose a Yes or No value to determine whether to use blank penalty on records from that source. 8. In the Default record priority option, enter a default record priority that will be used in the penalty score for every record that is not a member of a source. 9. Choose a Default apply blank penalty value (Yes or No). This determines whether to apply blank penalties to a record that is not a member of a source. 10. In the Blank penalty score table, choose a field from the Input Field column to which you want to assign blank penalty values. 11. In the Blank Penalty column, type a blank penalty value to attribute to any record containing a blank in the field you indicated in Input Field column. 16.4.7.5 Data Salvage Data salvaging temporarily copies data from a passenger record to the driver record after comparing the two records. The data that’s copied is data that is found in the passenger record but is missing or incomplete in the driver record. Data salvaging prevents blank matching or initials matching from matching records that you may not want to match. 421 2011-06-09
- 422. Data Quality For example, we have the following match group. If you did not enable data salvaging, the records in the first table would all belong to the same match group because the driver record, which contains a blank Name field, matches both of the other records. Record Name Postcode 123 Main St. 1 (driver) Address 54601 2 John Smith 123 Main St. 54601 3 Jack Hill 123 Main St. 54601 If you enabled data salvaging, the software would temporarily copy John Smith from the second record into the driver record. The result: Record #1 matches Record #2, but Record #1 does not match Record #3 (because John Smith doesn’t match Jack Hill). Record Name Address Postcode 1 (driver) John Smith (copied from record below) 123 Main St. 54601 2 John Smith 123 Main St. 54601 3 Jack Hill 123 Main St. 54601 The following example shows how this is used for a suppression source. Assume that the suppression source is a list of no-pandering addresses. In that case, you would set the suppression source to have the highest priority, and you would not enable data salvaging. That way, the software suppresses all records that match the suppression source records. For example, a suppress record of 123 Main St would match 123 Main St #2 and 123 Main St Apt C; both of these would be suppressed. 16.4.7.5.1 Data salvaging and initials When a driver record’s name field contains an initial, instead of a full name, the software may temporarily borrow the full name if it finds one in the corresponding field of a matching record. This is one form of data salvaging. For illustration, assume that the following three records represent potentially matching records (for example, the software has grouped these as members of a break group, based on address and ZIP Code data). Note: Initials salvaging only occurs with the given name and family name fields. 422 2011-06-09
- 423. Data Quality Record First name Last name Address Notes 357 J L 123 Main Driver 391 Juanita Lopez 123 Main 839 Joanne London 123 Main Lowest ranking record The first match comparison will be between the driver record (357) and the next highest ranking record (391). These two records will be called a match. Juanita and Lopez are temporarily copied to the name fields of record# 357. The next comparison will be between record 357 and the next lower ranking record (839). With data salvaging, the driver record’s name data is now Juanita Lopez (as “borrowed” from the first comparison). Therefore, record 839 will probably be considered not-to match record 357. By retaining more information for the driver record, data salvaging helps improve the quality of your matching results. Initials and suppress-type records However, if the driver record is a suppress-type record, you may prefer to turn off data salvaging, to retain your best chance of identifying all the records that match the initialized suppression data. For example, if you want to suppress names with the initials JL (as in the case above, you would want to find all matches to JL regardless of the order in which the records are encountered in the break group. If you have turned off data salvaging for the records of this suppression source, here is what happens during those same two match comparisons: Record First name Last name Address Notes 357 J L 123 Main Driver 391 Juanita Lopez 123 Main 839 Joanne London 123 Main Lowest ranking record The first match comparison will be between the driver record (357) and the next- highest ranking record (391). These two records will be called a match, since the driver record’s JL and Juanita Lopez will be called a match. The next comparison will be between the driver record (357) and the next lower ranking record (839). This time these two records will also be called a match, since the driver record’s JL will match Joanne London. Since both records 391 and 839 matched the suppress-type driver record, they are both designated as suppress matches, and, therefore, neither will be included in your output. 423 2011-06-09
- 424. Data Quality 16.4.7.5.2 To control data salvaging using a field You can use a field to control whether data salvage is enabled. If the field's value is Y for a record, data salvaging is enabled. Be sure to map the field into the Match transform that you want to use beforehand. 1. Open the Match Editor for a Match transform. 2. In the Transform Options window, click the Data Salvage tab. 3. Select the Enable data salvage option, and choose a default value for those records. The default value will be used in the cases where the field you choose is not populated for a particular record. 4. Select the Specify data salvage by field option, and choose a field from the drop-down menu. 16.4.7.5.3 To control data salvaging by source You can use membership in an input source to control whether data salvage is enabled or disabled for a particular record. Be sure to create your input sources beforehand. 1. Open the Match Editor for a Match transform. 2. In the Transform Options window, click the Data Salvage tab. 3. Select the Enable data salvage option, and choose a default value for those records. The default value will be used if a record's input source is not specified in the following steps. 4. Select the Specify data salvage by source option. 5. In the table, choose a Source and then a Perform Data Salvage value for each source you want to use. 16.4.8 Match criteria 16.4.8.1 Overview of match criteria Use match criteria in each match level to determine the threshold scores for matching and to define how to treat various types of data, such as numeric, blank, name data, and so on (your business rules). You can do all of this in the Criteria option group of the Match Editor. Match criteria To the Match transform, match criteria represent the fields you want to compare. For example, if you wanted to match on the first ten characters of a given name and the first fifteen characters of the family name, you must create two criteria that specify these requirements. 424 2011-06-09
- 425. Data Quality Criteria provide a way to let the Match transform know what kind of data is in the input field and, therefore, what types of operations to perform on that data. Pre-defined vs. custom criteria There are two types of criteria: • Pre-defined criteria are available for fields that are typically used for matching, such as name, address, and other data. By assigning a criteria to a field, the Match transform is able to identify what type of data is in the field, and allow it to perform internal operations to optimize the data for matching, without altering the actual input data. • Data Cleanse custom (user-defined, non party-data) output fields are available as pre-defined criteria. Map the custom output fields from Data Cleanse and the custom fields appear in the Match Editor's Criteria Fields tab. Any other types of data (such as part numbers or other proprietary data), for which a pre-defined criteria does not exist, should be designated as a custom criteria. Certain functions can be performed on custom keys, such as abbreviation, substring, numeric matching, but the Match transform cannot perform some cross-field comparisons such as some name matching functions. • Match criteria pre-comparison options The majority of your data standardization should take place in the address cleansing and Data Cleanse transforms. However, the Match transform can perform some preprocessing per criteria (and for matching purposes only; your actual data is not affected) to provide more accurate matches. The options to control this standardization are located in the Options and Multi Field Comparisons tabs of the Match editor. They include: • • • • • Convert diacritical characters Convert text to numbers Convert to uppercase Remove punctuation Locale For more information about these options, see the Match transform section of the Reference Guide. 16.4.8.1.1 To add and order a match criteria You can add as many criteria as you want to each match level in your Match transform. 1. Select the appropriate match level or Match Criteria option group in the Option Explorer of the Match Editor, and right-click. 2. Choose Criteria. 3. Enter a name for your criteria in the Criteria name box. You can keep the default name for pre-defined criteria, but you should enter a meaningful criteria name if you chose a Custom criteria. 4. On the Criteria Fields tab, in the Available criteria list, choose the criteria that best represents the data that you want to match on. If you don't find what you are looking for, choose the Custom criteria. 5. In the Criteria field mapping table, choose an input field mapped name that contains the data you want to match on for this criteria. 6. Click the Options tab. 425 2011-06-09
- 426. Data Quality 7. Configure the Pre-comparison options and Comparison rules. Be sure to set the Match score and No match score, because these are required. 8. If you want to enable multiple field (cross-field) comparison, click the Multiple Fields Comparisons tab, and select the Compare multiple fields option. a. Choose the type of multiple field comparison to perform: • All selected fields in other records: Compare each field to all fields selected in the table in all records. • The same field in other records: Compare each field only to the same field in all records. b. In the Additional fields to compare table, choose input fields that contain the data you want to include in the multiple field comparison for this criteria. Tip: You can use custom match criteria field names for multiple field comparison by typing in the Custom name column. Note: If you enable multiple field comparison, any appropriate match standard fields are removed from the Criteria field mapping table on the Criteria Fields tab . If you want to include them in the match process, add them in the Additional fields to compare table. 9. Configure the Pre-comparison options for multiple field comparison. 10. To order your criteria in the Options Explorer of the Match Editor (or the Match Table), select a criteria and click the Move Up or Move Down buttons as necessary. 16.4.8.2 Matching methods There are a number of ways to set up and order your criteria to get the matching results you want. Each of these ways have advantages and disadvantages, so consider them carefully. Match method Rule-based Allows you to control which criteria determines a match. This method is easy to set up. Weightedscoring Allows you to assign importance, or weight, to any criteria. However, weightedscoring evaluates every rule before determining a match, which might cause an increase in processing time. Combination method 426 Description Same relative advantages and disadvantages as the other two methods. 2011-06-09
- 427. Data Quality 16.4.8.2.1 Similarity score The similarity score is the percentage that your data is alike. This score is calculated internally by the application when records are compared. Whether the application considers the records a match depends on the Match and No match scores you define in the Criteria option group (as well as other factors, but for now let's focus on these scores). Example: This is an example of how similarity scores are determined. Here are some things to note: • The comparison table below is intended to serve as an example. This is not how the matching process works in the weighted scoring method, for example. • Only the first comparison is considered a match, because the similarity score met or exceeded the match score. The last comparison is considered a no-match because the similarity score was less than the no-match score. • When a single criteria cannot determine a match, as in the case of the second comparison in the table below, the process moves to the next criteria, if possible. Comparison No match Match Similarity score Matching? Smith > Smith 72 95 100% Yes Smith > Smitt 72 95 80% Depends on other criteria Smith > Smythe 72 95 72% No Smith > Jones 72 95 20% No 16.4.8.2.2 Rule-based method With rule-based matching, you rely only on your match and no-match scores to determine matches within a criteria. Example: This example shows how to set up this method in the Match transform. 427 2011-06-09
- 428. Data Quality Criteria Record A Record B No match Match Similarity score Given Name1 Mary Mary 82 101 100 Family Name Smith Smitt 74 101 80 E-mail [email protected] [email protected] 79 80 91 By entering a value of 101 in the match score for every criteria except the last, the Given Name1 and Family Name criteria never determine a match, although they can determine a no match. By setting the Match score and No match score options for the E-mail criteria with no gap, any comparison that reaches the last criteria must either be a match or a no match. A match score of 101 ensures that the criteria does not cause the records to be a match, because two fields cannot be more than 100 percent alike. Remember: Order is important! For performance reasons, you should have the criteria that is most likely to make the match or no-match decisions first in your order of criteria. This can help reduce the number of criteria comparisons. 16.4.8.2.3 Weighted-scoring method In a rule-based matching method, the application gives all of the criteria the same amount of importance (or weight). That is, if any criteria fails to meet the specified match score, the application determines that the records do not match. When you use the weighted scoring method, you are relying on the total contribution score for determining matches, as opposed to using match and no-match scores on their own. Contribution values Contribution values are your way of assigning weight to individual criteria. The higher the value, the more weight that criteria carries in determining matches. In general, criteria that might carry more weight than others include account numbers, Social Security numbers, customer numbers, Postcode1, and addresses. Note: All contribution values for all criteria that have them must total 100. You do not need to have a contribution value for all of your criteria. You can define a criteria's contribution value in the Contribution to weighted score option in the Criteria option group. 428 2011-06-09
- 429. Data Quality Contribution and total contribution score The Match transform generates the contribution score for each criteria by multiplying the contribution value you assign with the similarity score (the percentage alike). These individual contribution scores are then added to get the total contribution score. Weighted match score In the weighted scoring method, matches are determined only by comparing the total contribution score with the weighted match score. If the total contribution score is equal to or greater than the weighted match score, the records are considered a match. If the total weighted score is less than the weighted match score, the records are considered a no-match. You can set the weighted match score in the Weighted match score option of the Level option group. Example: The following table is an example of how to set up weighted scoring. Notice the various types of scores that we have discussed. Also notice the following: • When setting up weighted scoring, the No match score option must be set to -1, and the Match score option must be set to 101. These values ensure that neither a match nor a no-match can be found by using these scores. • We have assigned a contribution value to the E-mail criteria that gives it the most importance. Criteria Record A Record B No match Match Similarity score Contribution value Contribution score (similarity X contribution value) First Name Mary Mary -1 101 100 25 25 Last Name Smith Smitt -1 101 80 25 20 E-mail ms@ sap.com msmith@ sap.com -1 101 84 50 42 Total contribution score: 87 If the weighted match score is 87, then any comparison whose total contribution score is 87 or greater is considered a match. In this example, the comparison is a match because the total contribution score is 87. 16.4.8.2.4 Combination method This method combines the rule-based and weighted scoring methods of matching. 429 2011-06-09
- 430. Data Quality Contribution score (actual similarity X contribution value) Criteria Record A Record B No match Match Sim score Contribution value First Name Mary Mary 59 101 100 25 25 Last Name Smith Hope 59 101 22 N/A (No Match) N/A E-mail ms@ sap.com msmith@ sap.com 49 101 N/A N/A N/A Total contribution score N/A 16.4.8.3 Matching business rules An important part of the matching process is determining how you want to handle various forms of and differences in your data. For example, if every field in a record matched another record's fields, except that one field was blank and the other record's field was not, would you want these records to be considered matches? Figuring out what you want to do in these situations is part of defining your business rules. Match criteria are where you define most of your business rules, while some name-based options are set in the Match Level option group. 16.4.8.3.1 Matching on strings, abbreviations, and initials Initials and acronyms Use the Initials adjustment score option to allow matching initials to whole words. For example, "International Health Providers" can be matched to "IHP". Abbreviations Use the Abbreviation adjustment score option to allow matching whole words to abbreviations. For example, "International Health Providers" can be matched to "Intl Health Providers". String data Use the Substring adjustment score option to allow matching longer strings to shorter strings. For example, the string "Mayfield Painting and Sand Blasting" can match "Mayfield painting". 430 2011-06-09
- 431. Data Quality 16.4.8.3.2 Extended abbrevation matching Extended abbreviation matching offers functionality that handles situations not covered by the Initials adjustment score, Substring adjustment score, and Abbreviation adjustment score options. For example, you might encounter the following situations: • Suppose you have localities in your data such as La Crosse and New York. However, you also have these same localities listed as LaCrosse and NewYork (without spaces). Under normal matching, you cannot designate these (La Crosse/LaCrosse and New York/NewYork) as matching 100%; the spaces prevent this. (These would normally be 94 and 93 percent matching.) • Suppose you have Metropolitan Life and MetLife (an abbreviation and combination of Metropolitan Life) in your data. The Abbreviation adjustment score option cannot detect the combination of the two words. If you are concerned about either of these cases in your data, you should use the Ext abbreviation adjustment score option. How the adjustment score works The score you set in the Ext abbreviation adjustment score option tunes your similarity score to consider these types of abbreviations and combinations in your data. The adjustment score adds a penalty for the non-matched part of the words. The higher the number, the greater the penalty. A score of 100 means no penalty and score of 0 means maximum penalty. Example: Sim score when Adj score is 50 Sim score when Adj score is 100 String 1 String 2 Sim score when Adj score is 0 MetLife Metropolitan Life 58 79 100 MetLife Met Life 93 96 100 MetLife MetropolitanLife 60 60 60 Notes This score is due to string comparison. Extended Abbreviation scoring was not needed or used because both strings being compared are each one word. 16.4.8.3.3 Name matching Part of creating your business rules is to define how you want names handled in the matching process. The Match transform gives you many ways to ensure that variations on names or multiple names, for example, are taken into consideration. 431 2011-06-09
- 432. Data Quality Note: Unlike other business rules, these options are set up in the match level option group, because they affect all appropriate name-based match criteria. Two names; two persons With the Number of names that must match option, you can control how matching is performed on match keys with more than one name (for example, comparing "John and Mary Smith" to "Dave and Mary Smith"). Choose whether only one name needs to match for the records to be identified as a match, or whether the Match transform should disregard any persons other than the first name it parses. With this method you can require either one or both persons to match for the record to match. Two names; one person With the Compare Given_Name1 to Given_Name2 option, you can also compare a record's Given_Name1 data (first name) with the second record's Given_Name2 data (middle name). With this option, the Match transform can correctly identify matching records such as the two partially shown below. Typically, these record pairs represent sons or daughters named for their parents, but known by their middle name. Record # First name Middle name Last name Address 170 Leo Thomas Smith 225 Pushbutton Dr 198 Tom Smith 225 Pushbutton Dr Hyphenated family names With the Match on hyphenated family name option, you can control how matching is performed if a Family_Name (last name) field contains a hyphenated family name (for example, comparing "Smith-Jones" to "Jones"). Choose whether both criteria must have both names to match or just one name that must match for the records to be called a match. Match compound family names The Approximate Substring Score assists in setting up comparison of compound family names. The Approximate Substring score is assigned to the words that do not match to other words in a compared string.This option loosens some of the requirements of the Substring Adjustment score option in the following ways: • First words do not have to match exactly. • The words that do match can use initials and abbreviations adjustments (For example, Rodriguez and RDZ). • Matching words have to be in the same order, but there can be non-matching words before or after the matching words. • The Approximate Substring score is assigned the leftover words and spaces in the compared string. 432 2011-06-09
- 433. Data Quality The Approximate Substring option will increase the score for some matches found when using the Substring Matching Score. Example: When comparing CRUZ RODRIGUEZ and GARCIA CRUZ DE RDZ, the similarity scores are: • • • Without setting any adjusments, the Similarity score is 48. When you set the Substring adjustment score to 80 and the Abbreviation score to 80, the Similarity score is 66. When you set the Approximate substring adjustment score to 80 and the Abbreviation score to 80, the Similarity score is 91. 16.4.8.3.4 Numeric data matching Use the Numeric words match exactly option to choose whether data with a mixture of numbers and letters should match exactly. You can also specify how this data must match. This option applies most often to address data and custom data, such as a part number. The numeric matching process is as follows: 1. The string is first broken into words. The word breaking is performed on all punctuation and spacing, and then the words are assigned a numeric attribute. A numeric word is any word that contains at least one number from 0 to 9. For example, 4L is considered a numeric word, whereas FourL is not. 2. Numeric matching is performed according to the option setting that you choose (as described below). Option values and how they work Option value Description With this value, numeric words must match exactly; however, the position of the word is not important. For example: • Street address comparison: "4932 Main St # 101" and "# 101 4932 Main St" are considered a match. Any_Position 433 Street address comparison: "4932 Main St # 101" and "# 102 4932 Main St" are not considered a match. • Same_Position • Part description: "ACCU 1.4L 29BAR" and "ACCU 29BAR 1.4L" are considered a match. This value specifies that numeric words must match exactly; however, this option differs from the Any_Position option in that the position of the word is important. For example, 608-782-5000 will match 608-782-5000, but it will not match 782-608-5000. 2011-06-09
- 434. Data Quality Option value Description This value performs word breaking on all punctuation and spaces except on the decimal separator (period or comma) so that all decimal numbers are not broken. For example, the string 123.456 is considered a single numeric word as opposed to two numeric words. Any_Position_Consid er_Punctuation The position of the numeric word is not important; however, decimal separators do impact the matching process. For example: • Part description: "ACCU 29BAR 1.4L" and "ACCU 1.4L 29BAR" are considered a match. • • Any_Position_Ig nore_Punctuation Part description: "ACCU 1,4L 29BAR" and "ACCU 29BAR 1.4L" are not considered a match because there is a decimal indicator between the 1 and the 4 in both cases. Financial data: "25,435" and "25.435" are not considered a match. This value is similar to the Any_Position_Consider_Punctuation value, except that decimal separators do not impact the matching process. For example: • Part description: "ACCU 29BAR 1.4L" and "ACCU 1.4L 29BAR" are considered a match. • Part description: "ACCU 1,4L 29BAR" and "ACCU 29BAR 1.4L" are also considered a match even though there is a decimal indicator between the 1 and the 4. • Part description: "ACCU 29BAR 1.4L" and "ACCU 1.5L 29BAR" are not considered a match. 16.4.8.3.5 Blank field matching In your business rules, you can control how the Match transform treats field comparisons when one or both of the fields compared are blank. For example, the first name field is blank in the second record shown below. Would you want the Match transform to consider these records matches or no matches? What if the first name field were blank in both records? 434 2011-06-09
- 435. Data Quality Record #1 Record #2 John Doe _____ Doe 204 Main St 204 Main St La Crosse WI La Crosse WI 54601 54601 There are some options in the Match transform that allow you to control the way these are compared. They are: • • • • Both fields blank operation Both fields blank score One field blank operation One field blank score Blank field operations The "operation" options have the following value choices: Option Description Eval If you choose Eval, the Match transform scores the comparison using the score you enter at the One field blank score or Both fields blank score option. Ignore If you choose Ignore, the score for this field rule does not contribute to the overall weighted score for the record comparison. In other words, the two records shown above could still be considered duplicates, despite the blank field. Blank field scores The "Score" options control how the Match transform scores field comparisons when the field is blank in one or both records. You can enter any value from 0 to 100. To help you decide what score to enter, determine if you want the Match transform to consider a blank field 0 percent similar to a populated field or another blank field, 100 percent similar, or somewhere in between. Your answer probably depends on what field you're comparing. Giving a blank field a high score might be appropriate if you're matching on a first or middle name or a company name, for example. Example: Here are some examples that may help you understand how your settings of these blank matching options can affect the overall scoring of records. One field blank operation for Given_Name1 field set to Ignore 435 2011-06-09
- 436. Data Quality Note that when you set the blank options to Ignore, the Match transform redistributes the contribution allotted for this field to the other criteria and recalculates the contributions for the other fields. Fields compared Record A Record B % alike Contribution Score (per field) Postcode 54601 54601 100 20 (or 22) 22 Address 100 Water St 100 Water St 100 40 (or 44) 44 Family_Name Hamilton Hammilton 94 30 (or 33) 31 Given_Name1 Mary — 10 (or 0) — Weighted score: 97 One field blank operation for Given_Name1 field set to Eval; One field blank score set to 0 Fields compared Record A Record B % alike Contribution Score (per field) Postcode 54601 54601 100 20 20 Address 100 Water St 100 Water St 100 40 40 Family_Name Hamilton Hammilton 94 30 28 Given_Name1 Mary 0 10 0 Weighted score: 88 One field blank operation for Given_Name1 field set to Eval; One field blank score set to 100 Fields compared Record A Record B % alike Contribution Score (per field) Postcode 54601 54601 100 20 20 Address 100 Water St 100 Water St 100 40 40 Family_Name Hamilton Hammilton 94 30 28 Given_Name1 Mary 100 10 10 Weighted score: 98 436 2011-06-09
- 437. Data Quality 16.4.8.3.6 Multiple field (cross-field) comparison In most cases, you use a single field for comparison. For example, Field1 in the first record is compared with Field1 in the second record. However, there are situations where comparing multiple fields can be useful. For example, suppose you want to match telephone numbers in the Phone field against numbers found in fields used for Fax, Mobile, and Home. Multiple field comparison makes this possible. When you enable multiple field comparison in the Multiple Field Comparison tab of a match criteria in the Match Editor, you can choose to match selected fields against either all of the selected fields in each record, or against only the same field in each record. Note: By default, Match performs multiple field comparison on fields where match standards are used. For example, Person1_Given_Name1 is automatically compared to Person1_Given_Name_Match_Std1-6. Multiple field comparison does not need to be explicitly enabled, and no additional configuration is required to perform multiple field comparison against match standard fields. Comparing selected fields to all selected fields in other records When you compare each selected field to all selected fields in other records, all fields that are defined in that match criteria are compared against each other. Remember: “Selected” fields include the criteria field and the other fields you define in the Additional fields to compare table. • • If one or more field comparisons meets the settings for Match score, the two rows being compared are considered matches. If one or more field comparisons exceeds the No match score, the rule will be considered to pass and any other defined criteria/weighted scoring will be evaluated to determine if the two rows are considered matches. Example: Example of comparing selected fields to all selected fields in other records Your input data contains two firm fields. Row ID Firm1 Firm2 1 Firstlogic Postalsoft 2 SAP BusinessObjects Firstlogic With the Match score set to 100 and No match score set to 99, these two records are considered matches. Here is a summary of the comparison process and the results. • First, Row 1 Firm1 (Firstlogic) is compared to Row 2 Firm1 (SAP BusinessObjects). Normally, the rows would fail this comparison, but with multi-field comparison activated, a No Match decision is not made yet. 437 2011-06-09
- 438. Data Quality • Next, Row 1 Firm2 is compared to Row 2 Firm2 and so on until all other comparisons are made between all fields in all rows. Because Row 1 Firm1 (Firstlogic) and Row 2 Firm2 (Firstlogic) are 100% similar, the two records are considered matches. Comparing selected fields to the same fields in other records When you compare each selected field to the same field in other records, each field defined in the Multiple Field Comparison tab of a match criteria are compared only to the same field in other records. This sets up, within this criteria, what is essentially an OR condition for passing the criteria. Each field is used to determine a match: If Field_1, Field_2, or Field_3 passes the match criteria, consider the records a match. The No Match score for one field does not automatically fail the criteria when you use multi-field comparison. Remember: “Selected” fields include the criteria field and the other fields you define in the Additional fields to compare table. Example: Example of comparing selected fields to the same field in other records Your input data contains a phone, fax, and cell phone field. If any one of these input field's data is the same between thte rows, the records are found to be matches. Row ID Phone Fax Cell 1 608-555-1234 608-555-0000 608-555-4321 2 608-555-4321 608-555-0000 608-555-1111 With a Match score of 100 and a No match score of 99, the phone and the cell phone number would both fail the match criteria, if defined individually. However, because all three fields are defined in one criteria and the selected records being compared to the same records, the fact that the fax number is 100% similar calls these records a match. Note: In the example above, Row 1's cell phone and Row 2's phone would not be considered a match with the selection of the the same field to other records option because it only compares within the same field in this case. If this cross-comparison is needed, select the all selected fields in other records option instead. 16.4.8.3.7 Proximity matching Proximity matching gives you the ability to match records based on their proximity instead of comparing the string representation of data. You can match on geographic, numeric, and date proximity. • • 438 Match on Geographic proximity Match on numeric or date proximity 2011-06-09
- 439. Data Quality Match on Geographic proximity Geographic Proximity finds duplicate records based on geographic proximity, using latitude and longitude information. This is not driving distance, but Geographic distance. This option uses WGS 84 (GPS) coordinates. The Geographic proximity option can: • Search on objects within a radial range. This can help a company that wants to send a mailing out to customers within a certain distance from their business location. • Search on the nearest location. This can help a consumer find a store location closest to their address. Set up Geographic Proximity Matching - Criteria Fields To select the fields for Geographic Proximity matching, follow these steps: 1. Access the Match Editor, add a new criteria. 2. From Available Criteria, expand Geographic. 3. Select LATITUDE_LONGITUDE. This will make the two criteria fields available for mapping. 4. Map the correct latitude and longitude fields. You must map both fields. Set up Geographic Proximity matching - Criteria options You must have the Latitude and Longitude fields mapped before you can use this option. To perform geographic proximity matching, follow these steps: 1. From Compare data using, select Geo Proximity. This filters the options under Comparison Rules to show only applicable options. 2. Set the Distance unit option to one of the following: • Miles • Feet • Kilometers • Meters 3. Enter the Max Distance you want to consider for the range. 4. Set the Max Distance Score. Note: A distance equal to Max distance will receive a score of Max distance score. Any distance less than the Max distance will receive a proportional score between Max distance score and 100. For example, a proximity of 10 miles will have higher score than a 40 miles. 439 2011-06-09
- 440. Data Quality Match on numeric or date proximity The Match Transform's numeric proximity options find duplicates based on numerical closeness of data. You can find duplicates based on numeric values and date values. The following options are available in the Match Criteria Editor Options tab for numeric and date matching: Numeric difference Finds duplicates based on the numeric difference for numeric or date values. For example, you can use this option to find duplicates based on date values in a specific range (for example, plus or minus 35 days), regardless of character-based similarity. Numeric percent difference Finds duplicates based on the percentage of numeric difference for numeric values. Here are two examples where this might be useful : • • Finance data domain : You can search financial data to find all monthly mortgage payments that are within 5 percent of a given value. Product data domain, you can search product data to find all the steel rods that are within 10% tolerance of a given diameter. 16.4.9 Post-match processing 16.4.9.1 Best record A key component in most data consolidation efforts is salvaging data from matching records—that is, members of match groups—and posting that data to a best record, or to all matching records. You can perform these functions by adding a Best Record post-match operation. Operations happen within match groups The functions you perform with the Best Record operation involve manipulating or moving data contained in the master records and subordinate records of match groups. Match groups are groups of records that the Match transform has found to be matching, based on the criteria you have created. A master record is the first record in the Match group. You can control which record this is by using a Group Prioritization operation before the Best Record operation. Subordinate records are all of the remaining records in a match group. To help illustrate this use of master and subordinate records, consider the following match group: 440 2011-06-09
- 441. Data Quality Record Name #1 John Smith #2 John Smyth #3 #4 Phone Date Group rank 11 Apr 2001 Master 788-8700 12 Oct 1999 Subordinate John E. Smith 788-1234 22 Feb 1997 Subordinate J. Smith 788-3271 Subordinate Because this is a match group, all of the records are considered matching. As you can see, each record is slightly different. Some records have blank fields, some have a newer date, all have different phone numbers. A common operation that you can perform in this match group is to move updated data to all of the records in a match group. You can choose to move data to the master record, to all the subordinate members of the match group, or to all members of the match group. The most recent phone number would be a good example here. Another example might be to salvage useful data from matching records before discarding them. For example, when you run a drivers license file against your house file, you might pick up gender or date-of-birth data to add to your house record. Post higher priority records first The operations you set up in the Best Record option group should always start with the highest priority member of the match group (the master) and work their way down to the last subordinate, one at a time. This ensures that data can be salvaged from the higher-priority record to the lower priority record. So, be sure that your records are prioritized correctly, by adding a Group Prioritization post-match operation before your Best Record operation. 16.4.9.1.1 Best record strategies We provide you with strategies that help you set up some more common best record operation quickly and easily. If none of these strategies fit your needs, you can create a custom best record strategy, using your own Python code. Best record strategies act as a criteria for taking action on other fields. If the criteria is not met, no action is taken. Example: In our example of updating a phone field with the most recent data, we can use the Date strategy with the Newest priority to update the master record with the latest phone number in the match group. This 441 2011-06-09
- 442. Data Quality latter part (updating the master record with the latest phone number) is the action. You can also update all of the records in the match group (master and all subordinates) or only the subordinates. Restriction: The date strategy does not parse the date, because it does not know how the data is formatted. Be sure your data is pre-formatted as YYYYMMDD, so that string comparisons work correctly. You can also do this by setting up a custom strategy, using Python code to parse the date and use a date compare. Custom best record strategies and Python In the pre-defined strategies for the Best Record strategies, the Match transform auto-generates the Python code that it uses for processing. Included in this code, are variables that are necessary to manage the processing. Common variables The common variables you see in the generated Python code are: Variable Description SRC Signifies the source field. DST Signifies the destination field. RET Specifies the return value, indicating whether the strategy passed or failed (must be either "T" or "F"). NEWDST and NEWGRP variables Use the NEWDST and NEWGRP variables to allow the posting of data in your best-record action to be independent of the strategy fields. If you do not include these variables, the strategy field data must also be updated. Variable Description NEWDST New destination indicator. This string variable will have a value of "T" when the destination record is new or different than the last time the strategy was evaluated and a value of "F" when the destination record has not changed since last time. The NEWDST variable is only useful if you are posting to multiple destinations, such as ALL or SUBS in the Posting destination option. NEWGRP New group indicator. This string variable will have a value of "T" when the match group is different than the last time the strategy was evaluated and a value of "F" when the match group has not changed since last time. NEWDST example The following Python code was generated from a NON_BLANK strategy with options set this way: 442 2011-06-09
- 443. Data Quality Option Setting Best record strategy NON_BLANK Strategy priority Priority option not available for the NON_BLANK strategy. Strategy field NORTH_AMERICAN_PHONE1_NORTH_AMERICAN_PHONE_STANDARDIZED. Posting destination ALL Post only once per destination YES Here is what the Python code looks like. # Setup local temp variable to store updated compare condition dct = locals() # Store source and destination values to temporary variables # Reset the temporary variable when the destination changes if (dct.has_key('BEST_RECORD_TEMP') and NEWDST.GetBuffer() == u'F'): DESTINATION = dct['BEST_RECORD_TEMP'] else: DESTINATION = DST.GetField(u'NORTH_AMERICAN_PHONE1_NORTH_AMERICAN_PHONE_STANDARDIZED') SOURCE = SRC.GetField(u'NORTH_AMERICAN_PHONE1_NORTH_AMERICAN_PHONE_STANDARDIZED') if len(SOURCE.strip()) > 0 and len(DESTINATION.strip()) == 0: RET.SetBuffer(u'T') dct['BEST_RECORD_TEMP'] = SOURCE else: RET.SetBuffer(u'F') dct['BEST_RECORD_TEMP'] = DESTINATION # Delete temporary variables del SOURCE del DESTINATION Example: NEWDST and NEWGRP Suppose you have two match groups, each with three records. Match group Records Match group 1 Record A Record B Record C Match group 2 Record D Record E Record F Each new destination or match group is flagged with a "T". 443 2011-06-09
- 444. Data Quality NEWGRP NEWDST (T or F) (T or F) T (New match group) T (New destination "A") Record A > Record B F F A>C F T (New destination "B") B>A F F B>C F T (New destination "C") C>A F F C>B T (New match group) T (New destination "D") D>E F F D>F F T (New destination "E") E>D F F E>F F T (New destination "F") F>D F F F>E Comparison To create a pre-defined best record strategy Be sure to add a Best Record post-match operation to the appropriate match level in the Match Editor. Also, remember to map any pertinent input fields to make them available for this operation. This procedure allows you to quickly generate the criteria for your best record action. The available strategies reflect common use cases. 1. Enter a name for this Best Record operation. 2. Select a strategy from the Best record strategy option. 3. Select a priority from the Strategy priority option. The selection of values depends on the strategy you chose in the previous step. 4. Select a field from the Strategy field drop-down menu. The field you select here is the one that acts as a criteria for determining whether a best record action is taken. 444 2011-06-09
- 445. Data Quality Example: The strategy field you choose must contain data that matches the strategy you are creating. For example, if you are using a newest date strategy, be sure that the field you choose contains date data. To create a custom best record strategy 1. Add a best record operation to your Match transform. 2. Enter a name for your best record operation. 3. In the Best record strategy option, choose Custom. 4. Choose a field from the Strategy field drop-down list. 5. Click the View/Edit Python button to create your custom Python code to reflect your custom strategy. The Python Editor window appears. 16.4.9.1.2 Best record actions Best record actions are the functions you perform on data if a criteria of a strategy is met. Example: Suppose you want to update phone numbers of the master record. You would only want to do this if there is a subordinate record in the match group that has a newer date, which signifies a potentially new phone number for that person. The action you set up would tell the Match transform to update the phone number field in the master record (action) if a newer date in the date field is found (strategy). Sources and destinations When working with the best record operation, it is important to know the differences between sources and destinations in a best record action. The source is the field from which you take data and the destination is where you post the data. A source or destination can be either a master or subordinate record in a match group. Example: In our phone number example, the subordinate record has the newer date, so we take data from the phone field (the source) and post it to the master record (the destination). Posting once or many times per destination In the Best Record options, you can choose to post to a destination once or many times per action by setting the Post only once per destination option. 445 2011-06-09
- 446. Data Quality You may want your best record action to stop after the first time it posts data to the destination record, or you may want it to continue with the other match group records as well. Your choice depends on the nature of the data you’re posting and the records you’re posting to. The two examples that follow illustrate each case. If you post only once to each destination record, then once data is posted for a particular record, the Match transform moves on to either perform the next best record action (if more than one is defined) or to the next record. If you don’t limit the action in this way, all actions are performed each time the strategy returns True. Regardless of this setting, the Match transform always works through the match group members in priority order. When posting to record #1 in the figure below, without limiting the posting to only once, here is what happens: Match group Action Record #1 (master) Record #2 (subordinate) First, the action is attempted using, as a source, that record from among the other match group records that has the highest priority (record #2). Record #3 (subordinate) Next, the action is attempted with the next highest priority record (record #3) as the source. Record #4 (subordinate) Finally, the action is attempted with the lowest priority record (record #4) as the source. The results In the case above, record #4 was the last source for the action, and therefore could be a source of data for the output record. However, if you set your best record action to post only once per destination record, here is what happens: Match group Action Record #1 (master) First, the action is attempted using, as a source, that record from among the other match group records that has the highest priority (record #2). Record #2 (subordinate) If this attempt is successful, the Match transform considers this best record action to be complete and moves to the next best record action (if there is one), or to the next output record. If this attempt is not successful, the Match transform moves to the match group member with the next highest priority and attempts the posting operation. Record #3 (subordinate) 446 2011-06-09
- 447. Data Quality Match group Action Record #4 (subordinate) In this case, record #2 was the source last used for the best record action, and so is the source of posted data in the output record. To create a best record action The best record action is the posting of data from a source to a destination record, based on the criteria of your best record strategy. 1. Create a strategy, either pre-defined or custom. 2. Select the record(s) to post to in the Posting destination option. 3. Select whether you want to post only once or multiple times to a destination record in the Post only once per destination option. 4. In the Best record action fields table, choose your source field and destination field. When you choose a source field, the Destination field column is automatically populated with the same field. You need to change the destination field if this is not the field you want to post your data to. 5. If you want to create a custom best record action, choose Yes in the Custom column. You can now access the Python editor to create custom Python code for your custom action. 16.4.9.1.3 Destination protection The Best Record and Unique ID operations in the Match transform offer you the power to modify existing records in your data. There may be times when you would like to protect data in particular records or data in records from particular input sources from being overwritten. The Destination Protection tab in these Match transform operations allow you the ability to protect data from being modified. To protect destination records through fields 1. In the Destination Protection tab, select Enable destination protection. 2. Select a value in the Default destination protection option drop-down list. This value determines whether a destination is protected if the destination protection field does not have a valid value. 3. Select the Specify destination protection by field option, and choose a field from the Destination protection field drop-down list (or Unique ID protected field) . The field you choose must have a Y or N value to specify the action. Any record that has a value of Y in the destination protection field will be protected from being modified. 447 2011-06-09
- 448. Data Quality To protect destination records based on input source membership You must add an Input Source operation and define input sources before you can complete this task. 1. In the Destination Protection tab, select Enable destination protection. 2. Select a value in the Default destination protection option drop-down list. This value determines whether a destination (input source) is protected if you do not specifically define the source in the table below. 3. Select the Specify destination protection by source option. 4. Select an input source from the first row of the Source name column, and then choose a value from the Destination protected (or Unique ID protected) column. Repeat for every input source you want to set protection for. Remember that if you do not specify for every source, the default value will be used. 16.4.9.2 Unique ID A unique ID refers to a field within your data which contains a unique value that is associated with a record or group of records. You could use a unique ID, for example, in your company's internal database that receives updates at some predetermined interval, such as each week, month, or quarter. Unique ID applies to a data record in the same way that a national identification number might apply to a person; for example, a Social Security number (SSN) in the United States, or a National Insurance number (NINO) in the United Kingdom. It creates and tracks data relationships from run to run. With the Unique ID operation, you can set your own starting ID for new key generation, or have it dynamically assigned based on existing data. The Unique ID post-match processing operation also lets you begin where the highest unique ID from the previous run ended. Unique ID works on match groups Unique ID doesn't necessarily assign IDs to individual records. It can assign the same ID to every record in a match group (groups of records found to be matches). If you are assigning IDs directly to a break group, use the Group number field option to indicate which records belong together. Additionally, make sure that the records are sorted by group number so that records with the same group number value appear together. If you are assigning IDs to records that belong to a match group resulting from the matching process, the Group number field is not required and should not be used. Note: If you are assigning IDs directly to a break group and the Group number field is not specified, Match treats the entire data collection as one match group. 448 2011-06-09
- 449. Data Quality 16.4.9.2.1 Unique ID processing options The Unique ID post-match processing operation combines the update source information with the master database information to form one source of match group information. The operation can then assign, combine, split, and delete unique IDs as needed. You can accomplish this by using the Processing operation option. Operation Description Assigns a new ID to unique records that don't have an ID or to all members of a group that don't have an ID. In addition, the assign operation copies an existing ID if a member of a match group already has an ID. Each record is assigned a value. • Assign • • Records in a match group where one record had an input unique ID will share the value with other records in the match group which had no input value. The first value encountered will be shared. Order affects this; if you have a priority field that can be sequenced using ascending order, place a Prioritization post-match operation prior to the Unique ID operation. Records in a match group where two or more records had different unique ID input values will each keep their input value. If all of the records in a match group do not have an input unique ID value, then the next available ID will be assigned to each record in the match group. If the GROUP_NUMBER input field is used, then records with the same group number must appear consecutively in the data collection. Note: Use the GROUP_NUMBER input field only when processing a break group that may contain smaller match groups. If the GROUP_NUMBER field is not specified, Unique ID assumes that the entire collection is one group. 449 2011-06-09
- 450. Data Quality Operation Description Performs both an Assign and a Combine operation. AssignCom bine Each record is assigned a value. • Records that did not have an input unique ID value and are not found to match another record containing an input unique ID value will have the next available ID assigned to it. These are "add" records that could be unique records or could be matches, but not to another record that had previously been assigned a unique ID value. • Records in a match group where one or more records had an input unique ID with the same or different values will share the first value encountered with all other records in the match group. Order affects this; if you have a priority field that can be sequenced using ascending order, place a Prioritization post-match operation prior to the Unique ID operation. If the GROUP_NUMBER input field is used, then records with the same group number must appear consecutively in the data collection. Note: Use the GROUP_NUMBER input field only when processing a break group that may contain smaller match groups. If the GROUP_NUMBER field is not specified, Unique ID assumes that the entire collection is one group. Ensures that records in the same match group have the same Unique ID. For example, this operation could be used to assign all the members of a household the same unique ID. Specifically, if a household has two members that share a common unique ID, and a third person moves in with a different unique ID, then the Combine operation could be used to assign the same ID to all three members. Combine The first record in a match group that has a unique ID is the record with the highest priority. All other records in the match group are given this record’s ID (assuming the record is not protected). The Combine operation does not assign a unique ID to any record that does not already have a unique ID. It only combines the unique ID of records in a match group that already have a unique ID. If the GROUP_NUMBER input field is used, then records with the same group number must appear consecutively in the data collection. Note: Use the GROUP_NUMBER input field only when processing a break group that may contain smaller match groups. If the GROUP_NUMBER field is not specified, Unique ID assumes that the entire collection is one group. 450 2011-06-09
- 451. Data Quality Operation Description Deletes unique IDs from records that no longer need them, provided that they are not protected from being deleted. If you are using a file and are recycling IDs, this ID is added to the file. When performing a delete, records with the same unique ID should be grouped together. Delete When Match detects that a group of records with the same unique ID is about to be deleted: • If any of the records are protected, all records in the group are assumed to be protected. • If recycling is enabled, the unique ID will be recycled only once, even though a group of records had the same ID. Changes a split group's unique records, so that the records that do not belong to the same match group will have a different ID. The record with the group's highest priority will keep its unique ID. The rest will be assigned new unique IDs. For this operation, you must group your records by unique ID, rather than by match group number. For example: • Records in a match group where two or more records had different unique ID input values or blank values will each retain their input value, filled or blank depending on the record. • Records that did not have an input unique ID value and did not match any record with an input unique ID value will have a blank unique ID on output. • Records that came in with the same input unique ID value that no longer are found as matches have the first record output with the input value. Subsequent records are assigned new unique ID values. Split 16.4.9.2.2 Unique ID protection The output for the unique ID depends on whether an input field in that record has a value that indicates that the ID is protected. • • If the protected unique ID field is not mapped as an input field, Match assumes that none of the records are protected. There are two valid values allowed in this field: Y and N. Any other value is converted to Y. A value of N means that the unique ID is not protected and the ID posted on output may be different from the input ID. a value of Y means that the unique ID is protected and the ID posted on output will be the same as the input ID. • If the protected unique ID field is mapped as an input field, a value other than N means that the record's input data will be retained in the output unique ID field. These rules for protected fields apply to all unique ID processing operations. 451 2011-06-09
- 452. Data Quality 16.4.9.2.3 Unique ID limitations Because some options in the unique ID operation are based on reading a file or referring to a field value, there may be implications for when you are running a multi-server or real-time server environment and sharing a unique ID file. • • If you are reading from or writing to a file, the unique ID file must be on a shared file system. Recycled IDs are used in first-in, first-out order. When Match recycles an ID, it does not check whether the ID is already present in the file. You must ensure that a particular unique ID value is not recycled more than once. 16.4.9.2.4 To assign unique IDs using a file 1. In the Unique ID option group, select the Value from file option. 2. Set the file name and path in the File option. This file must be an XML file and must adhere to the following structure: <UniqueIdSession> <CurrentUniqueId>477</CurrentUniqueId> </UniqueIdSession> Note: The value of 477 is an example of a starting value. However, the value must be 1 or greater. 16.4.9.2.5 To assign a unique ID using a constant Similar to using a file, you can assign a starting unique ID by defining that value. 1. Select the Constant value option. 2. Set the Starting value option to the desired ID value. 16.4.9.2.6 Assign unique IDs using a field The Field option allows you to send the starting unique ID through a field in your data source or from a User-Defined transform, for example. The starting unique ID is passed to the Match transform before the first new unique ID is requested. If no unique ID is received, the starting number will default to 1. Caution: Use caution when using the Field option. The field that you use must contain the unique ID value you want to begin the sequential numbering with. This means that each record you process must contain this field, and each record must have the same value in this field. For example, suppose the value you use is 100,000. During processing, the first record or match group will have an ID of 100,001. The second record or match group receives an ID of 100,002, and so on. The value in the first record that makes it to the Match transform contains the value where the incrementing begins. 452 2011-06-09
- 453. Data Quality There is no way to predict which record will make it to the Match transform first (due to sorting, for example); therefore, you cannot be sure which value the incrementing will begin at. To assign unique IDs using a field 1. Select the Field option. 2. In the Starting unique ID field option, select the field that contains the starting unique ID value. 16.4.9.2.7 To assign unique IDs using GUID You can use Globally Unique Identifiers (GUID) as unique IDs. • Select the GUID option. Note: GUID is also known as the Universal Unique Identifier (UUID). The UUID variation used for unique ID is a time-based 36-character string with the format: TimeLow-TimeMid-TimeHighAndVersionClockSeqAndReservedClockSeqLow-Node For more information about UUID, see the Request for Comments (RFC) document. Related Topics • UUID RFC: http://www.ietf.org/rfc/rfc4122.txt 16.4.9.2.8 To recycle unique IDs If unique IDs are dropped during the Delete processing option, you can write those IDs back to a file to be used later. 1. In the Unique ID option group, set the Processing operation option to Delete. 2. Select the Value from file option. 3. Set the file name and path in the File option. 4. Set the Recycle unique IDs option to Yes. This is the same file that you might use for assigning a beginning ID number. Use your own recycled unique IDs If you have some IDs of your own that you would like to recycle and use in a data flow, you can enter them in the file you want to use for recycling IDs and posting a starting value for your IDs. Enter these IDs in an XML tag of <R></R>. For example: <UniqueIdSession> <CurrentUniqueId>477</CurrentUniqueId> <R>214</R> <R>378</R> </UniqueIdSession> 453 2011-06-09
- 454. Data Quality 16.4.9.2.9 Destination protection The Best Record and Unique ID operations in the Match transform offer you the power to modify existing records in your data. There may be times when you would like to protect data in particular records or data in records from particular input sources from being overwritten. The Destination Protection tab in these Match transform operations allow you the ability to protect data from being modified. To protect destination records through fields 1. In the Destination Protection tab, select Enable destination protection. 2. Select a value in the Default destination protection option drop-down list. This value determines whether a destination is protected if the destination protection field does not have a valid value. 3. Select the Specify destination protection by field option, and choose a field from the Destination protection field drop-down list (or Unique ID protected field) . The field you choose must have a Y or N value to specify the action. Any record that has a value of Y in the destination protection field will be protected from being modified. To protect destination records based on input source membership You must add an Input Source operation and define input sources before you can complete this task. 1. In the Destination Protection tab, select Enable destination protection. 2. Select a value in the Default destination protection option drop-down list. This value determines whether a destination (input source) is protected if you do not specifically define the source in the table below. 3. Select the Specify destination protection by source option. 4. Select an input source from the first row of the Source name column, and then choose a value from the Destination protected (or Unique ID protected) column. Repeat for every input source you want to set protection for. Remember that if you do not specify for every source, the default value will be used. 16.4.9.3 Group statistics The Group Statistics post-match operation should be added after any match level and any post-match operation for which you need statistics about your match groups or your input sources. This operation can also counts statistics from logical input sources that you have already identified with values in a field (pre-defined) or from logical sources that you specify in the Input Sources operation. 454 2011-06-09
- 455. Data Quality This operation also allows you to exclude certain logical sources based on your criteria. Note: If you choose to count input source statistics in the Group Statistics operation, Match will also count basic statistics about your match groups. Group statistics fields When you include a Group Statistics operation in your Match transform, the following fields are generated by default: • • • • GROUP_COUNT GROUP_ORDER GROUP_RANK GROUP_TYPE In addition, if you choose to generate source statistics, the following fields are also generated and available for output: • • • • SOURCE_COUNT SOURCE_ID SOURCE_ID_COUNT SOURCE_TYPE_ID Related Topics • Reference Guide: Transforms, Match, Output fields • Management Console Guide: Data Quality Reports, Match Source Statistics Summary report 16.4.9.3.1 To generate only basic statistics This task will generate statistics about your match groups, such as how many records in each match group, which records are masters or subordinates, and so on. 1. Add a Group Statistics operation to each match level you want, by selecting Post Match Processing in a match level, clicking the Add button, and selecting Group Statistics. 2. Select Generate only basic statistics. 3. Click the Apply button to save your changes. 16.4.9.3.2 To generate statistics for all input sources Before you start this task, be sure that you have defined your input sources in the Input Sources operation. Use this procedure if you are interested in generating statistics for all of your sources in the job. 1. Add a Group Statistics operation to the appropriate match level. 2. Select the Generate source statistics from input sources option. This will generate statistics for all of the input sources you defined in the Input Sources operation. 455 2011-06-09
- 456. Data Quality 16.4.9.3.3 To count statistics for input sources generated by values in a field For this task, you do not need to define input sources with the Input Sources operation. You can specify input sources for Match using values in a field. Using this task, you can generate statistics for all input sources identified through values in a field, or you can generate statistics for a sub-set of input sources. 1. Add a Group Statistics operation to the appropriate match level. 2. Select the Generate source statistics from source values option. 3. Select a field from the Logical source field drop-down list that contains the values for your logical sources. 4. Enter a value in the Default logical source value field. This value is used if the logical source field is empty. 5. Select one of the following: Option Description Count all sources Select to count all sources. If you select this option, you can click the Apply button to save your changes. This task is complete. Choose sources Select to define a sub-set of input sources to count. If you select this option, to count you can proceed to step 6 in the task. 6. Choose the appropriate value in the Default count flag option. Choose Yes to count any source not specified in the Manually define logical source count flags table. If you do not specify any sources in the table, you are, in effect, counting all sources. 7. Select Auto-generate sources to count sources based on a value in a field specified in the Predefined count flag field option. If you do not specify any sources in the Manually define logical source count flags table, you are telling the Match transform to count all sources based on the (Yes or No) value in this field. 8. In the Manually define logical source count flags table, add as many rows as you need to include all of the sources you want to count. Note: This is the first thing the Match transform looks at when determining whether to count sources. 9. Add a source value and count flag to each row, to tell the Match transform which sources to count. Tip: If you have a lot of sources, but you only want to count two, you could speed up your set up time by setting the Default count flag option to No, and setting up the Manually define logical source count flags table to count those two sources. Using the same method, you can set up Group Statistics to count everything and not count only a couple of sources. 456 2011-06-09
- 457. Data Quality 16.4.9.4 Output flag selection By adding an Output Flag Selection operation to each match level (Post Match Processing) you want, you can flag specific record types for evaluation or routing downstream in your data flow. Adding this operation generates the Select_Record output field for you to include in your output schema. This output field is populated with a Y or N depending on the type of record you select in the operation. Your results will appear in the Match Input Source Output Select report. In that report, you can determine which records came from which source or source group and how many of each type of record were output per source or source group. Record type Description Unique Records that are not members of any match group. No matching records were found. These can be from sources with a normal or special source. Single source masters Highest ranking member of a match group whose members all came from the same source. Can be from normal or special sources. Single source subordinates A record that came from a normal or special source and is a subordinate member of a match group. Multiple source masters Highest ranking member of a match group whose members came from two or more sources. Can be from normal or special sources. Multiple source subordinates A subordinate record of a match group that came from a normal or special source whose members came from two or more sources. Suppression matches Subordinate member of a match group that includes a higher-priority record that came from a suppress-type source. Can be from normal or special source. Suppression uniques Records that came from a suppress source for which no matching records were found. Suppression masters A record that came from a suppress source and is the highest ranking member of a match group. Suppression subordinates A record that came from a suppress-type source and is a subordinate member of a match group. 16.4.9.4.1 To flag source record types for possible output 1. In the Match editor, for each match level you want, add an Output Flag Select operation. 2. Select the types of records for which you want to populate the Select_Record field with Y. 457 2011-06-09
- 458. Data Quality The Select_Record output field can then be output from Match for use downstream in the data flow. This is most helpful if you later want to split off suppression matches or suppression masters from your data (by using a Case tranform, for example). 16.4.10 Association matching Association matching combines the matching results of two or more match sets (transforms) to find matches that could not be found within a single match set. You can set up association matching in the Associate transform. This transform acts as another match set in your data flow, from which you can derive statistics. This match set has two purposes. First, it provides access to any of the generated data from all match levels of all match sets. Second, it provides the overlapped results of multiple criteria, such as name and address, with name and SSN, as a single ID. This is commonly referred to as association matching. Group numbers The Associate transform accepts a group number field, generated by the Match transforms, for each match result that will be combined. The transform can then output a new associated group number. The Associate transform can operate either on all the input records or on one data collection at a time. The latter is needed for real-time support. Example: Association example Say you work at a technical college and you want to send information to all of the students prior to the start of a new school year. You know that many of the students have a temporary local address and a permanent home address. In this example, you can match on name, address, and postal code in one match set, and match on name and Social Security number (SSN), which is available to the technical college on every student, in another match set. Then, the Associate transform combines the two match sets to build associated match groups. This lets you identify people who may have multiple addresses, thereby maximizing your one-to-one marketing and mailing efforts. 16.4.11 Unicode matching Unicode matching lets you match Unicode data. You can process any non-Latin1 Unicode data, with special processing for Chinese, Japanese, Korean and Taiwanese (or CJKT) data. 458 2011-06-09
- 459. Data Quality Chinese, Japanese, Korean, and Taiwanese matching Regardless of the country-specific language, the matching process for CJKT data is the same. For example, the Match transform: • Considers half-width and full-width characters to be equal. • Considers native script numerals and Arabic numerals to be equal. It can interpret numbers that are written in native script. This can be controlled with the Convert text to numbers option in the Criteria options group. • Includes variations for popular, personal, and firm name characters in the referential data. • Considers firm words, such as Corporation or Limited, to be equal to their variations (Corp. or Ltd.) during the matching comparison process. To find the abbreviations, the transform uses native script variations of the English alphabets during firm name matching. • Ignores commonly used optional markers for province, city, district, and so on, in address data comparison. • Intelligently handles variations in a building marker. Japanese-specific matching capabilities With Japanese data, the Match transform considers: • Block data markers, such as chome and banchi, to be equal to those used with hyphenated data. • Words with or without Okurigana to be equal in address data. • Variations of no marker, ga marker, and so on, to be equal. • Variations of a hyphen or dashed line to be equal. Unicode match limitations The Unicode match functionality does not: • Perform conversions of simplified and traditional Chinese data. • Match between non-phonetic scripts like kanji, simplified Chinese, and so on. Route records based on country ID before matching Before sending Unicode data into the matching process, you must first, as best you can, separate out the data by country to separate match transforms. This can be done by using a Case transform to route country data based on the country ID. Tip: The Match wizard can do this for you when you use the multi-national strategy. Inter-script matching Inter-script matching allows you to process data that may contain more than one script by converting the scripts to Latin1. For example one record has Latin1 and other has katakana, or one has Latin and other has Cyrillic. Select Yes to enable Inter-script matching. If you prefer to process the data without converting it to Latin1, leave the Inter-script Matching option set No. Here are two examples of names matched using inter-script matching: 459 2011-06-09
- 460. Data Quality Name Can be matched to... Viktor Ivanov Виктор Иванов Takeda Noburu スッセ フレ Locale The Locale option specifies the locale setting for the criteria field. Setting this option is recommended if you plan to use the Text to Numbers feature to specify the locale of the data for locale-specific text-to-number conversion for the purpose of matching. Here a four examples of text-to-number conversion: Language Text Numbers French quatre mille cinq cents soixante-sept 4567 German dreitausendzwei 3002 Italian cento 100 Spanish ciento veintisiete 127 For more information on these matching options, see the Match Transform section of the Reference Guide 16.4.11.1 To set up Unicode matching 1. Use a Case transform to route your data to a Match transform that handles that type of data. 2. Open the AddressJapan_MatchBatchMatch transform configuration, and save it with a different name. 3. Set the Match engine option in the Match transform options to a value that reflects the type of data being processed. 4. Set up your criteria and other desired operations. For more information on Match Criteria options, see the Match Transform section of the Reference Guide. Example: • • • 460 When possible, use criteria for parsed components for address, firm, and name data, such as Primary_Name or Person1_Family_Name1. If you have parsed address, firm, or name data that does not have a corresponding criteria, use the Address_Data1-5, Firm_Data1-3, and Name_Data1-3 criteria. For all other data that does not have a corresponding criteria, use the Custom criteria. 2011-06-09
- 461. Data Quality 16.4.12 Phonetic matching You can use the Double Metaphone or Soundex functions to populate a field and use it for creating break groups or use it as a criteria in matching. Match criteria There are instances where using phonetic data can produce more matches when used as a criteria, than if you were to match on other criteria such as name or firm data. Matching on name field data produces different results than matching on phonetic data. For example: Name Comparison score Smith 72% similar Smythe Name Phonetic key (primary) Smith Comparison score SMO 100% similar Smythe SMO Criteria options If you intend to match on phonetic data, set up the criteria options this way Option Compare algorithm Field Check for transposed characters No Intials adjustment score 0 Substring adjustment score 0 Abbreviation adjustment score 461 Value 0 2011-06-09
- 462. Data Quality Match scores If you are matching only on the phonetic criteria, set your match score options like this: Option Value Match score 100 No match score 99 If you are matching on multiple criteria, including a phonetic criteria, place the phonetic criteria first in the order of criteria and set your match score options like this: Option Value Match score 101 No match score 99 Blank fields Remember that when you use break groups, records that have no value are not in the same group as records that have a value (unless you set up matching on blank fields). For example, consider the following two input records: Mr Johnson 100 Main St La Crosse WI 54601 Scott Johnson 100 Main St La Crosse WI 54601 After these records are processed by the Data Cleanse transform, the first record will have an empty first name field and, therefore, an empty phonetic field. This means that there cannot be a match, if you are creating break groups. If you are not creating break groups, there cannot be a match if you are not blank matching. Length of data The length you assign to a phonetic function output is important. For example: First name (last name) S (Johnson) S Scott (Johnson) 462 Output SKT 2011-06-09
- 463. Data Quality Suppose these two records represent the same person. In this example, if you break on more than one character, these records will be in different break groups, and therefore will not be compared. 16.4.13 Set up for match reports We offer many match reports to help you analyze your match results. For more information about these individual reports, see the Management Console Guide. Include Group Statistics in your Match transform If you are generating the Match Source Statistics Summary report, you must have a Group Statistics operation included in your Match and Associate transform(s). If you want to track your input source statistics, you may want to include an Input Sources operation in the Match transform to define your sources and, in a Group Statistics operation select to generate statistics for your input sources. Note: You can also generate input source statistics in the Group Statistics operation by defining input sources using field values. You do not necessarily need to include an Input Sources operation in the Match transform. Turn on report data generation in transforms In order to generate the data you want to see in match reports other than the Match Source Statistics report, you must set the Generate report statistics option to Yes in the Match and Associate transform(s). By turning on report data generation, you can get information about break groups, which criteria were instrumental in creating a match, and so on. Note: Be aware that turning on the report option can have an impact on your processing performance. It's best to turn off reports after you have thoroughly tested your data flow. Define names for match sets, levels, and operations To get the most accurate data in your reports, make sure that you have used unique names in the Match and Associate transforms for your match sets, levels, and each of your pre- and post-match operations, such as Group Prioritization and Group Statistics. This will help you better understand which of these elements is producing the data you are looking at. Insert appropriate output fields There are three output fields you may want to create in the Match transform, if you want that data posted in the Match Duplicate Sample report. They are: • • 463 Match_Type Group_Number 2011-06-09
- 464. Data Quality • Match_Score 16.5 Address Cleanse This section describes how to prepare your data for address cleansing, how to set up address cleansing, and how to understand your output after processing. Related Topics • How address cleanse works • Prepare your input data • Determine which transform(s) to use • Identify the country of destination • Set up the reference files • Define the standardization options • Supported countries (Global Address Cleanse) 16.5.1 How address cleanse works Address cleanse provides a corrected, complete, and standardized form of your original address data. With the USA Regulatory Address Cleanse transform and for some countries with the Global Address Cleanse transform, address cleanse can also correct or add postal codes. With the DSF2 Walk Sequencer transform, you can add walk sequence information to your data. What happens during address cleanse? The USA Regulatory Address Cleanse transform and the Global Address Cleanse transform cleanse your data in the following ways: • • • • Verify that the locality, region, and postal codes agree with one another. If your data has just a locality and region, the transform usually can add the postal code and vice versa (depending on the country). Standardize the way the address line looks. For example, they can add or remove punctuation and abbreviate or spell-out the primary type (depending on what you want). Identify undeliverable addresses, such as vacant lots and condemned buildings (USA records only). Assign diagnostic codes to indicate why addresses were not assigned or how they were corrected. (These codes are included in the Reference Guide). Reports The USA Regulatory Address Cleanse transform creates the USPS Form 3553 (required for CASS) and the NCOALink Summary Report. The Global Address Cleanse transform creates reports about 464 2011-06-09
- 465. Data Quality your data including the Canadian SERP—Statement of Address Accuracy Report, the Australia Post’s AMAS report, and the New Zealand SOA Report. Related Topics • The transforms • Input and output data and field classes • Prepare your input data • Determine which transform(s) to use • Define the standardization options • Reference Guide: Supported countries • Reference Guide: Data Quality Appendix, Country ISO codes and assignment engines • Reference Guide: Data Quality Fields, Global Address Cleanse fields • Reference Guide: Data Quality Fields, USA Regulatory Address Cleanse fields 16.5.1.1 The transforms The following table lists the address cleanse transforms and their purpose. Transform DSF2 Walk Sequencer Global Address Cleanse and engines 465 Description When you perform DSF2 walk sequencing in the software, the software adds delivery sequence information to your data, which you can use with presorting software to qualify for walk-sequence discounts. Remember: The software does not place your data in walk sequence order. Cleanses your address data from any of the supported countries (not for U.S. certification). You must set up the Global Address Cleanse transform in conjunction with one or more of the Global Address Cleanse engines (Canada, Global Address, or USA). With this transform you can create Canada Post's Software Evaluation and Recognition Program (SERP)—Statement of Address Accuracy Report, Australia Post's Address Matching Processing Summary report (AMAS), and the New Zealand Statement of Accuracy (SOA) report. 2011-06-09
- 466. Data Quality Transform Description USA Regulatory Address Cleanse Identifies, parses, validates, and corrects USA address data (within the Latin 1 code page) according to the U.S. Coding Accuracy Support System (CASS). Can create the USPS Form 3553 and output many useful codes to your records. You can also run in a non-certification mode as well as produce suggestion lists. Some options include: DPV, DSF2 (augment), eLOT, EWS, GeoCensus, LACSLink, NCOALink, RDI, SuiteLink, suggestion lists (not for certification), and Z4Change. Global Suggestion Lists Offers suggestions for possible address matches for your USA, Canada, and global address data. This transform is usually used for real time processing and does not standardize addresses. Use a Country ID transform before this transform in the data flow. Also, if you want to standardize your address data, use the Global Address Cleanse transform after the Global Suggestion Lists transform in the data flow. Country ID Identifies the country of destination for the record and outputs an ISO code. Use this transform before the Global Suggestion Lists transform in your data flow. (It is not necessary to place the Country ID transform before the Global Address Cleanse or the USA Regulatory Address Cleanse transforms.) 16.5.1.2 Input and output data and field classes Input data The address cleanse transforms accept discrete, multiline, and hybrid address line formats. Output data There are two ways that you can set the software to handle output data. Most use a combination of both. 466 2011-06-09
- 467. Data Quality Concept Multiline Discrete Description The first method is useful when you want to keep output address data in the same arrangement of fields as were input. The software applies intelligent abbreviation, when necessary, to keep the data within the same field lengths. Data is capitalized and standardized according to the way you set the standardization style options. The second method is useful when you want the output addresses broken down into smaller elements than you input. Also, you can retrieve additional fields created by the software, such as the error/status code. The style of some components is controlled by the standardization style options; most are not. The software does not apply any intelligent abbreviation to make components fit your output fields. When you set up the USA Regulatory Address Cleanse transform or the Global Address Cleanse transform, you can include output fields that contain specific information: Generated Field Address Class Generated Field Class Parsed: Contains the parsed input with some standardization applied. The fields subjected to standardization are locality, region, and postcode. Delivery Best: Contains the parsed data when the address is unassigned or the corrected data for an assigned address. Corrected: Contains the assigned data after directory lookups and will be blank if the address is not assigned. Dual Parsed, Best, and Corrected: Contain the DUAL address details that were available on input. Parsed: Contains the parsed input with some standardization applied. Official Best: Contains the information from directories defined by the Postal Service when an address is assigned. Contains the parsed input when an address is unassigned. Corrected: Contains the information from directories defined by the Postal Service when an address is assigned and will be blank if the address is not assigned. 16.5.2 Prepare your input data 467 2011-06-09
- 468. Data Quality Before you start address cleansing, you must decide which kind of address line format you will input. Both the USA Regulatory Address Cleanse transform and the Global Address Cleanse transform accept input data in the same way. Caution: The USA Regulatory Address Cleanse Transform does not accept Unicode data. If an input record has characters outside the Latin1 code page (character value is greater than 255), the USA Regulatory Address Cleanse transform will not process that data. Instead, the input record is sent to the corresponding standardized output field without any processing. No other output fields (component, for example) will be populated for that record. If your Unicode database has valid U.S. addresses from the Latin1 character set, the USA Regulatory Address Cleanse transform processes as usual. Accepted address line formats The following tables list the address line formats: multiline, hybrid, and discrete. Note: For all multiline and hybrid formats listed, you are not required to use all the multiline fields for a selected format (for example Multiline1-12). However, you must start with Multiline1 and proceed consecutively. You cannot skip numbers, for example, from Multiline1 to Multiline3. Multiline and multiline hybrid formats Example 1 Example 2 Example 3 Example 4 Example 5 Multiline1 Multiline1 Multiline1 Multiline1 Multiline1 Multiline2 Multiline2 Multiline2 Multiline2 Multiline2 Multiline3 Multiline3 Multiline3 Multiline3 Multiline3 Multiline4 Multiline4 Multiline4 Multiline4 Multiline4 Multiline5 Multiline5 Locality3 Multiline5 Multiline5 Multiline6 Multiline6 Locality2 Locality2 Multiline6 Multiline7 Multiline7 Locality1 Locality1 Locality1 Multiline8 Lastline Region1 Region1 Region1 Country (Optional) Country (Optional) Postcode (Global) or Postcode1 (USA Reg.) Postcode (Global) or Postcode1 (USA Reg.) Postcode (Global) or Postcode1 (USA Reg.) Country (Optional) Country (Optional) Country (Optional) Discrete line formats Example 1 Example 2 Example 3 Example 4 Address_Line Address_Line Address_Line Address_Line 468 2011-06-09
- 469. Data Quality Discrete line formats Example 1 Example 2 Example 3 Example 4 Lastline Locality3 (Global) Locality2 Locality1 Country (Optional) Locality2 Locality1 Region1 Locality1 Region1 Postcode (Global) or Postcode1 (USA Reg.) Region1 Postcode (Global) or Postcode1 (USA Reg.) Country (Optional) Postcode (Global) or Postcode1 (USA Reg.) Country (Optional) Country (Optional) 16.5.3 Determine which transform(s) to use You can choose from a variety of address cleanse transforms based on what you want to do with your data. There are transforms for cleansing global and/or U.S. address data, cleansing based on USPS regulations, using business rules to cleanse data and cleansing global address data transactionally. Related Topics • Cleanse global address data • Cleanse U.S. data only • Cleanse U.S. data and global data • Cleanse address data using multiple business rules • Cleanse your address data transactionally 16.5.3.1 Cleanse global address data To cleanse your address data for any of the software-supported countries (including Canada for SERP, Software Evaluation and Recognition Program, certification and Australia for AMAS, Address Matching Approval System, certification), use the Global Address Cleanse transform in your project with one or more of the following engines: • 469 Canada 2011-06-09
- 470. Data Quality • • Global Address USA Tip: Cleansing U.S. data with the USA Regulatory Address Cleanse transform is usually faster than with the Global Address Cleanse transform and USA engine. This scenario is usually true even if you end up needing both transforms. You can also use the Global Address Cleanse transform with the Canada, USA, Global Address engines in a real time data flow to create suggestion lists for those countries. Start with a sample transform configuration The software includes a variety of Global Address Cleanse sample transform configurations (which include at least one engine) that you can copy to use for a project. Related Topics • Supported countries (Global Address Cleanse) • Cleanse U.S. data and global data • Reference Guide: Transforms, Transform configurations 16.5.3.2 Cleanse U.S. data only To cleanse U.S. address data, use the USA Regulatory Address Cleanse transform for the best results. With this transform, and with DPV, LACSLink, and SuiteLink enabled, you can produce a CASS-certified mailing and produce a USPS Form 3553. If you do not intend to process CASS-certified lists, you should still use the USA Regulatory Address Cleanse transform for processing your U.S. data. Using the USA Regulatory Address Cleanse transform on U.S. data is more efficient than using the Global Address Cleanse transform. With the USA Regulatory Address Cleanse transform you can add additional information to your data such as DSF2, EWS, eLOT, NCOALink, and RDI. And you can process records one at a time by using suggestion lists. Start with a sample transform configuration The software includes a variety of USA Regulatory Address Cleanse sample transform configurations that can help you set up your projects. Related Topics • Reference Guide: Transforms, Data Quality transforms, Transform configurations • Introduction to suggestion lists 470 2011-06-09
- 471. Data Quality 16.5.3.3 Cleanse U.S. data and global data What should you do when you have U.S. addresses that need to be certified and also addresses from other countries in your database? In this situation, you should use both the Global Address Cleanse transform and the USA Regulatory Address Cleanse transform in your data flow. Tip: Even if you are not processing U.S. data for USPS certification, you may find that cleansing U.S. data with the USA Regulatory Address Cleanse transform is faster than with the Global Address Cleanse transform and USA engine. 16.5.3.4 Cleanse address data using multiple business rules When you have two addresses intended for different purposes (for example, a billing address and a shipping address), you should use two of the same address cleanse transforms in a data flow. One or two engines? When you use two Global Address Cleanse transforms for data from the same country, they can share an engine. You do not need to have two engines of the same kind. If you use one engine or two, it does not affect the overall processing time of the data flow. In this situation, however, you may need to use two separate engines (even if the data is from the same country). Depending on your business rules, you may have to define the settings in the engine differently for a billing address or for a shipping address. For example, in the Standardization Options group, the Output Country Language option can convert the data used in each record to the official country language or it can preserve the language used in each record. For example, you may want to convert the data for the shipping address but preserve the data for the billing address. 16.5.3.5 Cleanse your address data transactionally The Global Suggestion Lists transform, best used in transactional projects, is a way to complete and populate addresses with minimal data, or it can offer suggestions for possible matches. For example, the Marshall Islands and the Federated States of Micronesia were recently removed from the USA Address directory. Therefore, if you previously used the USA engine, you'll now have to use the Global Address engine. The Global Suggestion Lists transform can help identify that these countries are no longer in the USA Address directory. 471 2011-06-09
- 472. Data Quality This easy address-entry system is ideal in call center environments or any transactional environment where cleansing is necessary at the point of entry. It's also a beneficial research tool when you need to manage bad addresses from a previous batch process. Place the Global Suggestion Lists transform after the Country ID transform and before a Global Address Cleanse transform that uses a Global Address, Canada, and/or USA engine. Integrating functionality Global Suggestion Lists functionality is designed to be integrated into your own custom applications via the Web Service. If you are a programmer looking for details about how to integrate this functionality, see "Integrate Global Suggestion Lists" in the Integrator's Guide. Start with a sample transform configuration Data Quality includes a Global Suggestion Lists sample transform that can help you when setting up a project. Related Topics • Introduction to suggestion lists 16.5.4 Identify the country of destination The Global Address Cleanse transform includes Country ID processing. Therefore, you do not need to place a Country ID transform before the Global Address Cleanse transform in your data flow. In the Country ID Options option group of the Global Address Cleanse transform, you can define the country of destination or define whether you want to run Country ID processing. Constant country If all of your data is from one country, such as Australia, you do not need to run Country ID processing or input a discrete country field. You can tell the Global Address Cleanse transform the country and it will assume all records are from this country (which may save processing time). Assign default You'll want to run Country ID processing if you are using two or more of the engines and your input addresses contain country data (such as the two-character ISO code or a country name), or if you are using only one engine and your input source contains many addresses that cannot be processed by that engine. Addresses that cannot be processed are not sent to the engine. The transform will use the country you specify in this option group as a default. Related Topics • To set a constant country • Set a default country 472 2011-06-09
- 473. Data Quality 16.5.5 Set up the reference files The USA Regulatory Address Cleanse transform and the Global Address Cleanse transform and engines rely on directories (reference files) in order to cleanse your data. Directories To correct addresses and assign codes, the address cleanse transforms rely on databases called postal directories. The process is similar to the way that you use the telephone directory. A telephone directory is a large table in which you look up something you know (a person's name) and read off something you don't know (the phone number). In the process of looking up the name in the phone book, you may discover that the name is spelled a little differently from the way you thought. Similarly, the address cleanse transform looks up street and city names in the postal directories, and it corrects misspelled street and city names and other errors. Sometimes it doesn't work out. We've all had the experience of looking up someone and being unable to find their listing. Maybe you find several people with a similar name, but you don't have enough information to tell which listing was the person you wanted to contact. This type of problem can prevent the address cleanse transforms from fully correcting and assigning an address. Besides the basic address directories, there are many specialized directories that the USA Regulatory Address Cleanse transform uses: • • • • • • • • • • DPV® DSF2® Early Warning System (EWS) eLOT® GeoCensus LACSLink® NCOALink® RDI™ SuiteLink™ Z4Change These features help extend address cleansing beyond the basic parsing and standardizing. Define directory file locations You must tell the transform or engine where your directory (reference) files are located in the Reference Files option group. Your system administrator should have already installed these files to the appropriate locations based on your company's needs. Caution: Incompatible or out-of-date directories can render the software unusable. The system administrator must install weekly, monthly or bimonthly directory updates for the USA Regulatory Address Cleanse Transform; monthly directory updates for the Australia and Canada engines; and quarterly directory updates for the Global Address engine to ensure that they are compatible with the current software. 473 2011-06-09
- 474. Data Quality Substitution files If you start with a sample transform, the Reference Files options are filled in with a substitution variable (such as $$RefFilesAddressCleanse) by default. These substitution variables point to the reference data folder of the software directory by default. You can change that location by editing the substitution file associated with the data flow. This change is made for every data flow that uses that substitution file. Related Topics • USPS DPV® • USPS DSF2® • DSF2 walk sequencing • Early Warning System (EWS) • USPS eLOT® • GeoCensus (USA Regulatory Address Cleanse) • LACSLink® • NCOALink® overview • USPS RDI® • SuiteLink™ • Z4Change (USA Regulatory Address Cleanse) 16.5.5.1 View directory expiration dates in the trace log You can view directory expiration information for a current job in the trace log. To include directory expiration information in the trace log, perform the following steps. • • • Right click on the applicable job icon in Designer and select Execute. In the Execution Properties window, open the Execution Options tab (it should already be open by default). Select Print all trace messages. Related Topics • Using logs 16.5.6 Define the standardization options 474 2011-06-09
- 475. Data Quality Standardization changes the way the data is presented after an assignment has been made. The type of change depends on the options that you define in the transform. These options include casing, punctuation, sequence, abbreviations, and much more. It helps ensure the integrity of your databases, makes mail more deliverable, and gives your communications with customers a more professional appearance. For example, the following address was standardized for capitalization, punctuation, and postal phrase (route to RR). Input Output Multiline1 = route 1 box 44a Address_Line = RR 1 BOX 44A Multiline2 = stodard wisc Locality1 = STODDARD Region1 = WI Postcode1 = 54658 Global Address Cleanse transform In the Global Address Cleanse transform, you set the standardization options in the Standardization Options option group. You can standardize addresses for all countries and/or for individual countries (depending on your data). For example, you can have one set of French standardization options that standardize addresses within France only, and another set of Global standardization options that standardize all other addresses. USA Regulatory Address Cleanse transform If you use the USA Regulatory Address Cleanse transform, you set the standardization options on the "Options" tab in the Standardization Options section. Related Topics • Reference Guide: Transforms, Global Address Cleanse transform options (Standardization options) • Reference Guide: Transforms, USA Regulatory Address Cleanse (Standardization options) 16.5.7 Process Japanese addresses The Global Address Cleanse transform's Global Address engine parses Japanese addresses. The primary purpose of this transform and engine is to parse and normalize Japanese addresses for data matching and cleansing applications. Note: The Japan engine only supports kanji and katakana data. The engine does not support Latin data. 475 2011-06-09
- 476. Data Quality A significant portion of the address parsing capability relies on the Japanese address database. The software has data from the Ministry of Public Management, Home Affairs, Posts and Telecommunications (MPT) and additional data sources. The enhanced address database consists of a regularly updated government database that includes regional postal codes mapped to localities. Related Topics • Standard Japanese address format • Special Japanese address formats • Sample Japanese address 16.5.7.1 Standard Japanese address format A typical Japanese address includes the following components. Address component Japanese English Output field(s) Postal code 〒654-0153 654-0153 Postcode_Full Prefecture 兵庫県 Hyogo-ken Region1_Full City 神戸市 Kobe-shi Locality1_Full Ward 須磨区 Suma-ku Locality2_Full District 南落合 Minami Ochiai Locality3_Full Block number 1丁目 1 chome Primary_Name_Full1 Sub-block number 25番地 25 banchi Primary_Name_Full2 House number 2号 2 go Primary_Number_Full An address may also include building name, floor number, and room number. Postal code Japanese postal codes are in the nnn-nnnn format. The first three digits represent the area. The last four digits represent a location in the area. The possible locations are district, sub-district, block, 476 2011-06-09
- 477. Data Quality sub-block, building, floor, and company. Postal codes must be written with Arabic numbers. The post office symbol 〒 is optional. Before 1998, the postal code consisted of 3 or 5 digits. Some older databases may still reflect the old system. Prefecture Prefectures are regions. Japan has forty-seven prefectures. You may omit the prefecture for some well known cities. City Japanese city names have the suffix 市 (-shi). In some parts of the Tokyo and Osaka regions, people omit the city name. In some island villages, they use the island name with a suffix 島 (-shima) in place of the city name. In some rural areas, they use the county name with suffix 郡 (-gun) in place of the city name. Ward A city is divided into wards. The ward name has the suffix 区(-ku). The ward component is omitted for small cities, island villages, and rural areas that don't have wards. District A ward is divided into districts. When there is no ward, the small city, island village, or rural area is divided into districts. The district name may have the suffix 町 (-cho/-machi), but it is sometimes omitted. 町 has two possible pronunciations, but only one is correct for a particular district. In very small villages, people use the village name with suffix 村 (-mura) in place of the district. When a village or district is on an island with the same name, the island name is often omitted. Sub-district Primarily in rural areas, a district may be divided into sub-districts, marked by the prefix 字 (aza-). A sub-district may be further divided into sub-districts that are marked by the prefix 小字 (koaza-), meaning small aza. koaza may be abbreviated to aza. A sub-district may also be marked by the prefix 大字 (oaza-), which means large aza. Oaza may also be abbreviated to aza. Here are the possible combinations: • oaza • aza • oaza and aza • aza and koaza • oaza and koaza Note: The characters 大字(oaza-), 字(aza-), and 小字 (koaza-) are frequently omitted. 477 2011-06-09
- 478. Data Quality Sub-district parcel A sub-district aza may be divided into numbered sub-district parcels, which are marked by the suffix 部 (-bu), meaning piece. The character 部 is frequently omitted. Parcels can be numbered in several ways: • Arabic numbers (1, 2, 3, 4, and so on) 石川県七尾市松百町8部3番地1号 • Katakana letters in iroha order (イ, ロ, ハ, ニ, and so on) • • 石川県小松市里川町ナ部23番地 Kanji numbers, which is very rare (甲, 乙, 丙, 丁, and so on) 愛媛県北条市上難波甲部 311 番地 Sub-division A rural district or sub-district (oaza/aza/koaza) is sometimes divided into sub-divisions, marked by the suffix 地割 (-chiwari) which means division of land. The optional prefix is 第 (dai-) The following address examples show sub-divisions: 岩手県久慈市旭町10地割1番地 岩手県久慈市旭町第10地割1番地 Block number A district is divided into blocks. The block number includes the suffix 丁目 (-chome). Districts usually have between 1 and 5 blocks, but they can have more. The block number may be written with a Kanji number. Japanese addresses do not include a street name. 東京都渋谷区道玄坂２丁目２５番地１２号 東京都渋谷区道玄坂二丁目２５番地１２号 Sub-block number A block is divided into sub-blocks. The sub-block name includes the suffix 番地 (-banchi), which means numbered land. The suffix 番地 (-banchi) may be abbreviated to just 番 (-ban). House number Each house has a unique house number. The house number includes the suffix 号 (-go), which means number. Block, sub-block, and house number variations Block, sub-block, and house number data may vary. Possible variations include the following: Dashes The suffix markers 丁目(chome), 番地 (banchi), and 号(go) may be replaced with dashes. 東京都文京区湯島2丁目18番地12号 478 2011-06-09
- 479. Data Quality 東京都文京区湯島2-18-12 Sometimes block, sub-block, and house number are combined or omitted. 東京都文京区湯島2丁目18番12号 東京都文京区湯島2丁目18番地12 東京都文京区湯島2丁目18-12 No block number Sometimes the block number is omitted. For example, this ward of Tokyo has numbered districts, and no block numbers are included. 二番町 means district number 2. 東京都 千代田区 二番町 9番地 6号 Building names Names of apartments or buildings are often included after the house number. When a building name includes the name of the district, the district name is often omitted. When a building is well known, the block, sub-block, and house number are often omitted. When a building name is long, it may be abbreviated or written using its acronym with English letters. The following are the common suffixes: Suffix Romanized Translation ビルディング birudingu building ビルヂング birudingu building ビル biru building センター senta- center プラザ puraza plaza パーク pa-ku park タワー tawa- tower 会館 kaikan hall 棟 tou building (unit) 479 2011-06-09
- 480. Data Quality Suffix Romanized Translation 庁舎 chousha government office building マンション manshon condominium 団地 danchi apartment complex アパート apa-to apartment 荘 sou villa 住宅 juutaku housing 社宅 shataku company housing 官舎 kansha official residence Building numbers Room numbers, apartment numbers, and so on, follow the building name. Building numbers may include the suffix 号室 (-goshitsu). Floor numbers above ground level may include the suffix 階 (-kai) or the letter F. Floor numbers below ground level may include the suffix 地下n 階 (chika n kai) or the letters BnF (where n represents the floor number). An apartment complex may include multiple buildings called Building A, Building B, and so on, marked by the suffix 棟 (-tou). The following address examples include building numbers. • Third floor above ground • 東京都千代田区二番町９番地６号 バウエプタ３ F Second floor below ground • 東京都渋谷区道玄坂 2-25-12 シティバンク地下 2 階 Building A Room 301 • 兵庫県神戸市須磨区南落合 1-25-10 須磨パークヒルズ A 棟 301 号室 Building A Room 301 兵庫県神戸市須磨区南落合 1-25-10 須磨パークヒルズ A-301 480 2011-06-09
- 481. Data Quality 16.5.7.2 Special Japanese address formats Hokkaido regional format The Hokkaido region has two special address formats: • super-block • numbered sub-districts Super-block A special super-block format exists only in the Hokkaido prefecture. A super-block, marked by the suffix 条 (-joh), is one level larger than the block. The super-block number or the block number may contain a directional 北 (north), 南 (south), 東 (east), or 西 (west). The following address example shows a super-block 4 Joh. 北海道札幌市西区二十四軒 4 条４丁目１３番地７号 Numbered sub-districts Another Hokkaido regional format is numbered sub-district. A sub-district name may be marked with the suffix 線 (-sen) meaning number instead of the suffix 字 (-aza). When a sub-district has a 線 suffix, the block may have the suffix 号 (-go), and the house number has no suffix. The following is an address that contains first the sub-district 4 sen and then a numbered block 5 go. 北海道旭川市西神楽４線５号３番地１１ Accepted spelling Names of cities, districts and so on can have multiple accepted spellings because there are multiple accepted ways to write certain sounds in Japanese. Accepted numbering When the block, sub-block, house number or district contains a number, the number may be written in Arabic or Kanji. For example, 二番町 means district number 2, and in the following example it is for Niban-cho. 東京都千代田区二番町九番地六号 P.O. Box addresses P.O. Box addresses contain the postal code, Locality1, prefecture, the name of the post office, the box marker, and the box number. Note: The Global Address Cleanse transform recognizes P.O. box addresses that are located in the Large Organization Postal Code (LOPC) database only. The address may be in one of the following formats: 481 2011-06-09
- 482. Data Quality • Prefecture, Locality1, post office name, box marker (私書箱), and P.O. box number. • Postal code, prefecture, Locality1, post office name, box marker (私書箱), and P.O. box number. The following address example shows a P.O. Box address: The Osaka Post Office Box marker #1 大阪府大阪市大阪支店私書箱1号 Large Organization Postal Code (LOPC) format The Postal Service may assign a unique postal code to a large organization, such as the customer service department of a major corporation. An organization may have up to two unique postal codes depending on the volume of mail it receives. The address may be in one of the following formats: • Address, company name • Postal code, address, company name The following is an example of an address in a LOPC address format. 100-8798 東京都千代田区霞が関１丁目3 - 2日本郵政 株式会社 16.5.7.3 Sample Japanese address This address has been processed by the Global Address Cleanse transform and the Global Address engine. Input 0018521 北海道札幌市北区北十条西 1丁目 12 番地 3 号創生ビル 1 階 101 号室札幌私書箱センター Address-line fields Primary_Name1 Primary_Type1 丁目 Primary_Name2 12 Primary_Type2 482 1 番地 2011-06-09
- 483. Data Quality Address-line fields Primary_Number 3 Primary_Number_Description 号 Building_Name1 創生ビル Floor_Number 1 Floor_Description 階 Unit_Number 101 Unit_Description 号室 Primary_Address 1丁目12番地3号 Secondary_Address 創生ビル 1階 101号室 Primary_Secondary_Address 1丁目12番地3号 創生ビル 1階 101号室 Last line fields Country ISO_Country_Code_3Digit 392 ISO_Country_Code_2Char JP Postcode1 001 Postcode2 483 日本 8521 2011-06-09
- 484. Data Quality Last line fields Postcode_Full 001-8521 Region1 北海 Region1_Description 道 Locality1_Name 札幌 Locality1_Description 市 Locality2_Name 北 Locality2_Description 区 Locality3_Name 北十条西 Lastline 001-8521 北海道 札幌市 北区 北十条西 Firm Firm 札幌私書箱センター Non-parsed fields Status_Code Assignment_Type F Address_Type 484 S0000 S 2011-06-09
- 485. Data Quality 16.5.8 Process Chinese addresses The Global Address Cleanse transform's Global Address engine parses Chinese addresses. The primary purpose of this transform and engine is to parse and normalize addresses for data matching and cleansing applications. 16.5.8.1 Chinese address format Chinese Addresses are written starting with the postal code, followed by the largest administrative region (for example, province), and continue down to the smallest unit (for example, room number and mail receiver). When people send mail between different prefectures, they often include the largest administrative region in the address. The addresses contain detailed information about where the mail will be delivered. Buildings along the street are numbered sequentially, sometimes with odd numbers on one side and even numbers on the other side. In some instances both odd and even numbers are on the same side of the street. Postal Code In China, the Postal Code is 6-digit number to identify the target deliver point of the address, and often has the prefix 邮编 Country 中华人民共和国 (People's Republic of China)" is the full name of China, we often use the words " 中国 (PRC)" as an abbreviation of the country name. For mails delivered within China, the domestic addresses often omit the Country name of the target address Province In China, "Provinces" are similar to what a "state" is in the United States. China has 34 province-level divisions, including: • • • • Provinces(省 shěng) Autonomous regions(自治区 zìzhìqū) Municipalities(直辖市 zhíxiáshì) Special administrative regions(特别行政区 tèbié xíngzhèngqū) Prefecture Prefecture-level divisions are the second level of the administrative structure, including: • • • • 485 Prefectures (地区 dìqū) Autonomous prefectures (自治州 zìzhìzhōu) Prefecture-level cities (地级市dìjíshì) Leagues (盟méng) 2011-06-09
- 486. Data Quality County The county is the sub-division of Prefecture, including: • Counties (县 xiàn) • Autonomous counties (自治县 zìzhìxiàn) • County-level cities(县级市 xiànjíshì) • Districts (市辖区 shìxiáqū) • Banners (旗 qí) • Autonomous banners (自治旗 zìzhìqí) • Forestry areas (林区 línqū) • Special districts (特区 tèqū) Township Township level division includes: • Townships (乡 xiāng) • Ethnic townships (民族乡 mínzúxiāng) • Towns(镇 zhèn) • Subdistricts (街道办事处 jiēdàobànshìchù) • District public offices (区公所 qūgōngsuǒ) • Sumu(苏木 sūmù) • Ethnic sumu (民族苏木 mínzúsūmù) Village Village includes: • • • • Neighborhood committees(社区居民委员会 jūmínwěiyuánhùi) Neighborhoods or communities (社区) Village committees(村民委员会 cūnmínwěiyuánhùi) or Village groups (村民小组 cūnmínxiǎozǔ) Administrative villages(行政村 xíngzhèngcūn) Street information Specifies the delivery point where the mail receiver can be found within it. In China, The street information often has the form of Street (Road) name -> House number. For example, 上海市浦东新区晨晖路1001 号 • • Street name: The street name is usually followed by one of these suffixes 路, 大道, 街, 大街 and so on. House number: The house number is followed by the suffix 号, the house number is a unique number within the Street/Road. Residential community In China, residential community might be used for mail delivery. Especially for some famous residential communities in major cities, the street name and house number might be omitted. The residential community doesn't have a naming standard and it is not strictly required to be followed by a typical 486 2011-06-09
- 487. Data Quality marker. However, it is often followed by the typical suffixes, such as 新村, 小区 and so on (For example, 新村, 小区). Building name Building is often followed by the building marker, such as 大厦, 大楼 and so on, though is not strictly required (For example, 中华大厦). Building name in the residential communities is often represented by a number with a suffix of 号，幢 and so on (For example: 上海市浦东新区晨晖路100弄10号101室). Common metro address This address includes the District name, which is common for metropolitan areas in major cities. Address component Chinese English Output field Postcode 510030 510030 Postcode_Full Country 中国 China Country Province 广东省 Guangdong Province Region1_Full City name 广州市 Guangzhou City Locality1_Full District name 越秀区 Yuexiu District Locality2_Full Street name 西湖路 Xihu Road Primary_Name_Full1 House number 99 号 No. 99 Primary_Number_Full Rural address This address includes the Village name, which is common for rural addresses. Address component English Output field Postcode 5111316 5111316 Postcode_Full Country 中国 China Country Province 广东省 Guangdong Province Region1_Full City name 487 Chinese 广州市 Guangzhou City Locality1_Full 2011-06-09
- 488. Data Quality Address component Chinese English Output field County-level City name 增城市 Zengcheng City Locality2_Full Town name 荔城镇 Licheng Town Locality3_Full Village name 联益村 Lianyi Village Locality4_Full Street name 光大路 Guangda Road Primary_Name_Full1 House number 99 号 No. 99 Primary_Number_Full 16.5.8.2 Sample Chinese address This address has been processed by the Global Address Cleanse transform and the Global Address engine. Input 510830 广东省广州市花都区赤坭镇广源路 1 号星辰大厦 8 层 809 室 Address-Line fields Primary_Name1 Primary_Type1 路 Primary_Number 1 Primary_Number_Description 号 Building_Name1 星辰大厦 Floor_Number 488 广源 8 2011-06-09
- 489. Data Quality Address-Line fields Floor_Description 层 Unit_Number 809 Unit_Description 室 Primary_Address 广源路 1号 Secondary_Address 星辰大厦 8层809室 Primary_Secondary_Address 广源路 1号星辰大厦8层809室 Lastline fields Country Postcode_Full 510168 Region1 广东 Region1_Description 省 Locality1_Name 广州 Locality1_Description 市 Locality2_Name 花都 Locality2_Description 区 Locality3_Name 赤坭 Locality3_Description 489 中国 镇 2011-06-09
- 490. Data Quality Lastline fields Lastline 510830广东省广州市花都区赤坭镇 Non-parsed fields Status_Code S0000 Assignment_Type S Address_Type S 16.5.9 Supported countries (Global Address Cleanse) There are several countries supported by the Global Address Cleanse transform. The level of correction varies by country and by the engine that you use. Complete coverage of all addresses in a country is not guaranteed. For the Global Address engine, country support depends on which sets of postal directories you have purchased. For Japan, the assignment level is based on data provided by the Ministry of Public Management Home Affairs, Posts and Telecommunications (MPT). During Country ID processing, the transform can identify many countries. However, the Global Address Cleanse transform's engines may not provide address correction for all of those countries. Related Topics • Process U.S territories with the USA engine • Reference Guide: Country ISO codes and assignment engines 16.5.9.1 Process U.S territories with the USA engine 490 2011-06-09
- 491. Data Quality When you use the USA engine to process addresses from American Samoa, Guam, Northern Mariana Islands, Palau, Puerto Rico, and the U.S. Virgin Islands, the output region is AS, GU, MP, PW, PR, or VI, respectively. The output country, however, is the United States (US). If you do not want the output country as the United States when processing addresses with the USA engine, set the "Use Postal Country Name" option to No. These steps show you how to set the Use Postal Country Name in the Global Address Cleanse transform. 1. Open the Global Address Cleanse transform. 2. On the Options tab, expandStandardization Options > Country > Options. 3. For the Use Postal Country Name option, select No. Related Topics • Supported countries (Global Address Cleanse) 16.5.9.2 Set a default country Note: Run Country ID processing only if you are: • Using two or more of the engines and your input addresses contain country data (such as the two-character ISO code or a country name). • Using an engine that processes multiple countries (such as the EMEA or Global Address engine). • Using only one engine, but your input data contains addresses from multiple countries. 1. Open the Global Address Cleanse transform. 2. On the Options tab, expand Country ID Options, and then for the Country ID Mode option select Assign. This value directs the transform to use Country ID to assign the country. If Country ID cannot assign the country, it will use the value in Country Name. 3. For the Country Name option, select the country that you want to use as a default country. The transform will use this country only when Country ID cannot assign a country. If you do not want a default country, select None. 4. For the Script Code option, select the type of script code that represents your data. The LATN option provides script code for most types of data. However, if you are processing Japanese data, select KANA Related Topics • Identify the country of destination • To set a constant country 491 2011-06-09
- 492. Data Quality 16.5.9.3 To set a constant country 1. Open the Global Address Cleanse transform. 2. On the Options tab, expand Country ID Options, and then for the Country ID Mode option select Constant. This value tells the transform to take the country information from the Country Name and Script Code options (instead of running “Country ID” processing). 3. For the Country Name option, select the country that represents all your input data. 4. For the Script Code option, select the type of script code that represents your data. The LATN option provides script code for most types of data. However, if you are processing Japanese data, select KANA Related Topics • Identify the country of destination • Set a default country 16.5.10 New Zealand Certification New Zealand Certification enables you to process New Zealand addresses and qualify for mailing discounts with the New Zealand Post. 16.5.10.1 To enable New Zealand Certification You need to purchase the New Zealand directory data and obtain a customer number from the New Zealand Post before you can use the New Zealand Certification option. To process New Zealand addresses that qualify for mailing discounts: 1. In the Global Address Transform, enable Report and Analysis > Generate Report Data. 2. In the Global Address Cleanse Transform, set Country Options > Disable Certification to No. Note: The software does not produce the New Zealand Statement of Accuracy (SOA) report when this option is set to Yes. 492 2011-06-09
- 493. Data Quality 3. In the Global Address Transform, complete all applicable options in the Global Address > Report Options > New Zealand subgroup. 4. In the Global Address Cleanse Transform, set Engines > Global Address to Yes. After you run the job and produce the New Zealand Statement of Accuracy (SOA) report, you need to rename the New Zealand Statement of Accuracy (SOA) report and New Zealand Statement of Accuracy (SOA) Production Log before submitting your mailing. For more information on the required naming format, See New Zealand SOA Report and SOA production log file. Related Topics • Management Console Guide: New Zealand Statement of Accuracy (SOA) report • Reference Guide: Report options for New Zealand 16.5.10.2 New Zealand SOA Report and SOA production log file New Zealand Statement of Accuracy (SOA) Report The New Zealand Statement of Accuracy (SOA) report includes statistical information about address cleansing for New Zealand. New Zealand Statement of Accuracy (SOA) Production Log The New Zealand Statement of Accuracy (SOA) production log contains identical information as the SOA report in a pipe-delimited ASCII text file (with a header record). The software creates the SOA production log by extracting data from the Sendrightaddraccuracy table within the repository. The software appends a new record to the Sendrightaddraccuracy table each time a file is processed with the DISABLE_CERTIFICATION option set to No. If the DISABLE_CERTIFICATION option is set to Yes, the software does not produce the SOA report and an entry will not be appended to the Sendrightaddraccuracy table. Mailers must retain the production log file for at least 2 years. The default location of the SOA production log is <DataServicesInstallLocation>Business ObjectsBusinessObjects Data ServicesDataQualitycertificationsCertifica tionLogs. Mailing requirements The SOA report and production log are only required when you submit the data processed for a mailing and want to receive postage discounts. Submit the SOA production log at least once a month. Submit an SOA report for each file that is processed for mailing discounts. File naming format The SOA production log and SOA report must have a file name in the following format: Production Log [SOA% (9999)]_[SOA Expiry Date (YYYYMMDD)]_[SOA ID].txt 493 2011-06-09
- 494. Data Quality SOA Report [SOA% (9999)]_[SOA Expiry Date (YYYYMMDD)]_[SOA ID].PDF Example: An SOA with: SOA % = 94.3% SOA expiry date = 15 Oct 2008 SOA ID = AGD07_12345678 The file names will be: Production Log - 0943_20081015_AGD07_12345678.txt SOA Report - 0943_20081015_AGD07_12345678.pdf Related Topics • Management Console Guide: New Zealand Statement of Accuracy (SOA) report • Management Console Guide: Exporting New Zealand SOA certification logs 16.5.10.3 The New Zealand Certification blueprint Do the following to edit the blueprint, run the job for New Zealand Certification, and generate the SOA production log file: 1. Import nz_sendright_certification.atl located in the DataQualitycertifications folder in the location where you installed the software. The default location is <DataServicesInstallLo cation>Business ObjectsBusinessObjects Data ServicesDataQualitycertifications. The import adds the following objects to the repository: • The project DataQualityCertifications • The job Job_DqBatchNewZealand_SOAProductionLog • The dataflow DF_DqBatchNewZealand_SOAProductionLog • The datastore DataQualityCertifications • The file format DqNewZealandSOAProductionLog 2. Edit the datastore DataQualityCertifications. Follow the steps listed in Editing the datastore . 3. Optional: By default, the software places the SOA Production Log in <DataServicesInstallLo cation>Business ObjectsBusinessObjects Data ServicesDataQualitycerti ficationsCertificationLogs. If the default location is acceptable, ignore this step. If you want to output the production log file to a different location, edit the substitution parameter configu 494 2011-06-09
- 495. Data Quality ration. From the Designer access Tools > Substitution Parameter Configurations and change the path location in Configuration1 for the substitution parameter $$CertificationLogPath to the location of your choice. 4. Run the job Job_DqBatchNewZealand_SOAProductionLog. The job produces an SOA Production Log called SOAPerc_SOAExpDate_SOAId.txt in the default location or the location you specified in the substitution parameter configuration. 5. Rename the SOAPerc_SOAExpDate_SOAId.txt file using data in the last record in the log file and the file naming format described in New Zealand SOA Report and SOA production log file. Related Topics • New Zealand SOA Report and SOA production log file • Management Console Guide: New Zealand Statement of Accuracy (SOA) report 16.5.10.4 Editing the datastore After you download the blueprint .zip file to the appropriate folder, unzip it, and import the .atl file in the software, you must edit the DataQualityCertifications datastore. To edit the datastore: 1. Select the Datastores tab of the Local Object Library, right-click DataQualityCertifications and select Edit. 2. Click Advanced to expand the Edit Datastore DataQualityCertifications window. Note: Skip step 3 if you have Microsoft SQL Server 2000 or 2005 as a datastore database type. 3. Click Edit. 4. Find the column for your database type, change Default configuration to Yes, and click OK. Note: If you are using a version of Oracle other than Oracle 9i, perform the following substeps: a. In the toolbar, click Create New Configuration. b. Enter your information, including the Oracle database version that you are using, and then click OK. c. Click Close on the Added New Values - Modified Objects window. d. In the new column that appears to the right of the previous columns, select Yes for the Default configuration. e. Enter your information for the Database connection name, User name, and Password options. f. In DBO, enter your schema name. g. In Code Page, select cp1252 and then click OK. 495 2011-06-09
- 496. Data Quality 5. At the Edit Datastore DataQualityCertifications window, enter your repository connection information in place of the CHANGE_THIS values. (You may have to change three or four options, depending on your repository type.) 6. Expand the Aliases group and enter your owner name in place of the CHANGE_THIS value. If you are using Microsoft SQL Server, set this value to DBO. 7. Click OK. If the window closes without any error message, then the database is successfully connected. 16.5.11 Global Address Cleanse Suggestion List The Global Address Cleanse transform's Suggestion List processing option is used in transactional projects to complete and populate addresses that have minimal data. Suggestion lists can offer suggestions for possible matches if an exact match is not found. This option is beneficial in situations where a user wants to extract addresses not completely assigned by an automated process, and run through the system to find a list of possible matches. Based on the given input address, the Global Address Cleanse will perform an error-tolerant search in the address directory and return a list of possible matches. From the suggestion list returned, the user can select the correct suggestion and update the database accordingly. Note: • • No certification with Suggestion Lists: If you use the Canada engine or Global Address engine for Australia and New Zealand, you cannot certify your mailing for SERP, AMAS, or New Zealand certification. This option does not support processing of Japanese or Chinese address data. Start with a sample transform If you want to use the suggestion lists feature, it is best to start with the sample transforms that is configured for it. The sample transform, GlobalSuggestions_AddressCleanse is configured to cleanse Latin-1 address data in any supported country using the Suggestion List feature. Related Topics • Extracting data quality XML strings using extract_from_xml function 16.5.12 Global Suggestion List The Global Suggestion List transform allows the user to query addresses with minimal data (allows the use of wildcards), and it can offer a list of suggestions for possible matches. 496 2011-06-09
- 497. Data Quality It is a beneficial tool for a call center environment, where operators need to enter minimum input (i.e. number of keystrokes) to find the caller's delivery address. For example, if the operator is on the line with a caller from the United Kingdom, the application will prompt for the postcode and address range. Global Suggestion List is used to look-up the address with quick-entry The Global Suggestion List transform requires the two character ISO country code on input. Therefore, you may want to place a transform, such as the Country ID transform, that will output the ISO_Country_Code_2Char field before the Global Suggestion Lists transform. The Global Suggestion List transform is available for use with the Canada, Global Address, and USA engines. Note: No certification with suggestion lists: If you use the Canada engine, USA engine, or Global Address engine for Australia and New Zealand, you cannot certify your mailing for SERP, CASS, AMAS, or New Zealand certification. Start with a sample transform If you want to use the Global Suggestion List transform, it is best to start with one of the sample transforms that is configured for it. The following sample tranforms are available. Sample transform Description GlobalSuggestions A sample transform configured to generate a suggestion list for Latin-1 address data in any supported country. UKSuggestions A sample transform configured to generate a suggestion list for partial address data in the United Kingdom. 16.6 Beyond the basic address cleansing The USA Regulatory Address Cleanse transform offers many additional address cleanse features for U.S. addresses. These features extend address cleansing beyond the basic parsing and standardizing. 16.6.1 USPS DPV® Delivery Point Validation® is a USPS product developed to assist users in validating the accuracy of their address information. DPV compares Postcode2 information against the DPV directories to identify known addresses and potential problems with the address that may cause an address to become undeliverable. DPV is available for U.S. data in the USA Regulatory Address Cleanse transform only. 497 2011-06-09
- 498. Data Quality Note: DPV processing is required for CASS certification. If you are not processing for CASS certification, you can choose to run your jobs in non-certified mode and still enable DPV. Caution: If you choose to disable DPV processing, the software will not generate the CASS-required documentation and your mailing won't be eligible for postal discounts. Related Topics • To enable DPV • Non certified mode 16.6.1.1 Benefits of DPV DPV can be beneficial in the following areas: • • • • Mailing: DPV helps to screen out undeliverable-as-addressed (UAA) mail and helps to reduce mailing costs. Information quality: DPV increases the level of data accuracy through the ability to verify an address down to the individual house, suite, or apartment instead of only block face. Increased assignment rate: DPV may increase assignment rate through the use of DPV tiebreaking to resolve a tie when other tie-breaking methods are not conclusive. Preventing mail-order-fraud: DPV can eliminate shipping of merchandise to individuals who place fraudulent orders by verifying valid delivery addresses and Commercial Mail Receiving Agencies (CMRA). 16.6.1.2 DPV security The USPS has instituted processes that monitor the use of DPV. Each company that purchases the DPV functionality is required to sign a legal agreement stating that it will not attempt to misuse the DPV product. If a user abuses the DPV product, the USPS has the right to prohibit the user from using DPV in the future. 16.6.1.2.1 DPV false positive addresses The USPS has included false positive addresses in the DPV directories as an added security to prevent DPV abuse. Depending on what type of user you are and your license key codes, the software's behavior varies when it encounters a false positive address. The following table explains the behaviors for each user type: 498 2011-06-09
- 499. Data Quality User type Software behavior Read about: End users DPV processing is terminated. Obtaining DPV unlock code from SAP BusinessObjects Support End users with a stop processing alter- DPV processing con- Sending false positive logs to the USPS native agreement tinues. Service providers DPV processing con- Sending false positive logs to the USPS tinues. Related Topics • Stop Processing Alternative • Obtaining DPV unlock code from SAP BusinessObjects • Sending DPV false positive logs to the USPS 16.6.1.2.2 Stop Processing Alternative End users may establish a Stop Processing Alternative agreement with the USPS and SAP BusinessObjects. Establishing a stop processing agreement allows you to bypass any future directory locks. The Stop Processing Alternative is not an option in the software, it is a key code that you obtain from SAP BusinessObjects Support. First you must obtain the proper permissions from the USPS and then provide proof of permission to SAP BusinessObjects Support. Support will then provide a key code that disables the directory locking function in the software. Remember: When you obtain the Stop Processing Alternative key code from SAP BusinessObjects Support, enter it into the SAP BusinessObjects License Manager. With the Stop Processing Alternative key code in place, the software takes the following actions when a false positive is encountered: • Marks the record as a false positive. • Generates a log file containing the false positive address. • Notes the path to the log files in the error log. • Generates a US Regulatory Locking Report containing the path to the log file. • Continues processing your job. Even though your job continues processing, you are required to send the false positive log file to the USPS to notify them that a false positive address was detected. The USPS must release the list before you can use it for processing. Related Topics • Sending DPV false positive logs to the USPS 499 2011-06-09
- 500. Data Quality 16.6.1.2.3 DPV false positive logs The software generates a false positive log file any time it encounters a false positive record, regardless of how your job is set up. The software creates a separate log file for each mailing list that contains a false positive. If multiple false positives exist within one mailing list, the software writes them all to the same log file. DPV log file name and location The software stores DPV log files in the directory specified in the USPS Log Path option in the Reference Files group. Note: The USPS log path that you enter must be writable. An error is issued if you have entered a path that is not writable. Log file naming convention The software automatically names DPV false positive logs with the following format: dpvl####.log The #### portion of the naming format is a number between 0001 and 9999. For example, the first log file generated is dpvl0001.log, the next one is dpvl0002.log, and so on. Note: When you have set the data flow degree of parallelism to greater than 1, or you have enabled the run as a separate process option, the software generates one log per thread or process. During a job run, if the software encounters only one false positive record, one log will be generated. However, if it encounters more than one false positive record and the records are processed on different threads or processes, then the software will generate one log for each thread that processes a false positive record. Related Topics • Performance Optimization Guide: Using parallel execution 16.6.1.2.4 DPV locking for end users This locking behavior is applicable for end users or users who are DSF2 licensees that have DSF2 disabled in the job When the software finds a false positive address, DPV processing is discontinued for the remainder of the data flow. The software also takes the following actions: • • • • • • 500 Marks the record as a false positive address. Issues a message in the error log stating that a DPV false positive address was encountered. Includes the false positive address and lock code in the error log. Continues processing your data flow without DPV processing. Generates a lock code. Generates a false positive log. 2011-06-09
- 501. Data Quality • Generates a US Regulatory Locking Report that contains the false positive address and the lock code. (Report generation must be enabled in the USA Regulatory Address Cleanse transform.) To restore DPV functionality, users must obtain a DPV unlock code from SAP BusinessObjects Support. Related Topics • Obtaining DPV unlock code from SAP BusinessObjects 16.6.1.2.5 Obtaining DPV unlock code from SAP BusinessObjects These steps are applicable for end users who do not have a Stop Processing Alternative agreement with the USPS. When you receive a processing message that DPV false positive addresses are present in your address list, use the SAP BusinessObjectsUSPS Unlock Utility to obtain an unlock code. 1. Navigate to http://service.sap.com/bosp-unlock to open the SAP Service Market Place (SMP) unlock utility page. 2. Click Retrieve USPS Unlock Code. 3. Click Search and select an applicable Data Services system from the list. 4. Enter the lock code found in the dpvx.txt file (location is specified in the DPV Path option in the Reference Files group). 5. Select DPV as the lock type. 6. Select BOJ-EIM-DS as the component. 7. Enter the locking address that is listed in the dpvx.txt file. 8. Attach the dpvl####.log file (location is specified in the USPS Log Path option in the Reference Files group). 9. Click Submit. The unlock code displays. 10. Copy the unlock code and paste it into the dpvw.txt file, replacing all contents of the file with the unlock code (location is specified in the DPV path option of the Reference Files group). 11. Remove the record that caused the lock from the database, and delete the dpvl####.log file before processing the list again. Tip: Keep in mind that you can only use the unlock code one time. If the software detects another false-positive (even if it is the same record), you will need to retrieve a new LACSLink unlock code. Note: If an unlock code could not be generated, a message is still created and is processed by a Technical Customer Assurance engineer (during regular business hours). Note: If you are an end user who has a Stop Processing Alternative agreement, follow the steps to send the false positive log to the USPS. 501 2011-06-09
- 502. Data Quality 16.6.1.2.6 Sending DPV false positive logs to the USPS Service providers should follow these steps after receiving a processing message that DPV false positive addresses are present in their address list. End users with a Stop Processing Alternative agreement should follow these steps after receiving a processing message that DPV false positive addresses are present in their address list. 1. Send an email to the USPS NCSC at "[email protected]", and include the following information: • Type “DPV False Positive” as the subject line • Attach the dpvl####.log file or files that were generated by the software (location is specified in the USPS Log Path directory option in the Reference Files group) The USPS NCSC uses the information to determine whether the list can be returned to the mailer. 2. After the USPS NCSC has released the list that contained the locked or false positive record: • Delete the corresponding log file or files • Remove the record that caused the lock from the list and reprocess the file Note: If you are an end user who does not have a Stop Processing Alternative agreement, follow the steps to retrieve the DPV unlock code from SAP BusinessObjects Support. Related Topics • Obtaining DPV unlock code from SAP BusinessObjects 16.6.1.3 DPV monthly directories DPV directories are shipped monthly with the USPS directories in accordance with USPS guidelines. The directories expire in 105 days. The date on the DPV directories must be the same date as the Address directory. Do not rename any of the files. DPV will not run if the file names are changed. Here is a list of the DPV directories: • • • • • • 502 dpva.dir dpvb.dir dpvc.dir dpvd.dir dpv_vacant.dir dpv_no_stats.dir 2011-06-09
- 503. Data Quality 16.6.1.4 Required information in the job setup When you set up for DPV processing, the following options in the USPS License Information group are required: • • • • • • Customer Company Name Customer Company Address Customer Company Locality Customer Company Region Customer Company Postcode1 Customer Company Postcode2 16.6.1.5 To enable DPV Note: DPV is required for CASS. In addition to the required customer company information that you enter into the USPS License Information group, set the following options to perform DPV processing: 1. Open the USA Regulatory Address Cleanse transform. 2. Open the "Options" tab. Expand the Assignment Options group, and select Yes for the Enable DPV option. 3. In the Reference Files group, enter the path for your DPV directories in the DPV Path option. Note: DPV can run only when the location for all the DPV directories have been entered and none of the DPV directory files have been renamed. 4. Set a directory for the DPV log file in the USPS Path option. Use the substitution variable $$Certifi cationLogPath if you have it set up. 5. In the Report and Analysis group, select Yes for the Generate Report Data option. 16.6.1.6 DPV output fields Several output fields are available for reporting DPV processing results: • 503 DPV_CMRA 2011-06-09
- 504. Data Quality • • • • DPV_Footnote DPV_NoStats DPV_Status DPV_Vacant For full descriptions of these output fields, refer to the Reference Guide or view the Data Services Help information that appears when you open the Output tab of the USA Regulatory Address Cleanse transform. Related Topics • Reference Guide: Data Quality fields, USA Regulatory Address Cleanse fields, Output fields 16.6.1.7 Non certified mode You can set up your jobs with DPV disabled if you are not a CASS customer but you want a Postcode2 added to your addresses. The non-CASS option, Assign Postcode2 to Non DPV, enables the software to assign a Postcode2 when an address does not DPV-confirm. Caution: If you choose to disable DPV processing, the software does not generate the CASS-required documentation and your mailing won't be eligible for postal discounts. 16.6.1.7.1 Enable Non-Certified mode To run your job in non certified mode, follow these setup steps: 1. In the Assignment Options group, set the Enable DPV option to No. 2. In the Non Certified options group, set the Disable Certification to Yes. 3. In the Non Certified options group, set the Assign Postcode2 Not DPV Validated to Yes. Caution: The software blanks out all Postcode2 information in your data if you disable DPV processing and you disable the Assign Postcode2 Not DPV Validated option. This includes Postcode2 information provided in your input file. 16.6.1.8 DPV performance Due to additional time required to perform DPV processing, you may see a change in processing time. Processing time may vary with the DPV feature based on operating system, system configuration, and other variables that may be unique to your operating environment. 504 2011-06-09
- 505. Data Quality You can decrease the time required for DPV processing by loading DPV directories into system memory before processing. 16.6.1.8.1 Memory usage You may need to install additional memory on your operating system for DPV processing. We recommend a minimum of 768 MB to process with DPV enabled. To determine the amount of memory required to run with DPV enabled, check the size of the DPV 1 directories (recently about 600 MB ) and add that to the amount of memory required to run the software. The size of the DPV directories will vary depending on the amount of new data in each directory release. Make sure that your computer has enough memory available before performing DPV processing. To find the amount of disk space required to cache the directories, see the Supported Platforms document in the SAP BusinessObjects Support portal. Find link information in the SAP Business Objects Information resources table (see link below). Related Topics • SAP BusinessObjects information resources 16.6.1.8.2 Cache DPV directories To better manage memory usage when you have enabled DPV processing, choose to cache the DPV directories. 16.6.1.8.3 To cach DPV directories To set up your job for DPV caching, follow these steps: 1. In the Transform Performance group, set the Cache DPV Directories option to Yes. 2. In the same group, set the Insufficient Cache Memory Action to one of the following: Option Description Error Software issues an error and terminates the transform. Continue Software attempts to continue initialization without caching. 16.6.1.8.4 Running multiple jobs with DPV When running multiple DPV jobs and loading directories into memory, you should add a 10-second pause between jobs to allow time for the memory to be released. For more information about setting this properly, see your operating system manual. If you don't add a 10-second pause between jobs, there may not be enough time for your system to release the memory used for caching the directories from the first job. The next job waiting to process 1 505 The directory size is subject to change each time new DPV directories are installed. 2011-06-09
- 506. Data Quality may error out or access the directories from disk if there is not enough memory to cache directories. This may result in performance degradation. 16.6.1.9 DPV information in US Addressing Report The US Addressing Report automatically generates when you have enabled reporting in your job. The following sections of the US Addressing Report contain DPV information: • • DPV Return Codes Delivery Point Validation (DPV) Summary For information about the US Addressing Report, or other Data Quality reports, see the Management Console Guide. Related Topics • Management Console: Data Quality reports, US Addressing Report 16.6.1.10 DPV No Stats indicators The USPS uses No Stats indicators to mark addresses that fall under the No Stats category. The software uses the No Stats table when you have DPV or DSF2 turned on in your job. The USPS puts No Stats addresses in three categories: • • • Addresses that do not have delivery established yet. Addresses that receive mail as part of a drop. Addresses that have been vacant for a certain period of time. 16.6.1.10.1 No Stats table You must install the No Stats table (dpv_no_stats.dir) before the software performs DPV or DSF2 processing. The No Stats table is supplied by SAP BusinessObjects with the DPV directory install. The software automatically checks for the No Stats table in the directory folder that you indicate in your job setup. The software performs DPV and DSF2 processing based on the install status of the directory. dpv_no_stats.dir Installed 506 Results The software automatically outputs No Stats indicators when you include the DPV_NoStats output field in your job. 2011-06-09
- 507. Data Quality dpv_no_stats.dir Results Not installed The software automatically skips the No Stats processing and does not issue an error message. The software will perform DPV processing but won't populate the DPV_NoStat output field. 16.6.1.10.2 No Stats output field Use the DPV_NoStats output field to post No Stat indicator information to an output file. No Stat means that the address is a vacant property, it receives mail as a part of a drop, or it does not have an established delivery yet. Related Topics • DPV output fields 16.6.1.11 DPV Vacant indicators The software provides vacant information in output fields and reports using DPV vacant counts. The USPS DPV vacant lookup table is supplied by SAP BusinessObjects with the DPV directory install. The USPS uses DPV vacant indicators to mark addresses that fall under the vacant category. The software uses DPV vacant indicators when you have DPV or DSF2 enabled in your job. Tip: The USPS defines vacant as any delivery point that was active in the past, but is currently not occupied (usually over 90 days) and is not currently receiving mail delivery. The address could receive delivery again in the future. "Vacant" does not apply to seasonal addresses. 16.6.1.11.1 DPV address-attribute output field Vacant indicators for the assigned address are available in the DPV_Vacant output field. Note: The US Addressing Report contains DPV Vacant counts in the DPV Summary section. Related Topics • DPV output fields • Management Console: Data Quality reports, US Addressing Report 507 2011-06-09
- 508. Data Quality 16.6.2 LACSLink® LACSLink is a USPS product that is available for U.S. records with the USA Regulatory Address Cleanse transform only. LACSLink processing is required for CASS certification. LACSLink updates addresses when the physical address does not move but the address has changed. For example, when the municipality changes rural route addresses to street-name addresses. Rural route conversions make it easier for police, fire, ambulance, and postal personnel to locate a rural address. LACSLink also converts addresses when streets are renamed or post office boxes renumbered. LACSLink technology ensures that the data remains private and secure, and at the same time gives you easy access to the data. LACSLink is an integrated part of address processing; it is not an extra step. To obtain the new addresses, you must already have the old address data. Related Topics • How LACSLink works • To control memory usage for LACSLink processing • To disable LACSLink • LACSLink security 16.6.2.1 Benefits of LACSLink LACSLink processing is required for all CASS customers. If you process your data without LACSLink enabled, you won't get the CASS-required reports or postal discounts. 16.6.2.2 LACSLink security The USPS has instituted processes that monitor the use of LACSLink. Each company that purchases the LACSLink functionality is required to sign a legal agreement stating that it will not attempt to misuse the LACSLink product. If a user abuses the LACSLink product, the USPS has the right to prohibit the user from using LACSLink in the future. 508 2011-06-09
- 509. Data Quality 16.6.2.2.1 LACSLink false positive addresses The USPS has included false positive addresses in the LACSLink directories as an added security to prevent LACSLink abuse. Depending on what type of user you are and your license key codes, the software's behavior varies when it encounters a false positive address. The following table explains the behaviors for each user type: User type Software behavior Read about: End users LACSLink processing is terminated. Obtaining the LACSLink unlock code from SAP BusinessObjects Support End users with a Stop Processing LACSLink processing Alternative agreement continues. Sending false positive logs to the USPS Service providers Sending false positive logs to the USPS LACSLink processing continues. Related Topics • Stop Processing Alternative • Obtaining LACSLink unlock code from SAP BusinessObjects • Sending LACSLink false positive logs to the USPS 16.6.2.2.2 LACSLink false positive logs The software generates a false-positive log file any time it encounters a false positive record, regardless of how your job is set up. The software creates a separate log file for each mailing list that contains a false positive. If multiple false positives exist within one mailing list, the software writes them all to the same log file. LACSLink log file location The software stores LACSLink log files in the directory specified for the USPS Log Path in the Reference Files group. Note: The USPS log path that you enter must be writable. An error is issued if you have entered a path that is not writable. The software names LACSLink false positive logs lacsl###.log, where ### is a number between 001 and 999. For example, the first log file generated is lacsl001.log, the next one is lacsl002.log, and so on. Note: When you have set the data flow degree of parallelism to greater than 1, the software generates one log per thread. During a job run, if the software encounters only one false positive record, one log will be generated. However, if it encounters more than one false positive record and the records are 509 2011-06-09
- 510. Data Quality processed on different threads, then the software will generate one log for each thread that processes a false positive record. Related Topics • Performance Optimization Guide: Using parallel execution 16.6.2.2.3 LACSLink locking for end users This locking behavior is applicable for end users or users who are DSF2 licensees that have DSF2 disabled in the job. When the software finds a false positive address, LACSLink processing is discontinued for the remainder of the job processing. The software takes the following actions: • • • • • • • Marks the record as a false positive address. Issues a message in the error log that a LACSLink false positive address was encountered. Includes the false positive address and lock code in the error log. Continues processing your data flow without LACSLink processing. Generates a lock code. Generates a false positive error log. Generates a US Regulatory Locking Report that contains the false positive address and the lock code (Report generation must be enabled in the USA Regulatory Address Cleanse transform. To restore LACSLink functionality, users must obtain a LACSLink unlock code from SAP BusinessObjects Support. 16.6.2.2.4 Obtaining LACSLink unlock code from SAP BusinessObjects These steps are applicable for end users who do not have a Stop Processing Alternative agreement with the USPS. When you receive a processing message that LACSLink false positive addresses are present in your address list, use the SAP BusinessObjectsUSPS Unlock Utility to obtain an unlock code. 1. Navigate to http://service.sap.com/bosp-unlock to open the SAP Service Market Place (SMP) unlock utility page. 2. Click Retrieve USPS Unlock Code. 3. Click Search and select an applicable Data Services system from the list. 4. Enter the lock code found in the lacsx.txt file (location is specified in the LACSLink Path option in the Reference Files group). 5. Select LACSLink as the lock type. 6. Select BOJ-EIM-DS as the component. 7. Enter the locking address that is listed in the lacsx.txt file. 8. Attach the lacsl####.log file (location specified in the USPS Log Path option in the Reference Files group). 9. Click Submit. The unlock code displays. 510 2011-06-09
- 511. Data Quality 10. Copy the unlock code and paste it into the lacsw.txt file, replacing all contents of the file with the unlock code (location is specified in the LACSLink path option in the Reference Files group). 11. Remove the record that caused the lock from the database, and delete the lacsl####.log file before processing the list again. Tip: Keep in mind that you can only use the unlock code one time. If the software detects another false-positive (even if it is the same record), you will need to retrieve a new LACSLink unlock code. Note: If an unlock code could not be generated, a message is still created and is processed by a Technical Customer Assurance engineer (during regular business hours). Note: If you are an end user who has a Stop Processing Alternative agreement, follow the steps to send the false positive log to the USPS. 16.6.2.2.5 Sending LACSLink false positive logs to the USPS Service providers should follow these steps after receiving a processing message that LACSLink false positive addresses are present in their address list. End users with a Stop Processing Alternative agreement should follow these steps after receiving a processing message that LACSLink false positive addresses are present in their address list. 1. Send an email to the USPS at "[email protected]". Include the following: • Type “LACSLink False Positive” as the subject line • Attach the lacsl###.log file or files that were generated by the software (location specified in the USPS Log Files option in the Reference Files group). The USPS NCSC uses the information to determine whether or not the list can be returned to the mailer. 2. After the USPS NCSC has released the list that contained the locked or false positive record: • Delete the corresponding log file or files • Remove the record that caused the lock from the list and reprocess the file Note: If you are an end user who does not have a Stop Processing Alternative agreement, follow the steps to retrieve the LACSLink unlock code from SAP BusinessObjects Support. Related Topics • Obtaining LACSLink unlock code from SAP BusinessObjects 16.6.2.3 How LACSLink works 511 2011-06-09
- 512. Data Quality LACSLink provides a new address when one is available. LACSLink follows these steps when processing an address: 1. The USA Regulatory Address Cleanse transform standardizes the input address. 2. The transform looks for a matching address in the LACSLink data. 3. If a match is found, the transform outputs the LACSLink-converted address and other LACSLink information. Related Topics • To control memory usage for LACSLink processing • LACSLink® 16.6.2.4 Conditions for address processing The transform does not process all of your addresses with LACSLink when it is enabled. Here are the conditions under which your data is passed into LACSLink processing: • The address is found in the address directory, and it is flagged as a LACS-convertible record within the address directory. • The address is found in the address directory, and, even though a rural route or highway contract default assignment was made, the record wasn't flagged as LACS convertible. • The address is not found in the address directory, but the record contains enough information to be sent into LACSLink. For example, the following table shows an address that was found in the address directory as a LACS-convertible address. Original address After LACSLink conversion RR2 BOX 204 463 SHOWERS RD DU BOIS PA 15801 DU BOIS PA 15801-66675 16.6.2.5 Sample transform configuration 512 2011-06-09
- 513. Data Quality LACSLink processing is enabled by default in the sample transform configuration because it is required for CASS certification. The sample transform configuration is named USARegulatory_AddressCleanse and is found under the USA_Regulatory_Address_Cleanse group in the Object Library. 16.6.2.6 LACSLink directory files SAP Business Objects ships the LACSLink directory files with the U.S. National Directory update. The LACSLink directory files require about 600 MB of additional hard drive space. The LACSLink directories include the following: • • • • lacsw.txt lacsx.txt lacsy.ll lacsz.ll Caution: The LACSLink directories must reside on the hard drive in the same directory as the LACSLink supporting files. Do not rename any of the files. LACSLink will not run if the file names are changed. 16.6.2.6.1 Directory expiration and updates LACSLink directories expire in 105 days. LACSLink directories must have the same date as the other directories that you are using from the U.S. National Directories. 16.6.2.7 To enable LACSLink LACSLink is enabled by default in the USA Regulatory Address Cleanse transform. If you need to re-enable the option, follow these steps: 1. 2. 3. 4. Open the USA Regulatory Address Cleanse transform and open the "Options" tab. Expand the Processing Options group select Yes in the Enable LACSLink option. Enter the LACSLink path for the LACSLink Path option In the Reference Files group. You can use the substitution variable $$RefFilesAddressCleanse if you have it set up. 5. Complete the required fields in the USPS License Information group. 16.6.2.7.1 Required information in the job setup All users running LACSLink must include required information in the USPS License Information group. The required options include the following: • 513 Customer Company Name 2011-06-09
- 514. Data Quality • • • • • • Customer Company Address Customer Company Locality Customer Company Region Customer Company Postcode1 Customer Company Postcode2 Customer Company Phone 16.6.2.7.2 To disable LACSLink LACSLink is enabled by default in the USA Regulatory Address Cleanse transform configuration because it is required for CASS processing. Therefore, you must disable CASS certification in order to disable LACSLink. 1. In the USA Regulatory Address Cleanse transform configuration, open the "Options" tab. 2. Open the Non Certified Options group. 3. Select Yes for the Disable Certification option. 4. Open the Assignment Option group. 5. Select No for the Enable LACSLink option. Related Topics • LACSLink® 16.6.2.7.3 Reasons for errors If your job setup is missing information in the USPS License Information group, and you have DPV and/or LACSLink enabled in your job, you will get error messages based on these specific situations: Reason for error Missing required options Description When your job setup does not include the required parameters in the USPS License Information group, and you have DPV and/or LACSLink enabled in your job, the software issues a verification error. Unwritable Log File direcIf the path that you specified for the USPS Log Path option in the Reference tory Files group is not writable, the software issues an error. 16.6.2.8 LACSLink output fields Several output fields are available for reporting LACSLink processing results. 514 2011-06-09
- 515. Data Quality You must enable LACSLink, and include these output fields in your job setup, before the software posts information to these fields. Field name Length Description Returns the pre-conversion address, populated only when LACSLink is enabled and a LACSLink lookup was attempted. LACSLINK_QUERY 50 This address will be in the standard USPS format (as shown in USPS Publication 28). However, when an address has both a unit designator and secondary unit, the unit designator is replaced by the character “#”. blank: No LACSLink lookup attempted. 515 2011-06-09
- 516. Data Quality Field name Length Description Returns the match status for LACSLink processing: A = LACSLink record match. A converted address is provided in the address data fields. 00 = No match and no converted address. LACSLINK_RETURN_CODE 2 09 = LACSLink matched an input address to an old address, which is a "high-rise default" address; no new address is provided. 14 = Found a LACSLink record, but couldn't convert the data to a deliverable address. 92 = LACSLink record matched after dropping the secondary number from input address. blank = No LACSLink lookup attempted. 516 2011-06-09
- 517. Data Quality Field name Length Description Returns the conversion status of addresses processed by LACSLink. Y = Address converted by LACSLink (the LACSLink_Return_Code value is A). N = Address looked up with LACSLink but not converted. LACSLINK_INDICATOR 1 F = The address was a falsepositive. S = LACSLink conversion was made, but it was necessary to drop the secondary information. blank: No LACSLink lookup attempted. 16.6.2.9 To control memory usage for LACSLink processing The transform performance improves considerably if you cache the LACSLink directories. For the amount of disk space required to cache the directories, see the Supported Platforms document available in the SAP BusinessObjects Support > Documentation > Supported Platforms/PARs section of the SAP Service Marketplace: http://service.sap.com/bosap-support. If you do not have adequate system memory to load the LACSLink directories and the Insufficient Cache Memory Action is set to Error, a verification error message is displayed at run-time and the transform terminates. If the Continue option is chosen, the transform attempts to continue LACSLink processing without caching. Open the "Options" tab of your USA Regulatory Address Cleanse transform configuration in your data flow. Follow these steps to load the LACSLink directories into your system memory: 1. Open the Transform Performance option group. 2. Select Yes for the Cache LACSLink Directories option. Related Topics • LACSLink® 517 2011-06-09
- 518. Data Quality 16.6.2.10 LACSLink information in US Addressing Report The US Addressing Report automatically generates when you have enabled reporting in your job. The following table lists the LACSLink sections in the US Addressing Report: Section Information Locatable Address Conversion Record counts and percentages for the following information: (LACSLink) Summary • LACSLink converted addresses • Addresses not LACSLink converted LACSLink Return Codes Record counts and percentages for the following information: • Converted • Secondary dropped • No match • Can't convert • High-rise default 16.6.2.11 USPS Form 3553 The USPS Form 3553 reports LACSLink counts. The LACS/LACSLink field shows the number of records that have a LACSLink Indicator of Y or S, if LACSLink processing is enabled. If LACSLink processing is not enabled, this field shows the number of LACS code count. 16.6.3 SuiteLink™ SuiteLink is an option in the USA Regulatory Address Cleanse transform. SuiteLink uses a USPS directory that contains multiple files of specially indexed address information, like secondary numbers and unit designators, for locations identified as high-rise default buildings. With SuiteLink you can build accurate and complete addresses by adding suite numbers to high-rise business addresses. With the secondary address information added to your addresses, more of your pieces are sorted by delivery sequence and delivered with accuracy and speed. 518 2011-06-09
- 519. Data Quality SuiteLink is required for CASS SuiteLink is required when you process in CASS mode (and the Disable certification option is set to No). If you have disabled SuiteLink in your job setup, but you are in CASS mode, an error message is issued and processing does not continue. 16.6.3.1 Benefits of SuiteLink Businesses who depend on Web-site, mail, or in-store orders from customers will find that SuiteLink is a powerful money-saving tool. Also businesses who have customers that reside in buildings that house several businesses will appreciate getting their marketing materials, bank statements, and orders delivered right to their door. The addition of secondary number information to your addresses allows for the most efficient and cost-effective delivery sequencing and postage discounts. Note: SuiteLink is required for those preparing CASS-compliant mailing lists. 16.6.3.2 How SuiteLink works The software uses the data in the SuiteLink directories to add suite numbers to applicable addresses. The software matches a company name, a known high-rise address, and the CASS-certified postcode2 in your database to data in SuiteLink. When there is a match, the software creates a complete business address that includes the suite number. Example: Assign suite number This example shows a record that is processed through SuiteLink, and the output record with the assigned suite number. The input record contains: • Firm name (in FIRM input field) • Known high-rise address • CASS-certified postcode2 The SuiteLink directory contains: • • 519 secondary numbers unit designators 2011-06-09
- 520. Data Quality The output record contains: • the correct suite number Input record Output record Telera TELERA 910 E Hamilton Ave Fl2 910 E HAMILTON AVE STE 200 Campbell CA 95008 0610 CAMPBELL CA 95008 0625 16.6.3.3 SuiteLink directory The SuiteLink directory is distributed monthly. You must use SuiteLink directories with a zip4us.dir directory for the same month. (zip4us.dir path is entered in the Address Directory1 option of the Reference Files group in the USA Regulatory Address Cleanse transform. ) For example, the December 2011 SuiteLink directory can be used with only the December 2011 zip4us.dir directory. You cannot use a SuiteLink directory that is older than 60 days based on its release date. The software warns you 15 days before the directory expires. As with all directories, the software won't process your records with an expired SuiteLink directory. 16.6.3.4 To enable SuiteLink SuiteLink is enabled by default in any of the sample transform configurations that are set up to be CASS-compliant (and the Disable certification option is set to No). For example, if you use the USA Regulatory_AddressCleanse transform, SuiteLink is enabled. Note: Because SuiteLink is required for CASS processing, the Disable Certification option in the Non Certified Options group must be set to No. However, if you disable SuiteLink, you must also set the Disable Certification option to Yes. 1. Open the USA Regulatory Address Cleanse transform in your dataflow. 2. Open the "Options" tab. 520 2011-06-09
- 521. Data Quality 3. Expand the Assignment Options group and set the Enable SuiteLink option to Yes. 4. In the Reference Files group, enter the SuiteLink directory path in the SuiteLink Path option. You can use the substitution variable $$RefFilesAddressCleanse if you have it set up with the directory location that contains your SuiteLInk directories. 5. Optional: In the Transform Performance option group, set the Cache SuiteLink Directories option to Yes so that the SuiteLink directories are cached in memory. Note: Ensure that you have sufficient RAM to cache the SuiteLink directories before you enable this option. 16.6.3.5 Improve processing speed You may increase SuiteLink processing speed if you load the SuiteLink directories into memory. To activate this option, go to the Transform Performance group and set the Cache SuiteLink Directories to Yes. 16.6.3.6 SuiteLink return codes in US Addressing Report SuiteLink return code information is available in the US Addressing Report in the SuiteLink Return Codes section. The US Addressing Report shows the record count and percentage for the following return codes: A = Secondary exists and assignment made 00 = Lookup was attempted but no assignment 16.6.4 USPS DSF2® DSF2 is a USPS-licensed product that you can use to validate addresses, add delivery sequence information, and add DSF2 address attributes to addresses. There are two DSF2 features that are supported in Data Services: • • DSF2 Augment in the USA Regulatory Address Cleanse transform DSF2 Walk Sequence in the DSF2 Walk Sequencer transform Note: USPS DSF2 data is available only to USPS-certified DSF2 licensees. 521 2011-06-09
- 522. Data Quality Related Topics • DSF2 walk sequencing 16.6.4.1 Validate addresses DSF2 helps reduce the quantity of undeliverable-as-addressed (UAA) mail and keeps mailing costs down. DSF2 uses DPV® to validate addresses and identify inaccurate or incomplete addresses. Related Topics • USPS DPV® 16.6.4.2 Add address attributes DSF2 adds address attributes (information about the addresses) to your data. Use the attribute information to create more targeted mailings. 16.6.4.3 Add delivery sequence information DSF2 adds delivery sequence information to your data, which you can use to qualify for walk-sequence discounts. This information is sometimes called walk sequencing or pseudo sequencing. Related Topics • DSF2 walk sequencing • Pseudo sequencing 16.6.4.4 Benefits of DSF2 Those who want to target their mail to specific types of addresses and those who want to earn additional postal discounts will appreciate what DSF2 can do. 522 2011-06-09
- 523. Data Quality The DSF2 address-attribute data provides mailers with knowledge about the address above and beyond what is necessary to accurately format the addresses. Address-attribute data allows mailers to produce more targeted mailings. For example, If you plan to send out a coupon for your lawn-care service business, you do not want to send it to apartment dwellers (they may not have a lawn). You want your coupon to go to residential addresses that are not centralized in an apartment building. With the DSF2 information you can walk-sequence your mailings to achieve the best possible postal discounts by using the DSF2 Walk Sequencer transform. 16.6.4.5 Becoming a DSF2 licensee Before you can perform DSF2 processing in the software, you must complete the USPS DSF2 certification procedures and become licensed by the USPS. Part of certification is processing test jobs in Data Services to prove that the software complies with the license agreement. When you are ready to take these tests, contact SAP BusinessObjects Business User Support to obtain access to the DSF2 features in Data Services. Related Topics • DSF2 Certification 16.6.4.6 DSF2 directories DSF2 processing requires the following data: 523 2011-06-09
- 524. Data Quality Data Notes DPV directories The software uses DPV directories to verify addresses and identify inaccurate addresses. SAP BusinessObjects supplies the DPV directories with the U.S. National Directory delivery. Note: DPV directories are included with the DSF2 tables. Do not use the DPV directories included with the DSF2 tables. Use the DPV directories from SAP BusinessObjects with the U.S. National Directory delivery. eLOT directories The software uses eLOT directories to assign walk sequence numbers. SAP BusinessObjects supplies the eLOT directories with the U.S. National Directory delivery. Note: eLOT directories are included with the DSF2 tables. Do not use the eLOT directories included with the DSF2 tables. Use the eLOT directories from SAP BusinessObjects with the U.S. National Directory delivery. DSF2 tables The software uses DSF2 tables to assign address attributes. Note: DSF2 tables are supplied by the USPS and not SAP BusinessObjects. In addition, the DSF2 tables include DPV and eLOT directories. Do not use the DPV and eLOT directories included with the DSF2 tables. Use the DPV and eLOT directories from SAP BusinessObjects with the U.S. National Directory delivery. Delivery statistics file The software uses the delivery statistics file to provide counts of business and residential addresses per ZIP Code (Postcode1) per Carrier Route (Sortcode). SAP BusinessObjects supplies the delivery statistics file with the U.S. National Directory delivery. You must specify the location of these directory files in the USA Regulatory Address Cleanse transform, except for the delivery statistics file. Set the location of the delivery statistics file (dsf.dir) in the DSF2 Walk Sequencer transform. Also, to meet DSF2 requirements, you must install updated directories monthly. 16.6.4.7 DSF2 augment processing 524 2011-06-09
- 525. Data Quality Set up DSF2 augment processing in the USA Regulatory Address Cleanse transform. DSF2 processing requires DPV information, therefore, enable DPV in your job setup. If you plan to use the output information from the DSF2 augment processing for walk sequence processing, you must also enable eLOT. Note: DSF2 augment is available only in batch mode. You cannot add augment information to your data in real time. 16.6.4.7.1 DSF2 Augment directory expiration The DSF2 directories are distributed monthly. You must use the DSF2 directories with U.S. National directories that are labeled for the same month. For example, the May 2011 DSF2 directories can be used with only the May 2011 National directories. The DSF2 Augment data expires in 60 days instead of the 105 day expiration for the U.S. National directories. Because directories must all have the same base date (MM/YYYY), DSF2 users who have Augment or Both set for the DSF2 Mode option will have to update all of the U.S. National directories and other directories they use (such as LACSLink or DPV for example) at the same time as the DSF2 Augment directories. The software will remind users to update the directories with a warning message that appears 15 days before the directory expires. Remember: As with all directories, the software will not process your records with expired DSF2 directories. 16.6.4.7.2 Identify the DSF2 licensee When you perform DSF2 processing, you must provide the following information: The DSF2-licensed company and the client for whom the company is processing this job. You must complete the following options in the USPS License Information group for DSF2 processing: • • • • • • • • • • • • 525 DSF2 Licensee ID Licensee Name List Owner NAICS Code List ID Customer Company Name Customer Company Address Customer Company Locality Customer Company Region Customer Company Postcode1 Customer Company Postcode2 List Received Date List Return Date 2011-06-09
- 526. Data Quality Note: If you are performing DSF2 and NCOALink processing in the same instance of the USA Regulatory Address Cleanse transform, then the information that you enter in the USPS License Information group must apply to both DSF2 and NCOALink processing. If, for example, the List ID is different for DSF2 and NCOALink, you will need to include two USA Regulatory Address Cleanse transforms: One for NCOALink and another for DSF2. 16.6.4.7.3 To enable DSF2 Augment Before you can process with DSF2, you must first become a certified licensee. In addition to the required customer company information that you enter into the USPS License Information group, set the following options to perform DSF2 Augment processing: 1. In the USA Regulatory Address Cleanse transform, open the "Options" tab. 2. Expand the Report and Analysis group and set the Generate Report Data option to Yes. 3. Expand the Reference Files group and enter the path for the options DSF2 Augment Path, DPV Path, and eLOT Directory, or use the $$RefFilesAddressCleanse substitution variable if you have it set up. 4. Also in the Reference Files group, enter a path for the USPS Log Path option, or use the $$Certifi cationLogPath substitution variable if you have it set up. 5. Optional. Expand the Transform Performance group and set the Cache DPV Directories and Cache DSF2 Augment Directories to Yes. 6. Expand the Assignment Options group and set the Enable DSF2 Augment, Enable DPV, and Enable eLOT to Yes. 7. Include the DSF2 address attributes output fields in your output file setup. 16.6.4.7.4 DSF2 output fields When you perform DSF2 Augment processing in the software, address attributes are available in the following output fields for every address that was assigned. Be sure to include the fields containing information you'll need in your output file setup: • • • • • • • • • DSF2_Business_Indicator DSF2_Delivery_Type DSF2_Drop_Count DSF2_Drop_Indicator DSF2_Educational_Ind DSF2_LACS_Conversion_Ind DSF2_Record_Type DSF2_Seasonal_Indicator DSF2_Throwback_Indicator Note: A blank output in any of these fields means that the address was not looked up in the DSF2 directories. 526 2011-06-09
- 527. Data Quality Related Topics • Reference Guide: Data Quality fields, USA Regulatory Address Cleanse fields 16.6.4.7.5 Improve processing speed You can cache DSF2 data to improve DSF2 processing speed. To cache DSF2 data, Set the Cache DSF2 Augment Directories option in the Transform Performance group to Yes. The software caches only the directories needed for adding address attributes. 16.6.4.8 DSF2 walk sequencing When you perform DSF2 walk sequencing in the software, the software adds delivery sequence information to your data, which you can use with presorting software to qualify for walk-sequence discounts. Remember: The software does not place your data in walk sequence order. Include the DSF2 Walk Sequencer transform to enable walk sequencing. Related Topics • Reference Guide: Transforms, Data Quality transforms, DSF2® Walk Sequencer 16.6.4.8.1 Pseudo sequencing DSF2 walk sequencing is often called pseudo sequencing because it mimics USPS walk sequencing. Where USPS walk-sequence numbers cover every address, DSF2 processing provides pseudo sequence numbers for only the addresses in that particular file. 527 2011-06-09
- 528. Data Quality The software uses DSF2 data to assign sequence numbers for all addresses that are DPV-confirmed delivery points (DPV_Status = Y). Other addresses present in your output file that are not valid DPV-confirmed delivery points (DPV_Status = S, N, or D) will receive "0000" as their sequence number. All other addresses will have a blank sequence number. Note: When you walk-sequence your mail with the software, remember the following points: • Batch only. DSF2 walk sequencing is available only in batch mode. You cannot assign sequence numbers in real time. • Reprocess if you have made file changes. If your data changes in any way, you must re-assign sequence numbers. Sequence numbers are valid only for the data file as you process it at the time. 16.6.4.9 Break key creation Break keys create manageable groups of data. They are created when there are two or more fields to compare. The DSF2 Walk Sequencer transform automatically forms break groups before it adds walk sequence information to your data. The software creates break groups based on the Postcode1 and Sortcode_Route fields. Set options for how you want the software to configure the fields in the Data Collection Config group. Keeping the default settings optimizes the data flow and allows the software to make the break key consistent throughout the data. Option Default value Replace NULL with space Yes Right pad with spaces Yes 16.6.4.10 Enable DSF2 walk sequencing To enable DSF2 walk sequence, include the DSF2 Walk Sequencer transform in your data flow. 528 2011-06-09
- 529. Data Quality 16.6.4.10.1 Required information When you set up for DSF2 walk sequence processing, the following options in the USPS License Information group are required: • • • Licensee Name DSF2 Licensee ID List ID 16.6.4.10.2 To enable DSF2 walk sequencing The input file for the DSF2 Walk Sequencer transform must have been pre-processed with CASS-certified software (such as the USA Regulatory Address Cleanse transform). To obtain an additional postage discount, include the DSF2_Business_Indicator output field information from CASS-certified software. In addition to the required USPS License Information fields, make the following settings in the DSF2 Walk Sequencer transform: 1. Optional. Select Yes or No in the Common group, Run as Separate Process option. Select No if you are gathering DSF2 statistics. Select Yes to save processing time (if you don't need DSF2 statistics). 2. Enter the file path and file name (dsf.dir) to the Delivery Statistics directory in the DelStats Directory option in the Reference Files group. You may use the $$RefFilesAddressCleanse substitution parameter if you have it set up. 3. Enter the processing site location in the Site Location option of the Processing Options group. This is applicable only if you have more than one site location for DSF2 processing. 4. Make the following settings in the Data Collection Configuration group: • Select Yes or No in the Replace Null With Space option as desired. • Select Yes or No for the Right Pad With Spaces option as desired. • Select Yes or No for the Pre Sorted Data option (optional). We recommend that you keep the default setting of No so that Data Services sorts your data based on the break key fields (instead of using another software program). 16.6.4.11 DSF2 walk sequence input fields Here is a list of the DSF2 walk sequence input fields. Note: These fields must have been output from CASS-certified software processing before they can be used as input for the DSF2 Walk Sequencer transform: • • • • 529 Postcode1 Postcode2 Sortcode_Route LOT 2011-06-09
- 530. Data Quality • • • • LOT_Order Delivery_Point DPV_Status DSF2_Business_Indicator (optional) The software uses the information in these fields to determine the way the records should be ordered (walk sequenced) if they were used in a mailing list. The software doesn’t physically change the order of your database. The software assigns walk-sequence numbers to each record based on the information it gathers from these input fields. Note: All fields are required except for the DSF2_Business_Indicator field. The optional DSF2_Business_Indicator field helps the software determine if the record qualifies for saturation discounts. Saturation discounts are determined by the percentage of residential addresses in each carrier route. See the USPS Domestic Mail Manual for details about all aspects of business mailing and sorting discounts. Related Topics • Reference Guide: Transforms, DSF2® Walk Sequencer, Input fields 16.6.4.12 DSF2 walk-sequence output fields The software outputs walk-sequence number information to the following fields: • • • • • Active_Del_Discount Residential_Sat_Discount Sortcode_Route_Discount Walk_Sequence_Discount Walk_Sequence_Number Related Topics • Reference Guide: Data Quality fields, DSF2 Walk Sequencer, DSF2 Walk Sequencer output fields 16.6.4.13 DSF2 reporting There are reports and log files that the software generates for DSF2 augment and walk sequencing. Find complete information about these reports and log files in the Management Console Guide. 530 2011-06-09
- 531. Data Quality Delivery Sequence Invoice Report The USPS requires that you submit the Delivery Sequence Invoice report if you claim DSF2 walk-sequence discounts for this job. US Addressing Report • • The US Addressing Report is generated by the USA Regulatory Address Cleanse transform. The Second Generation Delivery Sequence File Summary and Address Delivery Types sections of the US Addressing Report shows counts and percentages of addresses in your file that match the various DSF2 categories (if NCOALink is enabled). The information is listed for pre and post NCOALink processing. DSF2 Augment Statistics Log File The USPS requires that DSF2 licensees save information about their processing in the DSF2 log file. The USPS dictates the contents of the DSF2 log file and requires that you submit it to them monthly. Log files are available to users with administrator or operator permissions. Related Topics • Management Console Guide: Administrator, Administrator management, Exporting DSF2 certification log • Management Console Guide: Data Quality reports, Delivery Sequence Invoice Report • Management Console Guide: Data Quality reports, US Addressing Report 16.6.4.13.1 DSF2 Augment Statistics Log File The DSF2 Augment Statistics Log File is stored in the repository. The software generates the log file to the repository where you can export them by using the Data Services Management Console (for Administrators or Operators only). The naming format for the log file is as follows: [DSF2_licensee_ID][mm][yy].dat The USPS dictates the contents of the DSF2 log file and requires that you submit it to them monthly. For details, see the DSF2 Licensee Performance Requirements document, which is available on the USPS RIBBS website (http://ribbs.usps.gov/dsf2/documents/tech_guides). You must submit the DSF2 log file to the USPS by the third business day of each month by e-mail. 16.6.5 NCOALink® overview The USPS Move Update standard helps users and the USPS to reduce the number of records that are returned because the address is out of date. NCOALink is a part of this effort. Move Updating is the 531 2011-06-09
- 532. Data Quality process of checking addresses against the National Change of Address (NCOA) database to make sure your data is updated with current addresses. When you process your data using NCOALink, you update your records for individuals or businesses that have moved and have filed a Change of Address (COA) form with the USPS. Other programs that are a part of Move Update, and that are supported in the USA Regulatory Address Cleanse transform, include, ANKLink®, and SuiteLink®. The USPS requires that your lists comply with Move Update standards in order for it to qualify for the discounted postal rates available for First-Class presorted mailings. You can meet this requirement through the NCOALink process. Note: Mover ID is the name under which SAP BusinessObjects Data Services is certified for NCOALink. Related Topics • About ANKLink • SuiteLink™ 16.6.5.1 The importance of move updating The USPS requires move updating on all First Class presorted mailings. To help mailers meet this requirement, the USPS offers certain options, including NCOALink. To keep accurate address information for your contacts, you must use a USPS method for receiving your contacts' new addresses. Not only is move updating good business, it is required for all First-Class mailers who claim presorted or automation rates. As the USPS expands move-updating requirements and more strictly enforces the existing regulations, move updating will become increasingly important. Related Topics • About ANKLink 16.6.5.2 Benefits of NCOALink By using NCOALink in the USA Regulatory Address Cleanse transform, you are updating the addresses in your lists with the latest move data. With NCOALink, you can: • • • 532 Improve mail deliverability. Reduce the cost and time needed to forward mail. Meet the USPS move-updating requirement for presorted First Class mail. 2011-06-09
- 533. Data Quality • Prepare for the possible expansion of move-update requirements. 16.6.5.3 How NCOALink works When processing addresses with NCOALink enabled, the software follows these steps: 1. The USA Regulatory Address Cleanse transform standardizes the input addresses. NCOALink requires parsed, standardized address data as input. 2. The software searches the NCOALink database for records that match your parsed, standardized records. 3. If a match is found, the software receives the move information, including the new address, if one is available. 4. The software looks up move records that come back from the NCOALink database to assign postal and other codes. 5. Depending on your field class selection, the output file contains: • The original input address. The complete and correct value found in the directories, standardized according to any settings that you defined in the Standardization Options group in the Options tab. (CORRECT) • The address components that have been updated with move-updated address data.(MOVE-UPDATED) Note: The transform looks for the move-updated address information in the U.S. National Directories. When the move-updated address is not found in the U.S. National Directories, the software populates the Move Updated fields with information found in the Move Update Directories only. The Move Updated fields that are populated as a result of standardizing against the U.S. National Directories will not be updated. • The move-updated address data if it exists and if it matches in the U.S. National directories. Or the field contains the original address data if a move does not exist or if the move does not match in the U.S. National Directories. (BEST) Based on the Apply Move to Standardized Fields option in the NCOALink group, standardized components can contain either original or move-updated addresses. 6. The software produces the reports and log files required for USPS compliance. 533 2011-06-09
- 534. Data Quality Example: 1. NCOALink requires parsed, standardized address data as input. Therefore, before NCOALink processing, the software performs its normal processing on the address data. 2. The software searches the NCOALink database for a record that matches your parsed, standardized record. 3. The software receives the move information, including the new address if one is available. 4. The software looks up the move record that comes back from the NCOALink database, to assign postal and other codes. 5. At your option, the software can either retain the old address and append the new, or replace the old address with the new. 6. The software produces the reports and log files that you will need for USPS compliance. 16.6.5.4 NCOALink provider levels NCOALink users fall in one of three categories of providers. Specify the service provider in the USPS License Information group of options under Provider Level. 534 2011-06-09
- 535. Data Quality Note: Only provider levels supported in your registered keycodes display in the selection list. Provider level Description Full Service Provider (FSP) Provides NCOALink processing to third parties. Limited Service Provider (LSP) Provides NCOALink processing to third parties and internally. End User Mailer (EUM) Provides NCOALInk processing to in-house lists only. 16.6.5.5 NCOALink brokers and list administrators An NCOALink user may have a broker or list administrator who owns the lists they are processing. When there is a broker or list administrator involved, add contact information in the NCOALink group under Contact Detail list > Contact Details. Broker Directs business to an NCOALink service provider. List Administrator List Administrator: Maintains and stores lists. List administrators are different than brokers in two ways: • • List administrators don't send move-updated files back to the list owner. List administrators may have an NCOALink license. If a list administrator, a broker, or both are involved in your job, you must complete Contact Detail List for each of them separately. You can duplicate a group of options by right-clicking the group name and choosing "Duplicate Option". 16.6.5.6 Address not known (ANKLink) Undeliverable-as-addressed (UAA) mail costs the mailing industry and the USPS a lot of money each year. The software provides NCOALink as an additional solution to UAA mail. With NCOALink, you also can have access to the USPS's ANKLink data. 16.6.5.6.1 About ANKLink NCOALink limited service providers and end users receive change of address data for the preceding 18 months. The ANKLink option enhances that information by providing additional data about moves that occurred in the previous months 19 through 48. 535 2011-06-09
- 536. Data Quality Tip: If you are an NCOALink full service provider you already have access to the full 48 months of move data (including the new addresses). Note: The additional 30 months of data that comes with ANKLink indicates only that a move occurred and the date of the move; the new address is not provided. The ANKLink data helps you make informed choices regarding a contact. If the data indicates that the contact has moved, you can choose to suppress that contact from the list or try to acquire the new address from an NCOALINK full service provider. If you choose to purchase ANKLink to extend NCOALINK information, then the DVD you receive from the USPS will contain both the NCOALink 18-month full change of address information and the additional 30 month ANKLink information which indicates that a move has occurred. If an ANKLink match exists, it is noted in the ANKLINK_RETURN_CODE output field and in the NCOALink Processing Summary report. 16.6.5.6.2 ANKLink data ANKLink is a subset of NCOALink. You can request ANKLink data from the USPS National Customer Support Center (NCSC) by calling 1-800-589-5766 or by e-mail at [email protected]. ANKLink data is not available from SAP BusinessObjects. The software detects if you're using ANKLink data. Therefore, you do not have to specify whether you're using ANKLink in your job setup. 16.6.5.6.3 ANKLink support for NCOALink provider levels The software supports three NCOALink provider levels defined by the USPS. Software options vary by provider level and are activated based on the software package that you purchased. The following table shows the provider levels and support: 536 2011-06-09
- 537. Data Quality Provider level Provide service to third parties COA data (months) Data reSupport ceived from for USPS ANKLink Full Service Yes. 48 Provider (FSP) Third party services must be at least 51% of all processing. Weekly No (no benefit) Limited Service Yes. 18 Provider (LSP) LSPs can both provide services to third parties and use the product internally. Weekly Yes End User Mailer No (EUM) Monthly Yes 18 Tip: If you are an NCOALink EUM, you may request an alternate stop processing agreement from the USPS. After you are approved by the USPS you may purchase the software's stop processing alternative functionality which allows DPV and LACSLink processing to continue after a false positive address record is detected. Related Topics • Stop Processing Alternative • DPV and LACSLink user types 16.6.5.7 Software performance In our tests, the software ran slower with NCOALink enabled than with it disabled. Your processing speed depends on the computer running the software and the percentage of input records affected by a move (more moves equals slower performance). Related Topics • Improving NCOALink processing performance 537 2011-06-09
- 538. Data Quality 16.6.5.8 Getting started with NCOALink Before you begin NCOALink processing you need to perform the following tasks: • Complete the USPS certification process to become an NCOALink service provider or end user. For information about certification, see the NCOALink Certification section following the link below. • Understand the available output strategies and performance optimization options. • Configure your job. Related Topics • NCOALink certification 16.6.5.9 What to expect from the USPS and SAP BusinessObjects NCOALink, and the license requirements that go with it, has created a new dimension in the relationship among mailers (you), the USPS, and vendors. It's important to be clear about what to expect from everyone. 16.6.5.9.1 Move updating is a business decision for you to make NCOALink offers an option to replace a person's old address with their new address. You as a service provider must decide whether you accept move updates related to family moves, or only individual moves. The USPS recommends that you make these choices only after careful thought about your customer relationships. Consider the following examples: • If you are mailing checks, account statements, or other correspondence for which you have a fiduciary responsibility, then move updating is a serious undertaking. The USPS recommends that you verify each move by sending a double postcard, or other easy-reply piece, before changing a financial record to the new address. • If your business relationship is with one spouse and not the other, then move updating must be handled carefully with respect to divorce or separation. Again, it may make sense for you to take the extra time and expense of confirming each move before permanently updating the record. 16.6.5.9.2 NCOALink security requirements Because of the sensitivity and confidentiality of change-of-address data, the USPS imposes strict security procedures on software vendors who use and provide NCOALink processing. One of the software vendor's responsibilities is to check that each list input to the USA Regulatory Address Cleanse transform contains at least 100 unique records. Therefore the USA Regulatory Address 538 2011-06-09
- 539. Data Quality Cleanse transform checks your input file for at least 100 unique records. These checks make verification take longer, but they are required by the USPS and they must be performed. If the software finds that your data does not have 100 unique records, it issues an error and discontinues processing. The process of checking for 100 unique records is a pre-processing step. So if the software does not find 100 unique records, there will be no statistics output or any processing performed on the input file. Related Topics • Getting started with NCOALink How the software checks for 100 unique records When you have NCOALink enabled in your job, the software checks for 100 unique records before any processing is performed on the data. The software checks the entire database for 100 unique records. If it finds 100 unique records, the job is processed as usual. However, if the software does not find 100 unique records, it issues an error stating that your input data does not have 100 unique records, or that there is not enough records to determine uniqueness. For the 100 unique record search, a record consists of all mapped input fields concatenated in the same order as they are mapped in the transform. Each record must be identical to another record for it to be considered alike (not unique). Example: Comparing records The example below illustrates how the software concatenates the fields in each record, and determines non-unique records. The first and last row in this example are not unique. 332 FRONT STREET NORTH LACROSSE WI 54601 332 FRONT STREET SOUTH LACROSSE WI 54601 331 FRONT STREET SOUTH LACROSSE WI 54601 332 FRONT STREET NORTH LACROSSE WI 54601 Finding unique records in multiple threads Sometimes input list have 100 unique records but the user still receives an error message stating that the list does not have 100 unique records. This can happen when there is a low volume of data in lists. To work around this problem, users can adjust the Degree of Parallelism (DOP) setting in their job. Low volume of data and DOP > 1 When an NCOALink job is set up with the DOP greater than 1, each thread checks for unique records within the first collection it processes and shares knowledge of the unique records it found with all other 539 2011-06-09
- 540. Data Quality threads. The first thread to finish processing it’s collection counts the unique records found by all threads up to that point in time and makes a decision regarding whether or not the 100 record minimum check has been satisfied. That thread may not necessarily be thread 1. For example, say your list has 3,050 records and you have the DOP set for 4. If the number of records per collection is 1000, each thread will have a collection of 1000 records except for the last thread which will only have 50 records. The thread processing 50 records is likely to finish its collection sooner and it may make the pass/fail decision before 100 unique records have been encountered. You may be able to successfully run this job if you lower the DOP. In this example, you could lower it to 3. 16.6.5.10 About NCOALink directories After you have completed the certification requirements and purchased the NCOALink product from the USPS, the USPS sends you the latest NCOALink directories monthly (if you’re an end user) or weekly (if you’re a limited or full service provider). The NCOALink directories are not provided by SAP BusinessObjects. The USPS requires that you use the most recent NCOALink directories available for your NCOALink jobs. Note: The NCOALink directories expire within 45 days. The software provides a DVD Verification (Installer) utility that installs (transfers and unpacks) the compressed files from the NCOALink DVD onto your system. The utility is available with a GUI (graphical user interface) or you can run it from a command line. If you are a service provider, then each day you run an NCOALink job, you must also download the daily delete file and install it in the same directory where your NCOALink directories are located. Related Topics • About the NCOALink daily delete file • To install NCOALink directories with the GUI 16.6.5.10.1 To install NCOALink directories with the GUI Prerequisites Ensure your system meets the following minimum requirements: • • • At least 60 GB of available disk space DVD drive Sufficient RAM. 1. Insert the USPS DVD containing the NCOALink directories into your DVD drive. 540 2011-06-09
- 541. Data Quality 2. Run the DVD Installer, located at $LINK_DIRbinncoadvdver.exe (Windows) or $LINK_DIR/bin/ncoadvdver (UNIX), where $LINK_DIR is the path to your software installation directory. For further installation details, see the online help available within the DVD Installer (choose Help > Contents). Note: For more information about required disk space for reference data, see the Product Availability Matrix at https://service.sap.com/PAM. Related Topics • SAP BusinessObjects information resources 16.6.5.10.2 To install NCOALink directories from the command line Prerequisites: Ensure your system meets the following minimum requirements: • • • At least 60 GB of available disk space DVD drive Sufficient RAM 1. Run the DVD Installer, located at $LINK_DIRbinncoadvdver.exe (Windows) or $LINK_DIR/bin/ncoadvdver (UNIX), where $LINK_DIR is the path to your installation directory. 2. To automate the installation process, use the ncoadvdver command with the following command line options: Option Description Windows UNIX -c /p:t -p:t Perform transfer. When using this option you must also specify the following: • DVD location with /d or -d • transfer location with /t or -t /p:u -p:u Perform unpack. When using this option, you must also specify the following: • DVD location with /d or -d • transfer location with /t or -t /p:v -p:v Perform verification. When using this option, you must also specify the transfer location with /t or -t. /d 541 Run selected processes in console mode (do not use the GUI). -d Specify DVD location. 2011-06-09
- 542. Data Quality Option Description Windows UNIX /t -t Specify transfer location. /nos -nos Do not stop on error (return failure code as exit status). /a -a Answer all warning messages with Yes. You can combine p options. For example, if you want to transfer, unpack, and verify all in the same process, enter /p:tuv or -p:tuv. After performing the p option specified, the program closes. Example: Your command line may look something like this: Windows ncoadvdver /p:tuv /d D: /t C:pwdirsncoa UNIX ncoadvdver [-c] [-a] [-nos] [-p:(t|u|v)][-d<path>] [-t <filename>] 16.6.5.11 About the NCOALink daily delete file If you are a service provider, then every day before you perform NCOALink processing, you must download the daily delete file and install it in the same directory as your NCOALink directories are located. The daily delete file contains records that are pending deletion from the NCOALink data. For example, if Jane Doe filed a change of address with the USPS and then didn’t move, Jane’s record would be in the daily delete file. Because the change of address is stored in the NCOALink directories, and they are updated only weekly or monthly, the daily delete file is needed in the interim, until the NCOALink directories are updated again. Note: If you are an end user, you only need the daily delete file for processing Stage I or II files. It is not required for normal NCOALink processing. Important points to know about the daily delete file: • 542 The software will fail verification if an NCOALink certification stage test is being performed and the daily delete file is not installed. 2011-06-09
- 543. Data Quality • • • USA Regulatory Address Cleanse transform supports only the ASCII version of the daily delete file. Do not rename the daily delete file. It must be named dailydel.dat. The software will issue a verification warning if the daily delete file is more than three days old. 16.6.5.11.1 To install the NCOALink daily delete file To download and install the NCOALink daily delete file, follow these steps: 1. Go to the USPS RIBBS site at http://ribbs.usps.gov/. 2. Click Move Update > NCOALink on the left side of the page. 3. Click Daily Delete Files and Daily Delete Header Files under Important Links. 4. Download the dailydel.dat file link and save it to the same location where your NCOALink directories are stored. 16.6.5.12 Output file strategies You can configure your output file to meet your needs. Depending on the Field Class Selection that you choose, components in your output file contain Correct, Move-updated, or Best information: • CORRECT: Outputs the original input address. The complete and correct value found in the directories, standardized according to any settings that you defined in the Standardization Options group in the Options tab. (CORRECT) • MOVE-UPDATED: Outputs the address components that have been updated with move-updated address data. Note: The transform looks for the move-updated address information in the U.S. National Directories. When the move-updated address is not found in the U.S. National Directories, the software populates the Move Updated fields with information found in the Move Update Directories only. The Move Updated fields that are populated as a result of standardizing against the U.S. National Directories will not be updated. • BEST: Outputs the move-updated address data if it exists and if it matches in the U.S. National directories. Or the field contains the original address data if a move does not exist or if the move does not match in the U.S. National Directories. Based on the Apply Move to Standardized Fields option setting in the NCOA option group, standardized components can contain original or move-updated addresses. By default the output option Apply Move to Standardized Fields is set to Yes and the software updates standardized fields to contain details about the updated address available through NCOALink. If you want to retain the old addresses in the standardized components and append the new ones to the output file, you must change the Apply Move to Standardized Fields option to No. Then you can use output fields such as NCOALINK_RETURN_CODE to determine whether a move occurred. One way to set up your output file is to replicate the input file format, then append extra fields for move data. 543 2011-06-09
- 544. Data Quality In the output records not affected by a move, most of the appended fields will be blank. Alternatively, you can create a second output file specifically for move records. Two approaches are possible: • Output each record once, placing move records in the second output file and all other records in the main output file. • Output move records twice; once to the main output file, and a second time to the second output file. Both of these approaches require that you use an output filter to determine whether a record is a move. 16.6.5.13 Improving NCOALink processing performance Many factors affect performance when processing NCOALink data. Generally the most critical factor is the volume of disk access that occurs. Often the most effective way to reduce disk access is to have sufficient memory available to cache data. Other critical factors that affect performance include hard drive speed, seek time, and the sustained transfer rate. When the time spent on disk access is minimized, the performance of the CPU becomes significant. Related Topics • Finding unique records in multiple threads 16.6.5.13.1 Operating systems and processors The computation involved in most of the software and NCOALink processing is very well-suited to the microprocessors found in most computers, such as those made by Intel and AMD. RISC style processors like those found in most UNIX systems are generally substantially slower for this type of computation. In fact a common PC can often run a single job through the software and NCOALink about twice as fast as a common UNIX system. If you’re looking for a cost-effective way of processing single jobs, a Windows server or a fast workstation can produce excellent results. Most UNIX systems have multiple processors and are at their best processing several jobs at once. You should be able to increase the degree of parallelism (DOP) in the data flow properties to maximize the processor or core usage on your system. Increasing the DOP depends on the complexity of the dataflow. 16.6.5.13.2 Memory NCOALink processing uses many gigabytes of data. The exact amount depends on your service provider level, the data format, and the specific release of the data from the USPS. In general, if performance is critical, and especially if you are an NCOALink full service provider and you frequently run very large jobs with millions of records, you should obtain as much memory as possible. You may want to go as far as caching the entire NCOALink data set. You should be able to cache the entire NCOALink data set using 20 GB of RAM, with enough memory left for the operating system. 544 2011-06-09
- 545. Data Quality 16.6.5.13.3 Data storage If at all possible, the hard drive you use for NCOALink data should be fully dedicated to that process, at least while your job is running. Other processes competing for the use of the same physical disk drive can greatly reduce your NCOALink performance. To achieve even higher transfer rates you may want to explore the possibility of using a RAID system (redundant array of independent discs). When the software accesses NCOALink data directly instead of from a cache, the most significant hard drive feature is the average seek time. 16.6.5.13.4 Data format The software supports both hash and flat file versions of NCOALink data. If you have ample memory to cache the entire hash file data set, that format may provide the best performance. The flat file data is significantly smaller, which means a larger share can be cached in a given amount of RAM. However, accessing the flat file data involves binary searches, which are slightly more time consuming than the direct access used with the hash file format. 16.6.5.13.5 Memory usage The optimal amount of memory depends on a great many factors. The “Auto” option usually does a good job of deciding how much memory to use, but in some cases manually adjusting the amount can be worthwhile. 16.6.5.13.6 Performance tips Many factors can increase or decrease NCOALink processing speed. Some are within your control and others may be inherent to your business. Consider the following factors: • Cache size—Using too little memory for NCOALink caching can cause unnecessary random file access and time-consuming hard drive seeks. Using far too much memory can cause large files to be read from the disk into the cache even when only a tiny fraction of the data will ever be used. The amount of cache that works best in your environment may require some testing to see what works best for your configuration and typical job size. • Directory location—It’s best to have NCOALink directories on a local solid state drive or a virtual RAM drive. Using a local solid state drive or virtual RAM drive eliminates all I/O for NCOALink while processing your job. If you have the directories on a hard drive, it’s best to use a defragmented local hard drive. The hard drive should not be accessed for anything other than the NCOALink data while you are running your job. • Match rate—The more records you process that have forwardable moves, the slower your processing will be. Retrieving and decoding the new addresses takes time, so updating a mailing list regularly will improve the processing speed on that list. • Input format—Ideally you should provide the USA Regulatory Address Cleanse transform with discrete fields for the addressee’s first, middle, and last name, as well as for the pre-name and post-name. If your input has only a name line, the transform will have to take time to parse it before checking NCOALink data. 545 2011-06-09
- 546. Data Quality • File size—Larger files process relatively faster than smaller files. There is overhead when processing any job, but if a job includes millions of records, a few seconds of overhead becomes insignificant. 16.6.5.14 To enable NCOALink processing You must have access to the following files: • NCOALink directories • Current version of the USPS daily delete file • DPV data files • LACSLink data files If you use a copy of the sample transform configuration, USARegulatoryNCOALink_AddressCleanse, NCOALink, DPV, and LACSLink are already enabled. 1. Open the USA Regulatory Address Cleanse transform and open the "Options" tab. 2. Set values for the options as appropriate for your situation. For more information about the USA Regulatory Address Cleanse transform fields, see the Reference Guide. The table below shows fields that are required only for specific provider levels. End user with Alternate stop processing Full or limited service provider yes yes yes List Owner NAICS Code 546 End user without alternate stop processing Licensee Name Option group Option name or subgroup yes yes yes 2011-06-09
- 547. Data Quality End user with Alternate stop processing Full or limited service provider no no yes Customer Company Name no yes yes Customer Company Address no yes yes Customer Company Locality no yes yes Customer Company Region no yes yes Customer Company Postcode1 no yes yes Customer Company Postcode2 no yes yes Customer Company Phone no no no List Processing Frequency yes yes yes List Received Date no no yes List Return Date no no yes Provider Level USPS License Information End user without alternate stop processing List ID Option group Option name or subgroup yes yes yes no All options are required, except Customer Parent Company Name and Customer Alternate Company Name. no All options are required, except Buyer Company Name and Postcode for Mail Entry. PAF Details subgroup no NCOALink Service Provider Options subgroup 547 no 2011-06-09
- 548. Data Quality Tip: If you are a service provider and you need to provide contact details for multiple brokers, expand the NCOALink group, right-click Contact Details and click Duplicate Option. An additional group of contact detail fields will be added below the original group. Related Topics • Reference Guide: USA Regulatory Address Cleanse transform • About NCOALink directories • About the NCOALink daily delete file • Output file strategies • Stop Processing Alternative 16.6.5.15 NCOALink log files The software automatically generates the USPS-required log files and names them according to USPS requirements. The software generates these log files to the repository where you can export them by using the Data Services Management Console. The software creates one log file per license ID. At the beginning of each month, the software starts new log files. Each log file is then appended with information about every NCOALink job processed that month for that specific license ID. The USPS requires that you save these log files for five years. The software produces the following move-related log files: • CSL (Customer Service log) • PAF (Processing Acknowlagement Form) customer Information log • BALA (Broker/Agent/List Administrator) log The following table shows the log files required for each provider level: Required for: Log file Limited or Full Service Providers Description CSL 548 End Users Yes Yes This log file contains one record per list that you process. Each record details the results of change-of-address processing. 2011-06-09
- 549. Data Quality Required for: Log file End Users Limited or Full Service Providers Description This log file contains the information that you provided for the PAF. PAF customer information log No Yes The log file lists each unique PAF entry. If a list is processed with the same PAF information, the information appears just once in the log file. When contact information for the list administrator has changed, then information for both the list administrator and the corresponding broker are written to the PAF log file. This log file contains all of the contact information that you entered for the broker or list administrator. The log file lists information for each broker or list administrator just once. BALA No Yes The USPS requires the Broker/Agent/List Administrator log file from service providers, even in jobs that do not involve a broker or list administrator. The software produces this log file for every job if you’re a certified service provider. Related Topics • Management Console Guide: NCOALink Processing Summary Report • Management Console Guide: Exporting NCOALink certification logs 16.6.5.15.1 Log file names The software follows the USPS file-naming scheme for the following log files: • Customer Service log • PAF Customer Information log • Broker/Agent/List Administrators log The table below describes the naming scheme for NCOALink log files. For example, P1234C10.DAT is a PAF Log file generated in December 2010 for a licensee with the ID 1234. 549 2011-06-09
- 550. Data Quality Character 1 Characters 2 -5 Character 6 Characters 7-8 Log type Platform ID Month Year B Broker log exactly four characters long 1 January C Customer service log 2 February P PAF log 3 March 4 April 5 May 6 June 7 July 8 August 9 September A October B November C December two characters , for example 10 for 2010 Extension .DAT 16.6.6 USPS eLOT® eLOT is available for U.S. records in the USA Regulatory Address Cleanse transform only. eLOT takes line of travel one step further. The original LOT narrowed the mail carrier's delivery route to the block face level (Postcode2 level) by discerning whether an address resided on the odd or even side of a street or thoroughfare. eLOT narrows the mail carrier's delivery route walk sequence to the house (delivery point) level. This allows you to sort your mailings to a more precise level. Related Topics • To enable eLOT 550 2011-06-09
- 551. Data Quality • Set up the reference files 16.6.6.1 To enable eLOT 1. Open the USA Regulatory Address Cleanse transform. 2. Open the "Options" tab, expand the Assignment Options group, and select Yes for the Enable eLOT option. 3. In the Reference Files group, set the path for your eLOT directory. You can use the subtitution varialble $$RefFilesAddressCleanse for this option if you have it set up. 16.6.7 Early Warning System (EWS) EWS helps reduce the amount of misdirected mail caused when valid delivery points are created between national directory updates. EWS is available for U.S. records in the USA Regulatory Address Cleanse transform only. 16.6.7.1 Overview of EWS The EWS feature is the solution to the problem of misdirected mail caused by valid delivery points that appear between national directory updates. For example, suppose that 300 Main Street is a valid address and that 300 Main Avenue does not exist. A mail piece addressed to 300 Main Avenue is assigned to 300 Main Street on the assumption that the sender is mistaken about the correct suffix. Now consider that construction is completed on a house at 300 Main Avenue. The new owner signs up for utilities and mail, but it may take a couple of months before the delivery point is listed in the national directory. All the mail intended for the new house at 300 Main Avenue will be mis-directed to 300 Main Street until the delivery point is added to the national directory. The EWS feature solves this problem by using an additional directory which informs CASS users of the existence of 300 Main Avenue long before it appears in the national directory. When using EWS processing, the previously mis-directed address now defaults to a 5-digit assignment. 551 2011-06-09
- 552. Data Quality 16.6.7.2 Start with a sample transform configuration If you want to use the USA Regulatory Address Cleanse transform with the EWS option turned on, it is best to start with the sample transform configuration for EWS processing named: USARegulatoryEWS_AddressCleanse. 16.6.7.3 EWS directory The EWS directory contains four months of rolling data. Each week, the USPS adds new data and drops a week's worth of old data. The USPS then publishes the latest EWS data. Each Friday, SAP BusinessObjects converts the data to our format (EWyymmdd.zip) and posts it on the SAP Buisiness User Support site at https://service.sap.com/bosap-downloads-usps. 16.6.7.4 To enable EWS EWS is already enabled when you use the software's EWS sample transform, USARegulatoryEWS_Ad dressCleanse. These steps show how to manually set EWS. 1. Open the USA Regulatory Address Cleanse transform. 2. Open the "Options" tab and expand the Assignment Options group. 3. select Enable for the Enable EWS option. 4. Expand the Reference Files group and enter a path for the EWS Directory option, or use the substitution variable $$RefFilesAddressCleanse if you have it set up. Related Topics • Early Warning System (EWS) 16.6.8 USPS RDI® The RDI option is available in the USA Regulatory Address Cleanse transform. RDI determines whether a given address is for a residence or non residence. 552 2011-06-09
- 553. Data Quality Parcel shippers can find RDI information to be very valuable because some delivery services charge higher rates to deliver to residential addresses. The USPS, on the other hand, does not add surcharges for residential deliveries. When you can recognize an address as a residence, you have increased incentive to ship the parcel with the USPS instead of with a competitor that applies a residential surcharge. According to the USPS, 91-percent of U.S. addresses are residential. The USPS is motivated to encourage the use of RDI by parcel mailers. You can use RDI if you are processing your data for CASS certification or if you are processing in a non-certified mode. In addition, RDI does not require that you use DPV processing. 16.6.8.1 Start with a sample transform If you want to use the RDI feature with the USA Regulatory Address Cleanse transform, it is best to start with the sample transform configuration, USARegulatoryRDI_AddressCleanse. Sample transforms are located in the Transforms tab of the Object Library. This sample is located under USA_Regulatory_Address_Cleanse transforms. 16.6.8.2 How RDI works After you install the USPS-supplied RDI directories (and then enable RDI processing), The software can determine if the address represented by an 11-digit postcode (Postcode1, Postcode2, and the DPBC) is a residential address or not. (The software can sometimes do the same with a postcode2.) The software indicates that an address is for a residence or not in the output component, RDI_INDICATOR. Using the RDI feature involves only a few steps: 1. Install USPS-supplied directories. 2. Specify where these directories are located. 3. Enable RDI processing in the software. 4. Run. Related Topics • To enable RDI 553 2011-06-09
- 554. Data Quality 16.6.8.2.1 Compatibility RDI has the following compatibility with other options in the software: • RDI is allowed in both CASS and non-CASS processing modes. • RDI is allowed with or without DPV processing. 16.6.8.3 RDI directory files RDI directories are available through the USPS. You purchase these directories directly from the USPS and install them according to USPS instructions to make them accessible to the software. RDI requires the following directories. File Description rts.hs11 For 11-digit postcode lookups (Postcode2 plus DPBC). This file is used when an address contains an 11-digit postcode. Determination is based on the delivery point. rts.hs9 For 9-digit postcode lookups (Postcode2). This file is based on a ZIP+4.This is possible only when the addresses for that ZIP+4 are for all residences or for no residences. 16.6.8.3.1 Specify RDI directory path In the Reference Files group, specify the location of your RDI directories in the RDI Path option. If RDI processing is disabled, the software ignores the RDI Path setting. 16.6.8.4 To enable RDI If you use a copy of the USARegulatoryRDI_AddressCleanse sample transform in your data flow, RDI is already enabled. However, if you are starting from a USA Regulatory Address Cleanse transform, make sure you enable RDI and set the location for the following RDI directories: rts.hs11 and rts.hs9. 554 2011-06-09
- 555. Data Quality 1. Open the USA Regulatory Address Cleanse transform. 2. In the "Options" tab expand the Reference Files group, and enter the location of the RDI directories in the RDI Path option, or use the substitution variable $$RefFilesAddressCleanse if you have it set up. 3. Expand the Assignment Options group, and select Yes for the Enable RDI option. 16.6.8.5 RDI output field For RDI, the software uses a single output component that is always one character in length. The RDI component is populated only when the Enable RDI option in the Assignment Options group is set to Yes. Job/Views field RDI_INDICATOR Length Description 1 This field contains the RDI value that consists of one of the following values: Y = The address is for a residence. N = The address is not for a residence. 16.6.8.6 RDI in reports A few of the software's reports have additional information because of the RDI feature. 16.6.8.6.1 CASS Statement, USPS Form 3553 The USPS Form 3553 contains an entry for the number of residences. (The CASS header record also contains this information.) 16.6.8.6.2 Statistics files The statistics file contains RDI counts and percentages. 555 2011-06-09
- 556. Data Quality 16.6.9 GeoCensus (USA Regulatory Address Cleanse) The GeoCensus option of the USA Regulatory Address Cleanse transform offers geographic and census coding for enhanced sales and marketing analysis. It is available for U.S. records only. Note: GeoCensus functionality in the USA Regulatory Address Cleanse transform will be deprecated in a future version. It is recommended that you upgrade any data flows that currently use the GeoCensus functionality to use the Geocoder transform. For instructions on upgrading from GeoCensus to the Geocoder transform, see the Upgrade Guide. Related Topics • How GeoCensus works • GeoCensus directories • To enable GeoCensus coding • Geocoding 16.6.9.1 How GeoCensus works By using GeoCensus, the USA Regulatory Address Cleanse transform can append latitude, longitude, and Census codes such as Census Tract/Block and Metropolitan Statistical Area (MSA) to your records, based on ZIP+4 codes. MSA is an aggregation of US counties into Metropolitan Statistical Areas assigned by the US Office of Management and Budget. You can apply the GeoCensus codes during address standardization and postcode2 assignment for simple, “one-pass” processing. The transform cannot, by itself, append demographic data to your records. The transform lays the foundation by giving you census coordinates via output fields. To append demographic information, you need a demographic database from another vendor. When you obtain one, we suggest that you use the matching process to match your records to the demographic database, and transfer the demographic information into your records. (You would use the MSA and Census block/tract information as match criteria, then use the Best Record transform to post income and other information.) Likewise, the transform does not draw maps. However, you can use the latitude and longitude assigned by the transform as input to third-party mapping applications. Those applications enable you to plot the locations of your customers and filter your database to cover a particular geographic area. 556 2011-06-09
- 557. Data Quality 16.6.9.2 The software provides census coordinates The software cannot, by itself, append demographic data to your records. The software simply lays the foundation by giving you census coordinates. To append demographic information, you need a demographic database from another vendor. When you get that, we suggest that you use our Match/Consolidate program to match your records to the demographic database and transfer the demographic information into your records. (In technical terms, you would use the MSA and Census block/tract information as match fields, then use the Group Posting feature to transfer income and other information. See the Match/Consolidate documentation for details and examples of group posting.) Likewise, the software does not draw maps. However, you can use the latitude and longitude assigned by the software as input to third-party mapping software. Those programs enable you to plot the locations of your customers and filter your database to cover a particular geographic area. 16.6.9.3 Get the most from the GeoCensus data You can combine GeoCensus with the functionality of mapping software to view your geo-enhanced information. It will help your organization build its sales and marketing strategies. Here are some of the ways you can use the GeoCensus data, with or without mapping products. 557 2011-06-09
- 558. Data Quality Information type How GeoCensus can help Market analysis You can use mapping applications to analyze market penetration, for instance. Companies striving to gain a clearer understanding of their markets employ market analysis. This way they can view sales, marketing, and demographic data on maps, charts, and graphs. The result is a more finely targeted marketing program. You will understand both where your customers are and the penetration you have achieved in your chosen markets. Predictive modeling and target marketing You can more accurately target your customers for direct response campaigns using geographic selections. Predictive modeling or other analytical techniques allow you to identify the characteristics of your ideal customer. This method incorporates demographic information used to enrich your customer database. From this analysis, it is possible to identify the best prospects for mailing or telemarketing programs. Media planning For better support of your advertising decisions, you may want to employ media planning. Coupling a visual display of key markets with a view of media outlets can help your organization make more strategic use of your advertising dollars. Territory management GeoCensus data provides a more accurate market picture for your organization. It can help you distribute territories and sales quotas more equitably. Direct sales Using GeoCensus data with market analysis tools and mapping software, you can track sales leads gathered from marketing activities. 16.6.9.4 GeoCensus directories The path and file names for the following directories must be defined in the Reference Files option group of the USA Regulatory Address Cleanse transform before you can begin GeoCensus processing. You can use the substitution variable $$RefFilesDataCleanse. 558 2011-06-09
- 559. Data Quality Directory name Description ageo1-10 Address-level GeoCensus directories are required if you choose Address for the Geo Mode option under the Assignment Options group. cgeo2.dir Centriod-level GeoCensus directory is required if you choose Centroid for the Geo Mode option under the Assignment Options group. 16.6.9.5 GeoCensus mode options To activate GeoCensus in the transform, you need to choose a mode in the Geo Mode option in the Assignment Options group. Mode Description Ad dress Processes Address-level GeoCensus only. Both Attempts to make an Address-level GeoCensus assignment first. If no assignment is made, it attempts to make a Centroid-level GeoCensus assignment. Cen troid Processes Centroid-level GeoCensus only. None Turns off GeoCensus processing. 16.6.9.6 GeoCensus output fields You must include at least one of the following generated output fields in the USA Regulatory Address Cleanse transform if you plan to use the GeoCensus option: • • • • • • • • 559 AGeo_CountyCode AGeo_Latitude AGeo_Longitude AGeo_MCDCode AGeo_PlaceCode AGeo_SectionCode AGeo_StateCode CGeo_BSACode 2011-06-09
- 560. Data Quality • • • • CGeo_Latitude CGeo_Longitude CGeo_Metrocode CGeo_SectionCode Find descriptions of these fields in the Reference Guide. 16.6.9.7 Sample transform configuration To process with the GeoCensus feature in the USA Regulatory Address Cleanse transform, it is best to start with the sample transform configuration created for GeoCensus. Find the sample configuration, USARegulatoryGeo_AddressCleanse, under USA_Regulatory_Address_Cleanse in the Object Library. 16.6.9.8 To enable GeoCensus coding If you use a copy of the USARegulatoryGeo_AddressCleanse sample transform file in your data flow, GeoCensus is already enabled. However, if you are starting from a USA Regulatory Address Cleanse transform, make sure you define the directory location and define the Geo Mode option. 1. Open the USA Regulatory Address Cleanse transform. 2. In the "Options" tab, expand the Reference Files group. 3. Set the locations for the cgeo.dir and ageo1-10.dir directories based on the Geo Mode you choose. 4. Expand the Assignment Options group, and select either Address, Centroid, or Both for the Geo Mode option. If you select None, the transform will not perform GeoCensus processing. Related Topics • GeoCensus (USA Regulatory Address Cleanse) 16.6.10 Z4Change (USA Regulatory Address Cleanse) The Z4Change option is based on a USPS directory of the same name. The Z4Change option is available in the USA Regulatory Address Cleanse transform only. 560 2011-06-09
- 561. Data Quality 16.6.10.1 Use Z4Change to save time Using the Z4Change option can save a lot of processing time, compared with running all records through the normal ZIP+4 assignment process. Z4Change is most cost-effective for databases that are large and fairly stable—for example, databases of regular customers, subscribers, and so on. In our tests, based on files in which five percent of records were affected by a ZIP+4 change, total batch processing time was one third the normal processing time. When you are using the transform interactively—that is, processing one address at a time—there is less benefit from using Z4Change. 16.6.10.2 USPS rules Z4Change is to be used only for updating a database that has previously been put through a full validation process. The USPS requires that the mailing list be put through a complete assignment process every three years. 16.6.10.3 Z4Change directory The Z4Change directory, z4change.dir, is updated monthly and is available only if you have purchased the Z4Change option for the USA Regulatory Address Cleanse transform. The Z4Change directory contains a list of all the ZIP Codes and ZIP+4 codes in the country. 16.6.10.4 Start with a sample transform If you want to use the Z4Change feature in the USA Regulatory Address Cleanse transform, it is best to start with the sample transform, USARegulatoryZ4Change_AddressCleanse. 561 2011-06-09
- 562. Data Quality 16.6.10.5 To enable Z4Change If you use a copy of the Z4Change transform configuration file sample(USARegulatoryZ4Change_Ad dressCleanse) in your data flow, Z4Change is already enabled. However, if you are starting from a USA Regulatory Address Cleanse transform, make sure you define the directory location and define the Z4Change Mode option. 1. Open the USA Regulatory Address Cleanse transform. 2. On the "Options" tab, expand the Reference Files group. 3. Set the location for the z4change.dir directory in the Z4Change Directory option. 4. Expand Z4Change options group and select Yes for the Enable Z4Change option. 5. In the Z4Change option group, enter the month and year that the input records were most recently ZIP+4 updated in the Last ZIP+4 Assign Date option. 16.6.11 Suggestion lists overview Suggestion List processing is used in transactional projects with the USA Regulatory Address Cleanse, Global Address Cleanse, and the Global Suggestion List transforms. Use suggestion lists to complete and populate addresses that have minimal data. Suggestion lists can offer suggestions for possible matches if an exact match is not found. This section is only about suggestion lists in the USA Regulatory Address Cleanse transform. Note: Suggestion list processing is not available for batch processing. In addition, if you have suggestion lists enabled, you are not eligible for CASS discounts and the software will not produce the required CASS documentation. Related Topics • Global Address Cleanse suggestion lists • Integrator's Guide: Using Data Services as a web service provider • Extracting data quality XML strings using extract_from_xml function 16.6.11.1 Introduction to suggestion lists 562 2011-06-09
- 563. Data Quality Ideally, when the USA Regulatory Address Cleanse transform looks up an address in the USPS postal directories (City/ZCF), it finds exactly one matching record with a matching combination of locality, region, and postcode. Then, during the look-up in the USPS national ZIP+4 directory, the software should find exactly one record that matches the address. Breaking ties Sometimes it's impossible to pinpoint an inut address to one matching record in the directory. At other times, the software may find several directory records that are near matches to the input data. When the software is close to a match, but not quite close enough, it assembles a list of the near matches and presents them as suggestions. When you choose a suggestion, the software tries again to assign the address. Example: Incomplete last line Given the incomplete last line below, the software could not reliably choose one of the four localities. But if you choose one, the software can proceed with the rest of the assignment process. Input record Possible matches in the City/ZCF directories Line1= 1000 vine La Crosse, WI 54603 Line2= lacr wi Lancaster, WI 53813 La Crosse, WI 54601 Larson, WI 54947 Example: Missing directional The same can happen with address lines. A common problem is a missing directional. In the example below, there is an equal chance that the directional could be North or South. The software has no basis for choosing one way or the other. Input record Possible matches in the ZIP+4 directory Line1 = 615 losey blvd 600-699 Losey Blvd N Line2 = 54603 600-698 Losey Blvd S Example: Missing suffix A missing suffix would cause similar behavior as in the example above. 563 2011-06-09
- 564. Data Quality Input record Possible matches in the ZIP+4 directory Line1 = 121 dorn 100-199 Dorn Pl Line2 = 54601 101-199 Dorn St Example: Misspelled street names A misspelled or incomplete street name could also result in the need to be presented with address suggestions. Input record Possible matches in the ZIP+4 directory Line1 = 4100 marl 4100-4199 Marshall 55421 Line2 = minneapolis mn 4100-4199 Maryland 55427 16.6.11.1.1 More information is needed When the software produces a suggestion list, you need some basis for selecting one of the possible matches. Sometimes you need more information before choosing a suggestion. Example • Operators taking information over the phone can ask for more information from the customer to decide which suggestion list to choose. • Operators entering data from a consumer coupon that is a little smudged may be able to choose a suggestion based on the information that is not smudged. 16.6.11.1.2 CASS rule The USPS does not permit SAP BusinessObjects Data Services to generate a USPS Form 3553 when suggestion lists are used in address assignment. The USPS suspects that users may be tempted to guess, which may result in misrouted mail that is expensive for the USPS to handle. Therefore, when you use the suggestion list feature, you cannot get a USPS Form 3553 covering the addresses that you assign. The form is available only when you process in batch mode with the Disable Certification option set to No. You must run addresses from real-time processes through a batch process in order to be CASS compliant. Then the software generates a USPS Form 3553 that covers your entire mailing database, and your list may be eligible for postal discounts. 564 2011-06-09
- 565. Data Quality 16.6.11.1.3 Integrating functionality Suggestion Lists functionality is designed to be integrated into your own custom applications via the Web Service. For information about integrating Data Services for web applications, see the Integrator's Guide. 16.6.11.1.4 Sample suggestion lists blueprint If you want to use the suggestion lists feature, it is best to start with one of the sample transforms configured for it. The sample transform is named USARegulatorySuggestions_Address_Cleanse. It is available for the USA Regulatory Address Cleanse transform. 16.6.12 Multiple data source statistics reporting Statistics based on logical groups For the USA Regulatory Address Cleanse transform, an input database can be a compilation of lists, with each list containing a field that includes a unique identifier. The unique identifier can be a name or a number, but it must reside in the same field across all lists. The software collects statistics for each list using the Data_Source_ID input field. You map the field that contains the unique identifier in your list to the software's Data_Source_ID input field. When the software generates reports, some of the reports will contain a summary for the entire list, and a separate summary per list based on the value mapped into the Data_Source_ID field. Restriction: For compliance with NCOALink reporting restrictions, the USA Regulatory Address Cleanse transform does not support processing multiple mailing lists associated with different PAFs. Therefore, for NCOALink processing, all records in the input file are considered to be a single mailing list and are reported as such in the Customer Service Log (CSL) file. Restriction: The Gather Statistics Per Data Source functionality is not supported when the Enable Parse Only or Enable Geo Only options in the Non Certified Options group are set to Yes. Related Topics • Gathering statistics per list 16.6.12.1 USPS certifications 565 2011-06-09
- 566. Data Quality The USA Regulatory Address Cleanse transform is CASS-certified. Therefore, when you process jobs with the USA Regulatory Address Cleanse transform (and it is set up correctly) you reap the benefits of that certification. If you integrate Data Services into your own software and you want to obtain CASS certification, you must follow the steps for CASS self-certification using your own software. You can also obtain licenses for DSF2 (Augment, Invoice, Sequence) and for NCOALink by using USA Regulatory Address Cleanse and DSF2 Walk Sequencer blueprints that are specifically set up for that purpose. Note: In this section we direct you to the USPS website and include names of documents and procedures. The USPS may change the address, procedure, or names of documents (and information required) without our prior knowledge. Therefore some of the information may become outdated. Related Topics • CASS self-certification • DSF2 Certification • Getting started with NCOALink 16.6.12.1.1 Completing USPS certifications The instructions below apply to USPS CASS self-certification, DSF2 license, and NCOALink license certification. During certification you must process files from the USPS to prove that your software is compliant with the requirements of your license agreement. The CASS, DSF2, and NCOALink certifications have two stages. Stage I is an optional test which includes answers that allow you to troubleshoot and prepare for the Stage II test. The Stage II test does not contain answers and is sent to the USPS for evaluation of the accuracy of your software configuration. 1. Complete the applicable USPS application (CASS, DSF2, NCOALink) and other required forms and return the information to the USPS. After you satisfy the initial application and other requirements, the USPS gives you an authorization code to purchase the CASS, DSF2, or NCOALink option. 2. Purchase the option from the USPS. Then submit the following information to SAP BusinessObjects: • your USPS authorization code (see step 1) • your NCOALink provider level (full service provider, limited service provider, or end user) (applicable for NCOALink only ) • your decision whether or not you want to purchase the ANKLink option (for NCOALink limited service provider or end user only) After you request and install the feature from SAP BusinessObjects, you are ready to request the applicable certification test from the USPS. The software provides blueprints to help you set up and run the certification tests. Import them from $$LINK_DIRDataQualityCertifications, where $$LINK_DIR is the software installation directory. 566 2011-06-09
- 567. Data Quality 3. Submit the Software Product Information form to the USPS and request a certification test. The USPS sends you test files to use with the blueprint. 4. After you successfully complete the certification tests, the USPS sends you the applicable license agreement. At this point, you also purchase the applicable product from SAP BusinessObjects. Related Topics • To set up the NCOALink blueprints • To set up the DSF2 certification blueprints • About ANKLink 16.6.12.1.2 Introduction to static directories Users who are self-certifying for CASS must use static directories. Those obtaining DSF2 licenses also need to use static directories. Static directories do not change every month with the regular directory updates. Instead, they can be used for certification for the duration of the CASS cycle. Using static directories ensures consistent results between Stage I and Stage II tests, and allows you to use the same directory information if you are required to re-test. You obtain static directories from SAP Business Objects. Note: If you do not use static directories when required, the software issues a warning. Static directories The following directories are available in static format: • • • • • • • • • • • zip4us.dir zip4us.shs zip4us.rev revzip4.dir city10.dir zcf10.dir dpv*.dir elot.dir ew*.dir SuiteLink directories LACSLink directories Obtaining static directories To request static directories, contact SAP Business User Support. Contact information (toll-free number and email address) is available at http://service.sap.com. 1. Click SAP Support Portal. 2. Click the "Help and Support " tab. 3. Click SAP BusinessObjects Support. 567 2011-06-09
- 568. Data Quality 4. Click Contact Support from the links at left. Static directories location It is important that you store your static directories separately from the production directories. If you store them in the same folder, the static directories will overwrite your production directories. We suggest that you create a folder named “static” to store your static directories. For example, save your static directories under $LINK_DIRDataQualityreferencestatic, where $LINK_DIR is the software's installation directory. Static directories safeguards To prevent running a production job using static directories, the software issues a verification warning or error under the following circumstances: • When the job has both static and non-static directories indicated. • When the release version of the zip4us.dir does not match the current CASS cycle in the software. • When the data versions in the static directories aren't all the same. For example, for CASS Cycle M the data versions in the static directories are M01. • When the job is set for self-certification but is not set up to use the static directories. • When the job is not set for self-certification but is set up to use the static directories. 16.6.12.1.3 To import certification blueprints The software includes blueprints to help you with certification testing. Additionally, the blueprints can be used to process a test file provided by the USPS during an audit. You need to first import the blueprints to the repository before you can use them in Data Services. To import the certification blueprints, follow these steps: 1. Open Data Services Designer. 2. Right-click in the Object Library area and select Repository > Import from file. 3. Go to $LINK_DIRDataQualitycertifications, where $LINK_DIR is the software installation directory. 4. Select the applicable blueprint and click Open. Note: A message appears asking for a pass phrase. The blueprints are not pass phrase protected, just click Import to advance to the next screen. 5. Click OK at the message warning that you are about to import the blueprints. Importing the blueprint files into the repository adds new projects, jobs, data flows, and flat file formats. The naming convention of the objects includes the certification type to indicate the associated certification test. 568 2011-06-09
- 569. Data Quality Related Topics • CASS self-certification blueprint • DSF2 Certification blueprints • NCOALink blueprints 16.6.12.1.4 CASS self-certification If you integrate Data Services into your own software, and you want to CASS-certify your software, you must obtain CASS certification on your own (self certification). You need to show the USPS that your software meets the CASS standards for accuracy of postal coding and address correction. You further need to show that your software can produce a facsimile of the USPS Form 3553 . You need a USPS Form 3553 to qualify mailings for postage discounts. Obtaining CASS certification on your own software is completely optional. However there are many benefits when your software is CASS certified. Visit the USPS RIBBS website at http://ribbs.usps.gov/index.cfm?page=cassmass for more information about CASS certification. Related Topics • Completing USPS certifications CASS self-certification process overview 1. Familiarize yourself with the CASS certification documentation and procedures located at http://ribbs.usps.gov/index.cfm?page=cassmass. 2. (Optional.) Download the CASS Stage I test from the RIBBS website. This is an optional step. You do not submit the Stage I test results to the USPS. Taking the Stage I test helps you analyze and correct any inconsistencies with the USPS-expected results before taking the Stage II test. 3. Import and make modifications to the CASS self-certification blueprint (us_cass_self_certifi cation.atl). The blueprint is located in $LINK_DIRDataQualityCertifications, where $$LINK_DIR is the software installation location. Edit the blueprint so it contains your static directories location, Stage I file location, your company name, and other settings that are required for CASS processing. 4. When you are satisfied that your Stage I test results compare favorably with the USPS-expected results, request the Stage II test from the USPS using the Stage II order form located on the RIBBS website. The USPS will place the Stage II test in your user area on the RIBBS website for you to download. 5. Download and unzip the Stage II test file to an output area. 6. After you run the Stage II file with the CASS self-certification blueprint, check that the totals on the USPS Form 3553 and the actual totals from the processed file match. 7. Compress the processed Stage II answer file (using WinZip for example) and name it using the same name as the downloaded Stage II file (step 5). 569 2011-06-09
- 570. Data Quality 8. Upload the processed Stage II answer file to your user area on the RIBBS website. The USPS takes about two weeks to grade your test. CASS self-certification blueprint SAP BusinessObjects provides a CASS self-certification blueprint. The blueprint contains the corresponding project, job, dataflow, and input/output formats. Additionally, the blueprint can be used to process a test file provided by the USPS during an audit. Import the us_cass_self_certification.atl blueprint from $LINK_DIRDataQualityCer tifications where $LINK_DIR is the software installation location. The table below contains the file names for the CASS self-certification blueprint: Object Name ATL file us_cass_self_certification.atl Project DataQualityCertificationCASS Job Job_DqBatchUSAReg_CASSSelfCert Dataflow DF_DqBatchUSAReg_CASSSelfCert Input file format DqUsaCASSSelfCert_In Output file format DqUsaCASSSelfCert _Out USPS Form 3553 required options for self certification The following options in the CASS Report Options group are required for CASS self certification. This information is included in the USPS Form 3553. 570 2011-06-09
- 571. Data Quality Option Description Company Name Cer- Specify the name of the company that owns the CASS-certified software. tified List Name Specify the name of the mailing list. List Owner Specify the name of the list owner. Note: Keep the CASS self-certification blueprints setting of “USPS”. Mailer Address(1-4) Specify the name and address of the person or organization for whom you are preparing the mailing (up to 29 characters per line). Software Version Specify the software name and version number that you are using to receive CASS self certification. Points to remember about CASS Remember these important points about CASS certification: • As an end user (you use Data Services to process your lists), you are not required to obtain CASS self certification because Data Services is already CASS certified. • CASS certification is given to software programs. You obtain CASS self certification if you have incorporated Data Services into your software program. • The CASS reports pertain to address lists. • CASS certification proves that the software can assign and standardize addresses correctly. 16.6.12.1.5 NCOALink certification The NCOALink certification consists of the following steps: 1. 2. 3. 4. Application and Self-Certification Statement Approval Software acquisition Testing and certification Execution of License Agreement This entire procedure is documented in the USPS Certification Procedures documents posted on the RIBBS website at http://ribbs.usps.gov/ncoalink/documents/tech_guides. Select either NCOALink End User Documents, NCOALink Limited Service Provider Documents, or NCOALink Full Service Provider Documents as applicable. 571 2011-06-09
- 572. Data Quality You must complete the appropriate service provider certification procedure for NCOALink in order to purchase the NCOALink product from the USPS. Related Topics • Getting started with NCOALink NCOALink software product information Use the information below to complete the Compliance Testing Product Information Form. Find this form on the RIBBS website at http://ribbs.usps.gov/ncoalink/documents/tech_guides. Click the Compliance Testing Form.doc link. Compliance Testing Product Information form Description Company Name & License Number Your specific information. The license number is the authorization code provided in your USPS approval letter. Company's NCOALink Product Name Mover ID for NCOALink Platform or Operating System Your specific information NCOALink Software Vendor SAP Americas, Inc. NCOALink Software Product Name Mover ID NCOALink Software Product Version ACE Address Matching ZIP+4 Product Name Contact SAP BusinessObjects Business User Support. Address Matching ZIP+4 Product Version Contact SAP BusinessObjects Business User Support. Address Matching ZIP+4 System Closed Is Software Hardware Dependent? No DPV® Product Name ACE DPV Product Version Contact SAP BusinessObjects Business User Support. LACSLink® Product Name ACE LACSLink Product Version Contact SAP BusinessObjects Business User Support. NCOALink Software options: Integrated or Standalone check boxes Integrated ANKLink Enhancement check box (applicable for Limited Service Providers and End Users) Check the box if you purchased the ANKLink option from SAP BusinessObjects. 572 2011-06-09
- 573. Data Quality Compliance Testing Product Information form Description HASH—FLAT—BOTH check boxes Indicate your preference. The software provides access to both file formats. NCOALink Level Option check boxes Check the appropriate box. Related Topics • Completing NCOALink certification • Data format Completing NCOALink certification During certification you must process files from the USPS to prove that you adhere to the requirements of your license agreement. NCOALink certification has two stages. Stage I is an optional test which includes answers that allow you to troubleshoot and prepare for the Stage II test. The Stage II test does not contain answers and is sent to the USPS for evaluation of the accuracy of your software configuration. Related Topics • To run the NCOALink certification jobs NCOALink blueprints SAP BusinessObjects provides NCOALink blueprints. The blueprints contain the corresponding projects, jobs, dataflows, and input/output formats. Additionally, the blueprints can be used to process a test file provided by the USPS during an audit. Import NCOALink blueprints from $LINK_DIRDataQualityCertification, where $LINK_DIR is the software installation location. The table below contains the file names for the Stage I NCOALink blueprints: Object ATL file us_ncoalink_stage_certification.atl Project DataQualityCertificationNCOALink Job Job_DqBatchUSAReg_NCOALinkStageI Dataflow DF_DqBatchUSAReg_NCOALinkStageI Input file format DqUsaNCOALinkStageI _in Output file format 573 Name DqUsaNCOALinkStageI _out 2011-06-09
- 574. Data Quality The table below contains the file names for the Stage II NCOALink blueprints: Object Name ATL file us_ncoalink_stage_certification.atl Project DataQualityCertificationNCOALink Job Job_DqBatchUSAReg_NCOALinkStageII Dataflow DF_DqBatchUSAReg_NCOALinkStageII Input file format DqUsaNCOALinkStageII _in Output file format DqUsaNCOALinkStageII _out To set up the NCOALink blueprints Before performing the steps below you must import the NCOALink blueprints. To set up the NCOALink Stage I and Stage II blueprints, follow the steps below. 1. In the Designer, select Tools > Substitution Parameter Configurations. The "Substitution Parameter Editor" opens. 2. Choose the applicable configuration from the Default Configuration drop list and enter values for your company's information and reference file locations. Click OK to close the Substitution Parameter Configurations tool. 3. Open the DataQualityCertificationsNCOALink project (which was imported with the blueprints). 4. Open the Job_DqBatchUSAReg_NCOALinkStageI job and then open the DF_Dq BatchUSAReg_NCOALinkStageI data flow. 5. Click the DqUsaNCOALinkStageI_in file to open the "Source File Editor". Find the Data Files(s) property group in the lower portion of the editor and make the following changes: a. In the Root Directory option, type the path or browse to the directory containing the input file. If you type the path, do not type a backslash () or forward slash (/) at the end of the file path. b. In the File name(s) option, change StageI.in to the name of the Stage file provided by the USPS. c. Exit the "Source File Editor". 6. Click the DqUsaNCOALinkStageI_out file to open the "Target File Editor". In the Data Files(s) property group make the following changes: a. In the Root Directory option, type the path or browse to the directory containing the output file. If you type the path, do not type a backslash () or forward slash (/) at the end of the file path. b. (Optional.) In the File name(s) option, change StageI.out to conform to your company's file naming convention. c. Exit the "Target File Editor". 574 2011-06-09
- 575. Data Quality 7. Double-click the USARegulatoryNCOALink_AddressCleanse transform to open the Transform Editor and click the "Options" tab. 8. Enter the correct path location to the reference files in the Reference Files group as necessary. Use the $$RefFilesAddressCleanse substitution variable to save time. 9. In the USPS License Information group, do the following: a. Enter a meaningful number in the List ID option. b. Enter the current date in the List Received Date and List Return Date options. c. Ensure that the provider level specified in the substitution parameter configuration by the $$USPSProviderLevel is accurate or specify the appropriate level (Full Service Provider, Limited Service Provider, or End User) in the Provider Level option. d. If you are a full service provider or limited service provider, complete the options in the NCOALink > PAF Details group and the NCOALink > Provider Options group. 10. When you are satisfied with the results of the Stage I test, repeat steps 4 through 9 to set up the Stage II objects found in the DF_DqBatchUSAReg_NCOALinkStage II data flow. Related Topics • Reference Guide: USPS license information options • DSF2 Certification blueprints • CASS self-certification blueprint • NCOALink blueprints • To import certification blueprints To run the NCOALink certification jobs Before you run the NCOALink certification jobs, ensure you have installed the DPV, LACSLink, and U.S. National directory files to the locations you specified during configuration and that the NCOALink DVD provided by the USPS is available. Running the Stage I job is optional; the results do not need to be sent to the USPS. However, running the Stage I job can help you ensure that you have configured the software correctly and are prepared to execute the Stage II job. 1. Use the NCOALink DVD Verification utility to install the NCOALink directories provided by the USPS. (See the link below for information about the NCOALink DVD Verification utility.) 2. Download the current version of the USPS daily delete file from http://ribbs.usps.gov/files/NCOALINK/index_dailyfiles.cfm. 3. Download the Stage I file from http://ribbs.usps.gov/ and uncompress it to the location you specified when setting up the certification job. Ensure the input file name in the source transform matches the name of the Stage I file from the USPS. 4. Execute the Stage I job and compare the test data with the expected results provided by the USPS in the Stage I input file. As necessary, make modifications to your configuration until you are satisfied with the results of your Stage I test. 575 2011-06-09
- 576. Data Quality 5. Download the Stage II file from the location specified by the USPS and uncompress it to the location you specified when setting up the certification job. Ensure the input file name in the transform matches the name of the Stage II file from the USPS. 6. Execute the Stage II job. Follow the specific instructions in the NCOALink Certification/Audit Instructions document that the USPS should have provided to you. 7. Compress the following results (using WinZip for example) and name it using the same name as the downloaded Stage II file (step 5): • Stage II output file • NCOALink Processing Summary Report • CASS Form 3553 • All log files generated in the $$CertificationLog path • • • Customer Service Log PAF (Service Providers only) Broker/Agent/List Administrator log (Service Providers only) 8. Send the results to the USPS for verification. Related Topics • Management Console Guide: Exporting NCOALink certification logs • To install NCOALink directories with the GUI • To install NCOALink directories from the command line • To install the NCOALink daily delete file 16.6.12.1.6 DSF2 Certification The DSF2 certification consists of the following steps: 1. 2. 3. 4. 5. Application and Self-Certification Statement Approval Documentation Requirements Stage I Interface Development DSF2 Testing and Certification Execution of License The entire process is detailed in the USPS DSF2 Certification Package document posted on the RIBBS website. Select the DSF2 Certification Package link on http://www.ribbs.usps.gov/dsf2/docu ments/tech_guides. The DSF2 Certification Package contains processes and procedures and the necessary forms for you to complete the five steps listed above. DSF2 Equipment Information for USPS certifications In the DSF2 Certification Package document, there is an Equipment Information form. You are required to provide information about the software you are using to certify for DSF2. Use the information in the following table as you complete the form for the DSF2 certification process. 576 2011-06-09
- 577. Data Quality Equipment Information form Description Interface Software Vendor SAP Americas, Inc. Interface Software Product Name ACE Interface Software Product Version Contact SAP BusinessObjects Business User Support. Address Matching ZIP+4 Product Name ACE Address Matching ZIP+4 Product Version Contact SAP BusinessObjects Business User Support. Address Matching ZIP+4 System Closed System Interface Hardware Vendor/Model/Type N/A The software is not hardware dependent Interface Hardware Operating System N/A The software is not hardware dependent Interface Hardware Serial Number N/A The software is not hardware dependent Find the DSF2 Certification Package document on the RIBBS website at www.ribbs.usps.gov/dsf2/documents/tech_guides. DSF2 Certification blueprints SAP BusinessObjects provides DSF2 certification blueprints for the three types of DSF2 certifications. The blueprints contain the corresponding projects, jobs, dataflows, and input/output formats. Additionally, the blueprints can be used to process a test file provided by the USPS during an audit. Import the DSF2 certification blueprints from $$LINK_DIRDataQualityCertifications where $$LINK_DIR is the software installation directory. The table below contains the file names for the USPS DSF2 Augment certification: Object Name ATL file us_dsf2_certification.atl Project DataQualityCertificationDSF2 Job Job_DqBatchUSAReg_DSF2Augment Dataflow DF_DqBatchUSAReg_DSF2Augment Input file format DqUsaDSF2Augment_in Output file format DqUsaDSF2Augment_out The table below contains the file names for the USPS DSF2 Invoice certification: 577 2011-06-09
- 578. Data Quality Project Name ATL file us_dsf2_certification.atl Project DataQualityCertificationDSF2 Job Job_DqBatchUSAReg_DSF2Invoice Dataflow DF_DqBatchUSAReg_DSF2Invoice Input file format DqUsaDSF2Invoice_in Output file format DqUsaDSF2Invoice_out The table below contains the file names for the USPS DSF2 Sequence certification: Project Name ATL file us_dsf2_certification.atl Project DataQualityCertificationDSF2 Job Job_DqBatchUSAReg_DSF2Sequence Dataflow DF_DqBatchUSAReg_DSF2Sequence Input file format DqUsaDSF2Sequence_in Output file format DqUsaDSF2Sequence_out To set up the DSF2 certification blueprints Before performing the steps below you must import the DSF2 blueprints. Follow these steps to set up the DSF2 Augment, Invoice, and Sequence certification blueprints. 1. In the Designer, select Tools > Substitution Parameter Configurations. The "Substitution Parameter Editor" opens. 2. Choose the applicable configuration from the Default Configuration drop list and enter values for your company's information and reference file locations. Note: DSF2 Augment only. Remember to enter the static directories location for the $$RefFilesUSPSStatic substitution variable. 3. Open the DataQualityCertificationDSF2 project (downloaded with the blueprint). 4. Expand the desired certification job and data flow. For example, if you are setting up for DSF2 Augment, expand the Job_DqBatchUSAReg_DSF2Augment job and then the DF_Dq BatchUSAReg_DSF2Augment data flow. 578 2011-06-09
- 579. Data Quality 5. Double-click the applicable input file format (*.in) to open the "Source File Editor". For example, for DSF2 Augment certification, double-click DSF2_Augment.in. 6. In the "Data Files(s)" property group make the following changes: a. In the Root Directory option, type the path or browse to the directory containing the input file. If you type the path, do not type a backslash or forward slash at the end of the file path. b. In the File name(s) option, change the input file name to the name of the file provided by the USPS. 7. Double-click the applicable output file format (*.out) to open the Target File Editor. For example, for DSF2 Augment certification, double-click DSF2_Augment.out. 8. In the Data Files(s) property group make the following changes: a. In the Root Directory option, type the path or browse to the directory containing the output file. If you type the path, do not type a backslash or forward slash at the end of the file path. b. (Optional) In the File name(s) option, change the output file name to conform to your company's file naming convention. 9. Click the USARegulatory_AddressCleanse transform to open the Transform Editor and click the "Options" tab. Note: For DSF2 Sequence and Invoice certifications, you will open the DSF2_Walk_Sequencer transform. 10. As necessary, in the Reference Files group, enter the correct path location to the reference files. For DSF2 Augment certification, use the $$RefFilesUSPSStatic substitution variable to save time. 11. In the CASS Report Options, update each option that is listed as “CHANGE_THIS” if applicable. Related Topics • DSF2 Certification blueprints • CASS self-certification blueprint • NCOALink blueprints • To import certification blueprints 16.6.12.2 Data_Source_ID field The software tracks statistics for each list based on the Data_Source_ID input field. Example: In this example there are 5 mailing lists combined into one list for input into the USA Regulatory Address Cleanse transform. Each list has a common field named List_ID, and a unique identifier in the List_ID field: N, S, E, W, C. The input mapping looks like this: 579 2011-06-09
- 580. Data Quality Transform Input Field Name Input Schema Column Name Type DATA_SOURCE_ID LIST_ID varchar(80) To obtain DPV statistics for each List_ID, process the job and then open the US Addressing report. The first DPV Summary section in the US Addressing report lists the Cumulative Summary, which reports the totals for the entire input set. Subsequent DPV Summary sections list summaries per Data_Source_ID. The example in the table below shows the counts and percentages for the entire database (cumulative summary) and for Data_Source_ID “N”. DPV Cumulative Summary Count % DPV Validated Addresses 1,968 3.94 214 4.28 Addresses Not DPV Valid 3,032 6.06 286 5.72 3 0.01 0 0.00 DPV Vacant Addresses 109 0.22 10 0.20 DPV NoStats Addresses 162 0.32 17 0.34 Statistic CMRA Validated Addresses DPV Summary for Data_Source_ID “N” % Related Topics • Group statistics reports 16.6.12.3 Gathering statistics per list Before setting up the USA Regulatory Address Cleanse transform to gather statistics per list, identify the field that uniquely identifies each list. For example, a mailing list that is comprised of more than one source might contain lists that have a field named LIST_ID that uniquely identifies each list. 1. Open the USA Regulatory Address Cleanse transform in the data flow and then click the "Options" tab. 2. Expand the Report and Analysis group and select Yes for the Generate Report Data and the Gather Statistics Per Data Source options. 3. Click the "Input" tab and click the "Input Schema Column Name" field next to the Data_Source_ID field for uniquely identifying a list. A drop menu appears. 580 2011-06-09
- 581. Data Quality 4. Click the drop menu and select the input field from your database that you've chosen as the common field for uniquely identifying a list. In the scenario above, that would be the LIST_ID field. 5. Continue with the remaining job setup tasks and execute your job. 16.6.12.4 Physical Source Field and Cumulative Summary Some reports include a report per list based on the Data_Source_ID field (Identified in the report footer by “Physical Source Field”), and a summary of the entire list (identified in the report footer by “Cumulative Summary”). However, the Address Standardization, Address Information Code, and USA Regulatory Locking reports will not include a Cumulative Summary. The records in these reports are sorted by the Data Source ID value. Note: When you enable NCOALink, the software reports a summary per list only for the following sections of the NCOALink Processing Summary Report: • • • NCOALink Move Type Summary NCOALink Return Code Summary ANKLink Return Code Summary Special circumstances There are some circumstances when the words “Cumulative Summary” and“ Physical Source Field” will not appear in the report footer sections. • • When the Gather Statistics Per Data Source option is set to No When the Gather Statistics Per Data Source option is set to Yes and there is only one Data Source ID value present in the list but it is empty 16.6.12.4.1 USPS Form 3553 and group reporting The USPS Form 3553 includes a summary of the entire list and a report per list based on the Data_Source_ID field. Example: Cumulative Summary The USPS Form 3553 designates the summary for the entire list with the words “Cumulative Summary”. It appears in the footer as highlighted in the Cumulative Summary report sample below. In addition, the Cumulative Summary of the USPS Form 3553 contains the total number of lists in the job in Section B, field number 5, "Number of Lists" (highlighted below). 581 2011-06-09
- 582. Data Quality Example: Physical Source Field The USPS Form 3553 designates the summary for each Individual list with the words "Physical Source Field" followed by the Data Source ID value. It appears in the footer as highlighted in the sample below. The data in the report is for that list only. 582 2011-06-09
- 583. Data Quality 16.6.12.4.2 Group statistics reports Reports that show both cumulative statistics (summaries for the entire mailing list) and group statistics (based on the Physical Source Field) include the following reports: • • • Address Validation Summary Address Type Summary US Addressing Reports that do not include a Cumulative Summary include the following: • • • Address Information Code Summary Address Standardization US Regulatory Locking Related Topics • Data_Source_ID field 16.7 Data Quality support for native data types The Data Quality transforms generally process incoming data types as character data. Therefore, if a noncharacter data type is mapped as input, the software converts the data to a character string before passing it through the Data Quality transforms. Some Data Quality data types are recognized and processed as the same data type as they were input. For example, if a date type field is mapped to a Data Quality date type input field, the software has the following advantages: • • Sortation: The transform recognizes and sorts the incoming data as the specified data type. Efficiency: The amount of data being converted to character data is reduced making processing more efficient. Related Topics • Data Quality transforms • Data types 16.7.1 Data Quality data type definitions The Data Quality transforms have four field attributes to define the field: • 583 Name 2011-06-09
- 584. Data Quality • • • Type Length Scale These attributes are listed in the Input and output tab of the transform editor. In the Input tab, the attribute Name is listed under the Transform Input Field Name column. The Type, Length, and Scale attributes are listed under the Type column in the format <type>(<length>, <scale>). The Output tab also contains the four field attributes listed above. The attribute Name is listed under the Field_Name column. The Type, Length, and Scale attributes are listed under the Type column in the format <type>(<length>, <scale>). 16.8 Data Quality support for NULL values The Data Quality transforms process NULL values as NULL. A field that is NULL is passed through processing with the NULL marker preserved unless there is data available to populate the field on output. When there is data available, the field is output with the data available instead of NULL. The benefit of this treatment of NULL is that the Data Quality transforms treat a NULL marker as unknown instead of empty. Note: If all fields of a record contain NULL, the transform will not process the record, and the record will not be a part of statistics and reports. Related Topics • Data Quality transforms • NULL values and empty strings 584 2011-06-09
- 585. Design and Debug Design and Debug This section covers the following Designer features that you can use to design and debug jobs: • Use the View Where Used feature to determine the impact of editing a metadata object (for example, at table). See which data flows use the same object. • Use the View Data feature to view sample source, transform, and target data in a data flow after a job executes. • Use the Interactive Debugger to set breakpoints and filters between transforms within a data flow and view job data row-by-row during a job execution. • Use the Difference Viewer to compare the metadata for similar objects and their properties. • Use the auditing data flow feature to verify that correct data is processed by a source, transform, or target object. Related Topics • Using View Where Used • Using View Data • Using the interactive debugger • Comparing Objects • Using Auditing 17.1 Using View Where Used When you save a job, work flow, or data flow the software also saves the list of objects used in them in your repository. Parent/child relationship data is preserved. For example, when the following parent data flow is saved, the software also saves pointers between it and its three children: • • a query transform • 585 a table source a file target 2011-06-09
- 586. Design and Debug You can use this parent/child relationship data to determine what impact a table change, for example, will have on other data flows that are using the same table. The data can be accessed using the View Where Used option. For example, while maintaining a data flow, you may need to delete a source table definition and re-import the table (or edit the table schema). Before doing this, find all the data flows that are also using the table and update them as needed. To access the View Where Used option in the Designer you can work from the object library or the workspace. 17.1.1 Accessing View Where Used from the object library You can view how many times an object is used and then view where it is used. 17.1.1.1 To access parent/child relationship information from the object library 1. View an object in the object library to see the number of times that it has been used. The Usage column is displayed on all object library tabs except: • Projects • Jobs • Transforms Click the Usage column heading to sort values. For example, to find objects that are not used. 2. If the Usage is greater than zero, right-click the object and select View Where Used. 586 2011-06-09
- 587. Design and Debug The "Output" window opens. The Information tab displays rows for each parent of the object you selected. The type and name of the selected object is displayed in the first column's heading. The As column provides additional context. The As column tells you how the selected object is used by the parent. Other possible values for the As column are: • For XML files and messages, tables, flat files, etc., the values can be Source or Target • For flat files and tables only: As Lookup() Lookup table/file used in a lookup function Lookup_ext() Lookup table/file used in a lookup_ext function Lookup_seq() • Description Lookup table/file used in a lookup_seq function For tables only: As Description Comparison Table used in the Table Comparison transform Key Generation Table used in the Key Generation transform 3. From the "Output" window, double-click a parent object. The workspace diagram opens highlighting the child object the parent is using. Once a parent is open in the workspace, you can double-click a row in the output window again. • If the row represents a different parent, the workspace diagram for that object opens. • If the row represents a child object in the same parent, this object is simply highlighted in the open diagram. This is an important option because a child object in the "Output" window might not match the name used in its parent. You can customize workspace object names for sources and targets. The software saves both the name used in each parent and the name used in the object library. The Information tab on the "Output" window displays the name used in the object library. The names of objects used in parents can only be seen by opening the parent in the workspace. 587 2011-06-09
- 588. Design and Debug 17.1.2 Accessing View Where Used from the workspace From an open diagram of an object in the workspace (such as a data flow), you can view where a parent or child object is used: • To view information for the open (parent) object, select View > Where Used, or from the tool bar, select the View Where Used button. The "Output" window opens with a list of jobs (parent objects) that use the open data flow. • To view information for a child object, right-click an object in the workspace diagram and select the View Where Used option. The "Output" window opens with a list of parent objects that use the selected object. For example, if you select a table, the "Output" window displays a list of data flows that use the table. 17.1.3 Limitations • This feature is not supported in central repositories. • Only parent and child pairs are shown in the Information tab of the Output window. For example, for a table, a data flow is the parent. If the table is also used by a grandparent (a work flow for example), these are not listed in the Output window display for a table. To see the relationship between a data flow and a work flow, open the work flow in the workspace, then right-click a data flow and select the View Where Used option. • The software does not save parent/child relationships between functions. • If function A calls function B, and function A is not in any data flows or scripts, the Usage in the object library will be zero for both functions. The fact that function B is used once in function A is not counted. • If function A is saved in one data flow, the usage in the object library will be 1 for both functions A and B. • • 588 Transforms are not supported. This includes custom ABAP transforms that you might create to support an SAP applications environment. The Designer counts an object's usage as the number of times it is used for a unique purpose. For example, in data flow DF1 if table DEPT is used as a source twice and a target once the object library displays its Usage as 2. This occurrence should be rare. For example, a table is not often joined to itself in a job design. 2011-06-09
- 589. Design and Debug 17.2 Using View Data View Data provides a way to scan and capture a sample of the data produced by each step in a job, even when the job does not execute successfully. View imported source data, changed data from transformations, and ending data at your targets. At any point after you import a data source, you can check on the status of that data—before and after processing your data flows. Use View Data to check the data while designing and testing jobs to ensure that your design returns the results you expect. Using one or more View Data panes, you can view and compare sample data from different steps. View Data information is displayed in embedded panels for easy navigation between your flows and the data. Use View Data to look at: • Sources and targets View Data allows you to see data before you execute a job. Armed with data details, you can create higher quality job designs. You can scan and analyze imported table and file data from the object library as well as see the data for those same objects within existing jobs. Of course after you execute the job, you can refer back to the source data again. • Transforms • Lines in a diagram Note: • • View Data displays blob data as <blob>. View Data is not supported for SAP IDocs. For SAP and PeopleSoft, the Table Profile tab and Column Profile tab options are not supported for hierarchies. Related Topics • Viewing data passed by transforms • Using the interactive debugger 17.2.1 Accessing View Data 17.2.1.1 To View data for sources and targets 589 2011-06-09
- 590. Design and Debug You can view data for sources and targets from two different locations: 1. View Data button View Data buttons appear on source and target objects when you drag them into the workspace. Click the View data button (magnifying glass icon) to open a View Data pane for that source or target object. 2. Object library View Data in potential source or target objects from the Datastores or Formats tabs. Open a View Data pane from the object library in one of the following ways: • Right-click a table object and select View Data. • Right-click a table and select Open or Properties. The Table Metadata, XML Format Editor, or Properties window opens. From any of these windows, you can select the View Data tab. To view data for a file, the file must physically exist and be available from your computer's operating system. To view data for a table, the table must be from a supported database. Related Topics • Viewing data in the workspace 17.2.2 Viewing data in the workspace View Data can be accessed from the workspace when magnifying glass buttons appear over qualified objects in a data flow. This means: For sources and targets, files must physically exist and be accessible from the Designer, and tables must be from a supported database. To open a View Data pane in the Designer workspace, click the magnifying glass button on a data flow object. 590 2011-06-09
- 591. Design and Debug A large View Data pane appears beneath the current workspace area. Click the magnifying glass button for another object and a second pane appears below the workspace area (Note that the first pane area shrinks to accommodate the presence of the second pane). You can open two View Data panes for simultaneous viewing. When both panes are filled and you click another View Data button, a small menu appears containing window placement icons. The black area in each icon indicates the pane you want to replace with a new set of data. Click a menu option and the data from the latest selected object replaces the data in the corresponding pane. The description or path for the selected View Data button displays at the top of the pane. • For sources and targets, the description is the full object name: • ObjectName ( Datastore.Owner ) for tables • FileName ( File Format Name ) for files • For View Data buttons on a line, the path consists of the object name on the left, an arrow, and the object name to the right. For example, if you select a View Data button on the line between the query named Query and the target named ALVW_JOBINFO(joes.DI_REPO), the path would indicate: Query -> ALVW_JOBINFO(Joes.DI_REPO) You can also find the View Data pane that is associated with an object or line by: • 591 Rolling your cursor over a View Data button on an object or line. The Designer highlights the View Data pane for the object. 2011-06-09
- 592. Design and Debug • Looking for grey View Data buttons on objects and lines. The Designer displays View Data buttons on open objects with grey rather than white backgrounds. Related Topics • Viewing data passed by transforms 17.2.3 View Data Properties You can access View Data properties from tool bar buttons or the right-click menu. View Data displays your data in the rows and columns of a data grid. The number of rows displayed is determined by a combination of several conditions: • Sample size — The number of rows sampled in memory. Default sample size is 1000 rows for imported source and target objects. Maximum sample size is 5000 rows. Set sample size for sources and targets from Tools > Options > Designer > General > View Data sampling size. When using the interactive debugger, the software uses the Data sample rate option instead of sample size. • Filtering • Sorting If your original data set is smaller or if you use filters, the number of returned rows could be less than the default. You can see which conditions have been applied in the navigation bar. Related Topics • Filtering • Sorting • Starting and stopping the interactive debugger 17.2.3.1 Filtering You can focus on different sets of rows in a local or new data sample by placing fetch conditions on columns. 592 2011-06-09
- 593. Design and Debug 17.2.3.1.1 To view and add filters 1. In the View Data tool bar, click the Filters button, or right-click the grid and select Filters. The Filters window opens. 2. Create filters. The Filters window has three columns: a. Column—Select a name from the first column. Select {remove filter} to delete the filter. b. Operator—Select an operator from the second column. c. Value—Enter a value in the third column that uses one of the following data type formats Data Type Format Integer, double, real standard date yyyy.mm.dd time hh24:mm:ss datetime yyyy.mm.dd hh24:mm.ss varchar 'abc' 3. In the Concatenate all filters using list box, select an operator (AND, OR) for the engine to use in concatenating filters. Each row in this window is considered a filter. 4. To see how the filter affects the current set of returned rows, click Apply. 5. To save filters and close the Filters window, click OK. Your filters are saved for the current object and the local sample updates to show the data filtered as specified in the Filters dialog. To use filters with a new sample, see Using Refresh. Related Topics • Using Refresh 17.2.3.1.2 To add a filter for a selected cell 1. Select a cell from the sample data grid. 593 2011-06-09
- 594. Design and Debug 2. In the View Data tool bar, click the Add Filter button, or right-click the cell and select Add Filter. The Add Filter option adds the new filter condition, <column> = <cell value>, then opens the Filters window so you can view or edit the new filter. 3. When you are finished, click OK. To remove filters from an object, go to the View Data tool bar and click the Remove Filters button, or right-click the grid and select Remove Filters. All filters are removed for the current object. 17.2.3.2 Sorting You can click one or more column headings in the data grid to sort your data. An arrow appears on the heading to indicate sort order: ascending (up arrow) or descending (down arrow). To change sort order, click the column heading again. The priority of a sort is from left to right on the grid. To remove sorting for an object, from the tool bar click the Remove Sort button, or right-click the grid and select Remove Sort. Related Topics • Using Refresh 17.2.3.3 Using Refresh To fetch another data sample from the database using new filter and sort settings, use the Refresh command. After you edit filtering and sorting, in the tool bar click the Refresh button in the tool bar, or right-click the data grid and select Refresh. 594 2011-06-09
- 595. Design and Debug To stop a refresh operation, click the Stop button. While the software is refreshing the data, all View Data controls except the Stop button are disabled. 17.2.3.4 Using Show/Hide Columns You can limit the number of columns displayed in View Data by using the Show/Hide Columns option from: • The tool bar. • The right-click menu. • The arrow shortcut menu, located to the right of the Show/Hide Columns tool bar button. This option is only available if the total number of columns in the table is ten or fewer. Select a column to display it. You can also "quick hide" a column by right-clicking the column heading and selecting Hide from the menu. 17.2.3.4.1 To show or hide columns 1. Click the Show/Hide columns tool bar button, or right-click the data grid and select Show/Hide Columns. The Column Settings window opens. 2. Select the columns that you want to display or click one of the following buttons: Show, Show All, Hide, or Hide All. 3. Click OK. 17.2.3.5 Opening a new window To see more of the data sample that you are viewing in a View Data pane, open a full-sized View Data window. From any View Data pane, click the Open Window tool bar button to activate a separate, 595 2011-06-09
- 596. Design and Debug full-sized View Data window. Alternatively, you can right-click and select Open in new window from the menu. 17.2.4 View Data tool bar options The following options are available on View Data panes. Icon Description Open in new window Opens the View Data pane in a larger window. See Opening a new window. Save As Saves the data in the View Data pane. Print Prints View Data pane data. Copy Cell Copies View Data pane cell data. Refresh data Fetches another data sample from existing data in the View Data pane using new filter and sort settings. See Using Refresh. Open Filters window Opens the Filters window. See Filtering. Add a Filter See To add a filter for a selected cell. Remove Filter Removes all filters in the View Data pane. Remove Sort 596 Option Removes sort settings for the object you select. See Sorting. 2011-06-09
- 597. Design and Debug Icon Option Description Show/hide navigation Shows or hides the navigation bar which appears below the data table. Show/hide columns See Using Show/Hide Columns 17.2.5 View Data tabs The View Data panel for objects contains three tabs: • • • Data tab Profile tab Column Profile tab Use tab options to give you a complete profile of a source or target object. The Data tab is always available. The Profile and Relationship tabs are supported with the Data Profiler. Without the Data Profiler, the Profile and Column Profile tabs are supported for some sources and targets (see Release Notes for more information). Related Topics • Viewing the profiler results 17.2.5.1 Data tab The Data tab allows you to use the properties of View Data. It also indicates nested schemas such as those used in XML files and messages. When a column references nested schemas, that column is shaded yellow and a small table icon appears in the column heading. Related Topics • View Data Properties 17.2.5.1.1 To view a nested schema 1. Double-click a cell. 597 2011-06-09
- 598. Design and Debug The data grid updates to show the data in the selected cell or nested table. In the Schema area, the selected cell value is marked by a special icon. Also, tables and columns in the selected path are displayed in blue, while nested schema references are displayed in grey. In the Data area, data is shown for columns. Nested schema references are shown in angle brackets, for example <CompanyName>. 2. Continue to use the data grid side of the panel to navigate. For example: • Select a lower-level nested column and double-click a cell to update the data grid. • Click the at the top of the data grid to move up in the hierarchy. • See the entire path to the selected column or table displayed to the right of the Drill Up button. Use the path and the data grid to navigate through nested schemas. 17.2.5.2 Profile tab If you use the Data Profiler, the Profile tab displays the profile attributes that you selected on the Submit Column Profile Request option. The Profile tab allows you to calculate statistical information for any set of columns you choose. This optional feature is not available for columns with nested schemas or for the LONG data type. Related Topics • Executing a profiler task 17.2.5.2.1 To use the Profile tab without the Data Profiler 1. Select one or more columns. Select only the column names you need for this profiling operation because Update calculations impact performance. You can also right-click to use the Select All and Deselect All menu options. 2. Click Update. 3. The statistics appear in the Profile grid. The grid contains six columns: 598 2011-06-09
- 599. Design and Debug Column Description Column Names of columns in the current table. Select names from this column, then click Update to populate the profile grid. Distinct Values The total number of distinct values in this column. NULLs The total number of NULL values in this column. Min Of all values, the minimum value in this column. Max Of all values, the maximum value in this column. Last Updated The time that this statistic was calculated. Sort values in this grid by clicking the column headings. Note that Min and Max columns are not sortable. In addition to updating statistics, you can click the Records button on the Profile tab to count the total number of physical records in the object you are profiling. The software saves previously calculated values in the repository and displays them until the next update. 17.2.5.3 Column Profile tab The Column Profile tab allows you to calculate statistical information for a single column. If you use the Data Profiler, the Relationship tab displays instead of the Column Profile. Note: This optional feature is not available for columns with nested schemas or the LONG data type. Related Topics • To view the relationship profile data generated by the Data Profiler 17.2.5.3.1 To calculate value usage statistics for a column 1. Enter a number in the Top box. This number is used to find the most frequently used values in the column. The default is 10, which means that the software returns the top 10 most frequently used values. 2. Select a column name in the list box. 3. Click Update. The Column Profile grid displays statistics for the specified column. The grid contains three columns: 599 2011-06-09
- 600. Design and Debug Column Description Value A "top" (most frequently used) value found in your specified column, or "Other" (remaining values that are not used as frequently). Total The total number of rows in the specified column that contain this value. Percentage The percentage of rows in the specified column that have this value compared to the total number of values in the column. The software returns a number of values up to the number specified in the Top box, plus an additional value called "Other." So, if you enter 5 in the Top box, you will get up to 6 returned values (the top 5 used values in the specified column, plus the "Other" category). Results are saved in the repository and displayed until you perform a new update. For example, statistical results in the preceding table indicate that of the four most frequently used values in the Name column, 50 percent use the value Item3, 20 percent use the value Item2, and so on. You can also see that the four most frequently used values (the "top four") are used in 90 percent of all cases, as only 10 percent is shown in the Other category. For this example, the total number of rows counted during the calculation for each top value is 1000. 17.3 Using the interactive debugger The Designer includes an interactive debugger that allows you to examine and modify data row-by-row (during a debug mode job execution) by placing filters and breakpoints on lines in a data flow diagram. The interactive debugger provides powerful options to debug a job. Note: A repository upgrade is required to use this feature. 600 2011-06-09
- 601. Design and Debug 17.3.1 Before starting the interactive debugger Like executing a job, you can start the interactive debugger from the Debug menu when a job is active in the workspace. Select Start debug, set properties for the execution, then click OK. The debug mode begins. The Debug mode provides the interactive debugger's windows, menus, and tool bar buttons that you can use to control the pace of the job and view data by pausing the job execution using filters and breakpoints. While in debug mode, all other Designer features are set to read-only. To exit the debug mode and return other Designer features to read/write, click the Stop debug button on the interactive debugger toolbar. All interactive debugger commands are listed in the Designer's Debug menu. The Designer enables the appropriate commands as you progress through an interactive debugging session. Before you start a debugging session, however, you might want to set the following: • Filters and breakpoints • Interactive debugger port between the Designer and an engine. 17.3.1.1 Setting filters and breakpoints You can set any combination of filters and breakpoints in a data flow before you start the interactive debugger. The debugger uses the filters and pauses at the breakpoints you set. If you do not set predefined filters or breakpoints: • The Designer will optimize the debug job execution. This often means that the first transform in each data flow of a job is pushed down to the source database. Consequently, you cannot view the data in a job between its source and the first transform unless you set a predefined breakpoint on that line. • You can pause a job manually by using a debug option called Pause Debug (the job pauses before it encounters the next transform). Related Topics • Push-down optimizer 17.3.1.1.1 To set a filter or breakpoint 1. In the workspace, open the job that you want to debug. 2. Open one of its data flows. 601 2011-06-09
- 602. Design and Debug 3. Right-click the line that you want to examine and select Set Filter/Breakpoint. A line is a line between two objects in a workspace diagram. The Breakpoint window opens. Its title bar displays the objects to which the line connects. 4. Set and enable a filter or a breakpoint using the options in this window. A debug filter functions as a simple Query transform with a WHERE clause. Use a filter to reduce a data set in a debug job execution. Note that complex expressions are not supported in a debug filter. Place a debug filter on a line between a source and a transform or two transforms. If you set a filter and a breakpoint on the same line, The software applies the filter first. The breakpoint can only see the filtered rows. Like a filter, you can set a breakpoint between a source and transform or two transforms. A breakpoint is the location where a debug job execution pauses and returns control to you. Choose to use a breakpoint with or without conditions. • • If you use a breakpoint without a condition, the job execution pauses for the first row passed to a breakpoint. If you use a breakpoint with a condition, the job execution pauses for the first row passed to the breakpoint that meets the condition. A breakpoint condition applies to the after image for UPDATE, NORMAL and INSERT row types and to the before image for a DELETE row type. Instead of selecting a conditional or unconditional breakpoint, you can also use the Break after 'n' row(s) option. In this case, the execution pauses when the number of rows you specify pass through the breakpoint. 5. Click OK. The Breakpoint enabled icon appears on the selected line. The software provides the following filter and breakpoint conditions: Icon Description Breakpoint disabled Breakpoint enabled Filter disabled Filter enabled Filter and breakpoint disabled 602 2011-06-09
- 603. Design and Debug Icon Description Filter and breakpoint enabled Filter enabled and breakpoint disabled Filter disabled and breakpoint enabled In addition to the filter and breakpoint icons that can appear on a line, the debugger highlights a line when it pauses there. A red locator box also indicates your current location in the data flow. For example, when you start the interactive debugger, the job pauses at your breakpoint. The locator box appears over the breakpoint icon as shown in the following diagram: A View Data button also appears over the breakpoint. You can use this button to open and close the View Data panes. As the debugger steps though your job's data flow logic, it highlights subsequent lines and displays the locator box at your current position. Related Topics • Panes 603 2011-06-09
- 604. Design and Debug 17.3.1.2 Changing the interactive debugger port The Designer uses a port to an engine to start and stop the interactive debugger. The interactive debugger port is set to 5001 by default. 17.3.1.2.1 To change the interactive debugger port setting 1. Select Tools > Options > Designer > Environment. 2. Enter a value in the InteractiveDebugger box. 3. Click OK. 17.3.2 Starting and stopping the interactive debugger A job must be active in the workspace before you can start the interactive debugger. You can se