SlideShare a Scribd company logo
SCons: an introduction


              Dean Giberson <dean@deangiberson.com>


                                   Dec 17, 2008




    If you've tried putting a build system together for a modern game you know
that you are facing a monumental, never ending task. Once you do get a basic
pipeline working, modern games require source assets that are in the range of
100 Gb of raw data. Tracking this amount of data takes time and resources.
    SCons (http://www.scons.org) is a Python based make replacement; with
it you can tame your dependencies and data sizes. What follows is quick art
centric introduction.

1 The Canonical Example
The SCons documentation1 would give an example like this:

     env = Environment()

     env.Program( 'HelloWorld', ['HelloWorld.cpp'])


   follow this and provide a compilable HelloWorld.cpp and you would have a
program ready to go. But how would you extend this to use your own pipeline
tools?

2 Adding Compressed Textures
Lets assume that you want to compress textures for use in a game. NVidia
Textures2 Tools can be used on your TGA/PNG/JPG/DDS images. For exam-
ple you want to convert this image3 into a DXT5 compressed image for use on
target:

     nvcompress -color -bc3 gunmap.png gunmap.dds

  1 http://www.scons.org/doc/0.97/HTML/scons-user/book1.html
  2 http://developer.nvidia.com/object/texture_tools.html
  3 Images provide by G3D Engine Data Collection http://g3d-cpp.sourceforge.net/.



                                           1
Once we understand how to do this from the command line adding it to
SCons is quite easy.


     import SCons

     env = Environment(ENV = os.environ)

     env['NVCOMPRESS'] = 'nvcompress'
     env['NVCOMPRESSTYPE'] = SCons.Util.CLVar('-color')
     env['NVCOMPRESSFLAGS'] = SCons.Util.CLVar('-bc3')
     env['NVCOMPRESSCOM'] = '$NVCOMPRESS $NVCOMPRESSFLAGS $SOURCE $TARGET'

     NvCompressAction = Action( '$NVCOMPRESSCOM')
     NvCompressBuilder = Builder( action=NvCompressAction,
                                  suffix='.dds')

     env['BUILDERS']['NvCompress'] = NvCompressBuilder


     env.NvCompress( 'image', 'image.png')




                                  2
3 A Quick walk-through to understand what is
  happening
     import SCons


    We import the SCons python module into our script. Normally you don't
have to do this, but I'm using the 'SCons.Util.CLVar' class provided by SCons
so I import the module to gain access to it.


     env = Environment(ENV = os.environ)


   Construct a SCons Environment object. SCons does not copy the system
environment by default, this is by design as a build environment should be as
explicit as possible. For now I'll just copy the system environment, but please
note that this is bad practice.


     env['NVCOMPRESS'] = 'nvcompress'
     env['NVCOMPRESSTYPE'] = SCons.Util.CLVar('-color')
     env['NVCOMPRESSFLAGS'] = SCons.Util.CLVar('-bc3')
     env['NVCOMPRESSCOM'] = '$NVCOMPRESS $NVCOMPRESSFLAGS $SOURCE $TARGET'


    Now begin populating the Environment with data. Environment objects
are class instances that behave like dictionaries. We can add to them through
assignment, and query the contents using normal Python dictionary functions.
    In our case I'm lling four slots with:
   • the name of the executable ('NVCOMPRESS'),
   • the default type ag ('NVCOMPRESSTYPE'),
   • the default compression ag('NVCOMPRESSFLAGS'),

   • a template command line ('NVCOMPRESSCOM').
     SCons uses late binding of variables found within environment strings,
     any sub-string that starts with a '$' character is interpreted at the calling
     site for the current value within the current environment. This acts like a
     controlled dynamic scoping system.
     I'll come back to what this means in practical terms in a moment, but
     for now accept that our default command line evaluates to:    'nvcompress
     -color -bc3 $SOURCE $TARGET'             .


                                       3
NvCompressAction = Action( '$NVCOMPRESSCOM')
     NvCompressBuilder = Builder( action=NvCompressAction,
                                  suffix='.dds')


    These three lines are the core of our SCons extension; an Action object is
created, and the command line string is set. The evaluation of this command
follows the same rules as for environment variables set earlier.
    Then a Builder is constructed using the Action object just created. We also
set the default extension for all targets.


     env['BUILDERS']['NvCompress'] = NvCompressBuilder


    The Environment is then extended with this builder, and a name is given
for calls to it.


     env.NvCompress( 'image', 'image.png')


   The only thing left is to construct a Node in the dependency tree for the
target le.
   SCons is a three stage process:
  1. Create or gather a collection of tools and set the environment,
  2. Create a tree of dependencies with sources and targets,
  3. The dependencies are scanned, then for any out of date data, the action
     is called.
     Each of these steps is discrete, and must be occur in this order. Step
     1 happens during the setup phase of SCons (default tools are scanned
     for and basic environments are constructed) and to a lesser extent within
     the running SConstruct script itself (as I've just shown). Step 2 happens
     within the SConstruct script and continues until the end of the script.
     After control returns to the SCons system the Step 3 begins and the
     dependency tree is scanned and the actions are triggered. Always in this
     order.
     It's for this reason that I say that a dependency node is constructed from
     the last line of the SConstruct. It doesn't actually run 'nvcompress' at
     this point. Only an object, representing the potential to run 'nvcompress',
     is constructed and added to the system. Work is done later by an internal
     SCons class, 'Taskmaster.


                                      4
4 More les, more controls
One of the great things about SCons is that it's embedded with Python. A
consequence of this choice is that you, the tool writer, have access to all of
Pythons functions.
   I've shown you how to add one Builder and access that Builder for a single
target texture. Games are not made from one texture, you're going to want
access to many textures and that means several targets.     Glob
                                                              to the rescue.


      from glob import glob

      for tex in glob( r'./*/*.png'):
          target = tex.replace('.png','.dds')
          env.NvCompress( target, tex)


   This will nd all of the source textures in subdirectories, and add them to
the dependency graph. Any textures that are added are found automatically.
   The same strategy can be applied to other types of data as well. Python
has great libraries for XML, SQL, even direct in memory structure access or
peeking in compressed les. You will not have to drop out of Python for very
many reasons. Check the module documentation4 for details.

5 Other compression types
The previous method of adding dependencies will add each texture into the
pipeline with the same compression options. Normally you will want a range
of compression methods for textures. I want to focus on the clearest method,
changing the values in the Environment.
    I mentioned earlier that the evaluation of strings set for Builders and Actions
acts like dynamic scoping for variables. This feature allows us to change the
functionality of a call by changing values when the dependency node is built.


      env.NvCompress( 'image2', 'image.png', NVCOMPRESSFLAGS='-bc2')


   Which will result in this command:      'nvcompress -color -bc2 image.png
image2.dds'.
    Adding a method to match lename patterns in a database (or text le)
gives us a simple way to control the compression of individual textures.


      # Global texture compression options
  4 http://docs.python.org/modindex.html



                                           5
# format 'glob,opts'
     .envmaps*.png,-bc1 # Does not include cubemaps
     .nmaps*.png,-bc3n

   This simple text le has a line for each le pattern, a comma (',') and
the compression option for the command line. Comments start with a hash
character ('#') and continue to the end of the line. A parser for this format is
easy to write.

     from glob import glob
     from fnmatch import fnmatch

     gCompressionOptions = []

     f = open('texture_options.txt','rt')
     try:
          for line in f:
              line = line.split('#')[0]
              if line != '':
                  (pattern,options) = line.split(',')
                  gCompressionOptions.append( (pattern,options))
     finally:
          f.close()

     for tex in glob( r'.**.png'):
         hasCustomPattern = False
         target = tex.replace('.png','.dds')
         for pat,opt in gCompressionOptions:
             if fnmatch(tex,pat):
                 opt = opt.strip()
                 env.NvCompress( target, tex, NVCOMPRESSFLAGS=opt)
                 hasCustomPattern = True
         if not hasCustomPattern:
             env.NvCompress( target, tex)

   Once we have the patterns into an array it's simple to check if any les found
by the glob matches a given pattern. If there is a match, set the compression
options for that texture. If not the default is used.

6 Exploring dependencies
A core strength of SCons is it's dependency system. This system is not normally
based on time stamps but on a one way hash of the le contents (MD5). Using
hash values allows for stronger connections between assets.

                                       6
In order to follow what SCons is doing with dependency checking you can
use the 'debug=explain'       command line option. This option will print out
information about dependency checks and commands being run.
    The interesting thing about using hash values for dependency check is that
you can't use  touch   to force a recompile of an asset, you must change the
contents of the le or force SCons to rebuild an asset.
    On the ip side of this, you get control of the derived le from both the
contents of the source les and the contents of the command line used to build
the derived le. SCons combines all of these values into a string that represents
the derived target asset, if any sources have changed that targets action is
invoked.
    To get a view of the dependency tree use the  tree=derived      option.



     scons: Reading SConscript files ...
     special .envmapsgunmap.dds .envmapsgunmap.png -bc1
     special .nmapsgunmap.dds .nmapsgunmap.png -bc3n
     normal .texgunmap.dds .texgunmap.png
     scons: done reading SConscript files.
     scons: Building targets ...
     scons: `.' is up to date.
     +-.
       +-envmaps
       | +-envmapsgunmap.dds
       +-nmaps
       | +-nmapsgunmap.dds
       +-tex
         +-texgunmap.dds
     scons: done building targets.




7 Sharing the results
Strong dependency systems are great for personal development, you can be sure
that you only need to build the minimum following a change. This doesn't help
on a large team; if an artist makes a change then once submitted every person
on the team needs to build that same asset. In my case building a compressed
texture takes 9.5 seconds, if you have 20 other people on your team, you team
will spend 190 seconds for each texture change. The odd part of this result is
that every person is trying to get the same result from the same source.
    SCons took the strong hash key info and took it to the next logical level. If
everyone calculates the hash the same way, then it's possible to store copies of
the result in a shared location under that name. Then when you go to build, do


                                       7
a quick check rst. If the asset exists under that name in this cache then just
copy it; if not then build the asset and place it into the cache under it's hash
name.
    The result is a distributed build cache. And you can take advantage of it
out of the box. Create a location with a lot of available disk space, available for
everyone on your team, some server. Next place this line into you SConstruct
le.


      env.CacheDir(r'x:Locationofcachedir')


    Now your derived les are shared, and only one person, normally the origi-
nator of the source art, needs to build the nal result. In most cases, this saves
hours of cumulative time for a large team of people. You results will be better
if you have a continuous integration server for art.

8 The script we have
Taking all of these changes into account we get the following script.


      import SCons

      env = Environment(ENV = os.environ)

      env.CacheDir(r'x:Locationofcachedir')

      env['NVCOMPRESS'] = 'nvcompress'
      env['NVCOMPRESSTYPE'] = SCons.Util.CLVar('-color')
      env['NVCOMPRESSFLAGS'] = SCons.Util.CLVar('-bc3')
      env['NVCOMPRESSCOM'] = '$NVCOMPRESS $NVCOMPRESSFLAGS $SOURCE $TARGET'

      NvCompressAction = Action( '$NVCOMPRESSCOM')
      NvCompressBuilder = Builder( action=NvCompressAction,
                                   suffix='.dds')

      env['BUILDERS']['NvCompress'] = NvCompressBuilder

      from glob import glob
      from fnmatch import fnmatch

      gCompressionOptions = []

      f = open('texture_options.txt','rt')
      try:

                                        8
for line in f:
              line = line.split('#')[0]
              if line != '':
                  (pattern,options) = line.split(',')
                  gCompressionOptions.append( (pattern,options))
     finally:
         f.close()

     for tex in glob( r'.**.png'):
         hasCustomPattern = False
         target = tex.replace('.png','.dds')
         for pat,opt in gCompressionOptions:
             if fnmatch(tex,pat):
                 opt = opt.strip()
                 env.NvCompress( target, tex, NVCOMPRESSFLAGS=opt)
                 hasCustomPattern = True
         if not hasCustomPattern:
             env.NvCompress( target, tex)



    This will build all of your textures found in sub directories, using the com-
pression options found in the  texture_options.txt      le, and share the results
with colleagues using a shared drive.
    I hope you see how easy it is to build a strong build system for art given the
right tools.
    SCons is not just for code, but can be used for art builds as well. Coupled
with a continuous integration server, having a quick, robust pipeline is within
your grasp.




                                        9

More Related Content

PPTX
Version control
Shahriar Iqbal Chowdhury
 
PPTX
Transformaciones modelo a modelo: ATL (ParteII)
Ricardo Tesoriero
 
PDF
Course 102: Lecture 13: Regular Expressions
Ahmed El-Arabawy
 
PPTX
Sizmek ad builder capabilities deck
Leah Daly
 
PDF
Project 2 how to install and compile os161
Xiao Qin
 
PDF
Course 102: Lecture 9: Input Output Internals
Ahmed El-Arabawy
 
PPTX
Process and thread
Bach Khoa Da Nang
 
Version control
Shahriar Iqbal Chowdhury
 
Transformaciones modelo a modelo: ATL (ParteII)
Ricardo Tesoriero
 
Course 102: Lecture 13: Regular Expressions
Ahmed El-Arabawy
 
Sizmek ad builder capabilities deck
Leah Daly
 
Project 2 how to install and compile os161
Xiao Qin
 
Course 102: Lecture 9: Input Output Internals
Ahmed El-Arabawy
 
Process and thread
Bach Khoa Da Nang
 

What's hot (9)

PDF
Course 102: Lecture 20: Networking In Linux (Basic Concepts)
Ahmed El-Arabawy
 
PDF
Fauna Auxiliar.pdf
NelsonFernandes140519
 
PPT
Clear case
Rahul Keshri
 
PDF
Windows引导过程
Mike Wang
 
PDF
Cerita Perjanjian Lama untuk anak-anak
FreeChildrenStories
 
PPTX
Linux architecture
ICI Bucharest - roTLD
 
PDF
Version Control with Git & GitHub
Piet Cordemans
 
PDF
Manual de configuracion de redes
mmri
 
PDF
Pengenalan Git
fajran
 
Course 102: Lecture 20: Networking In Linux (Basic Concepts)
Ahmed El-Arabawy
 
Fauna Auxiliar.pdf
NelsonFernandes140519
 
Clear case
Rahul Keshri
 
Windows引导过程
Mike Wang
 
Cerita Perjanjian Lama untuk anak-anak
FreeChildrenStories
 
Linux architecture
ICI Bucharest - roTLD
 
Version Control with Git & GitHub
Piet Cordemans
 
Manual de configuracion de redes
mmri
 
Pengenalan Git
fajran
 
Ad

Viewers also liked (18)

PDF
Confrontation Pipeline and SCons
slantsixgames
 
PPT
Introduction to s cons
dcshi
 
PPTX
Presentación1
Astrid Valdes
 
PPTX
Técnicas de Comunicación Oral
yesibrandao24
 
PDF
Digital divide
Andrew Campbell
 
PDF
Federal Health Care Spending:Why Is It Growing? What Could Be Done About It?
Congressional Budget Office
 
PPT
DIFICULTADES DEL APRENDIZAJE
Belén Quinaluisa
 
PDF
Time-based Animation Games
University of West Florida
 
PPTX
"Корреляция между эффективными коммуникациями и прибыльностью бизнеса", Натал...
rabota.ua
 
PPTX
How Linkedin uses Gamification to improve its Business!
Alok Rodinhood Kejriwal
 
PPTX
Introduction to dry needling
Subhanjan Das
 
ZIP
Continuous Integration, Build Pipelines and Continuous Deployment
Christopher Read
 
PDF
Living Brands
Young & Rubicam
 
PDF
60 Content Predictions for 2015 by Content Marketing Institute
Content Marketing Institute
 
PDF
11 Inspirational Quotes from the Top CEOs of the Modern Business World
Matt Burke
 
PPT
The Executioner's Tale
Christina Wodtke
 
PDF
Marco Común de Competencia Digital Docente - 2017
INTEF
 
PPTX
The French Revolution of 1789
Tom Richey
 
Confrontation Pipeline and SCons
slantsixgames
 
Introduction to s cons
dcshi
 
Presentación1
Astrid Valdes
 
Técnicas de Comunicación Oral
yesibrandao24
 
Digital divide
Andrew Campbell
 
Federal Health Care Spending:Why Is It Growing? What Could Be Done About It?
Congressional Budget Office
 
DIFICULTADES DEL APRENDIZAJE
Belén Quinaluisa
 
Time-based Animation Games
University of West Florida
 
"Корреляция между эффективными коммуникациями и прибыльностью бизнеса", Натал...
rabota.ua
 
How Linkedin uses Gamification to improve its Business!
Alok Rodinhood Kejriwal
 
Introduction to dry needling
Subhanjan Das
 
Continuous Integration, Build Pipelines and Continuous Deployment
Christopher Read
 
Living Brands
Young & Rubicam
 
60 Content Predictions for 2015 by Content Marketing Institute
Content Marketing Institute
 
11 Inspirational Quotes from the Top CEOs of the Modern Business World
Matt Burke
 
The Executioner's Tale
Christina Wodtke
 
Marco Común de Competencia Digital Docente - 2017
INTEF
 
The French Revolution of 1789
Tom Richey
 
Ad

Similar to SCons an Introduction (20)

PDF
Armitage – The Ultimate Attack Platform for Metasploit
Ishan Girdhar
 
PDF
Viktor Tsykunov: Azure Machine Learning Service
Lviv Startup Club
 
PDF
Alex Smola at AI Frontiers: Scalable Deep Learning Using MXNet
AI Frontiers
 
PDF
Deep Learning for Computer Vision: Software Frameworks (UPC 2016)
Universitat Politècnica de Catalunya
 
PPT
DB Conan 1.0
raka.angga
 
PDF
maxbox starter60 machine learning
Max Kleiner
 
PDF
Power ai tensorflowworkloadtutorial-20171117
Ganesan Narayanasamy
 
PPT
Introduction to NS2 - Cont..
cscarcas
 
PPTX
Tensor flow
Nikhil Krishna Nair
 
PPTX
Tutorial: Image Generation and Image-to-Image Translation using GAN
Wuhyun Rico Shin
 
PPTX
Final training course
Noor Dhiya
 
PDF
An Experiment with Checking the glibc Library
Andrey Karpov
 
PDF
Linux Kernel, tested by the Linux-version of PVS-Studio
PVS-Studio
 
PDF
NativeBoost
ESUG
 
PDF
Cppcheck
PVS-Studio
 
PDF
nlp dl 1.pdf
nyomans1
 
PDF
Build a Complex, Realtime Data Management App with Postgres 14!
Jonathan Katz
 
PPT
Network Simulator Tutorial
cscarcas
 
PDF
maXbox Starter 42 Multiprocessing Programming
Max Kleiner
 
PDF
Keras and TensorFlow
NopphawanTamkuan
 
Armitage – The Ultimate Attack Platform for Metasploit
Ishan Girdhar
 
Viktor Tsykunov: Azure Machine Learning Service
Lviv Startup Club
 
Alex Smola at AI Frontiers: Scalable Deep Learning Using MXNet
AI Frontiers
 
Deep Learning for Computer Vision: Software Frameworks (UPC 2016)
Universitat Politècnica de Catalunya
 
DB Conan 1.0
raka.angga
 
maxbox starter60 machine learning
Max Kleiner
 
Power ai tensorflowworkloadtutorial-20171117
Ganesan Narayanasamy
 
Introduction to NS2 - Cont..
cscarcas
 
Tensor flow
Nikhil Krishna Nair
 
Tutorial: Image Generation and Image-to-Image Translation using GAN
Wuhyun Rico Shin
 
Final training course
Noor Dhiya
 
An Experiment with Checking the glibc Library
Andrey Karpov
 
Linux Kernel, tested by the Linux-version of PVS-Studio
PVS-Studio
 
NativeBoost
ESUG
 
Cppcheck
PVS-Studio
 
nlp dl 1.pdf
nyomans1
 
Build a Complex, Realtime Data Management App with Postgres 14!
Jonathan Katz
 
Network Simulator Tutorial
cscarcas
 
maXbox Starter 42 Multiprocessing Programming
Max Kleiner
 
Keras and TensorFlow
NopphawanTamkuan
 

More from slantsixgames (11)

PPTX
Supersize your production pipe enjmin 2013 v1.1 hd
slantsixgames
 
DOC
Ask the Producers Feb 8th
slantsixgames
 
PPTX
Maximize Your Production Effort (English)
slantsixgames
 
PPTX
Maximize Your Production Effort (Chinese)
slantsixgames
 
PDF
Confrontation Audio GDC 2009
slantsixgames
 
PDF
Audio SPU Presentation
slantsixgames
 
PDF
Collision Detection an Overview
slantsixgames
 
PDF
Modern Graphics Pipeline Overview
slantsixgames
 
PDF
PPU Optimisation Lesson
slantsixgames
 
PDF
Event System Presentation
slantsixgames
 
PDF
Supersize Your Production Pipe
slantsixgames
 
Supersize your production pipe enjmin 2013 v1.1 hd
slantsixgames
 
Ask the Producers Feb 8th
slantsixgames
 
Maximize Your Production Effort (English)
slantsixgames
 
Maximize Your Production Effort (Chinese)
slantsixgames
 
Confrontation Audio GDC 2009
slantsixgames
 
Audio SPU Presentation
slantsixgames
 
Collision Detection an Overview
slantsixgames
 
Modern Graphics Pipeline Overview
slantsixgames
 
PPU Optimisation Lesson
slantsixgames
 
Event System Presentation
slantsixgames
 
Supersize Your Production Pipe
slantsixgames
 

Recently uploaded (20)

PDF
Research-Fundamentals-and-Topic-Development.pdf
ayesha butalia
 
PDF
GDG Cloud Munich - Intro - Luiz Carneiro - #BuildWithAI - July - Abdel.pdf
Luiz Carneiro
 
PDF
Get More from Fiori Automation - What’s New, What Works, and What’s Next.pdf
Precisely
 
PDF
OFFOFFBOX™ – A New Era for African Film | Startup Presentation
ambaicciwalkerbrian
 
PDF
CIFDAQ's Market Wrap : Bears Back in Control?
CIFDAQ
 
PDF
Data_Analytics_vs_Data_Science_vs_BI_by_CA_Suvidha_Chaplot.pdf
CA Suvidha Chaplot
 
PDF
Automating ArcGIS Content Discovery with FME: A Real World Use Case
Safe Software
 
PPTX
cloud computing vai.pptx for the project
vaibhavdobariyal79
 
PDF
Presentation about Hardware and Software in Computer
snehamodhawadiya
 
PDF
How Open Source Changed My Career by abdelrahman ismail
a0m0rajab1
 
PPTX
Applied-Statistics-Mastering-Data-Driven-Decisions.pptx
parmaryashparmaryash
 
PPTX
OA presentation.pptx OA presentation.pptx
pateldhruv002338
 
PDF
Trying to figure out MCP by actually building an app from scratch with open s...
Julien SIMON
 
PDF
Brief History of Internet - Early Days of Internet
sutharharshit158
 
PPTX
The Future of AI & Machine Learning.pptx
pritsen4700
 
PDF
Software Development Methodologies in 2025
KodekX
 
PDF
Peak of Data & AI Encore - Real-Time Insights & Scalable Editing with ArcGIS
Safe Software
 
PDF
Accelerating Oracle Database 23ai Troubleshooting with Oracle AHF Fleet Insig...
Sandesh Rao
 
PDF
AI-Cloud-Business-Management-Platforms-The-Key-to-Efficiency-Growth.pdf
Artjoker Software Development Company
 
PDF
The Future of Mobile Is Context-Aware—Are You Ready?
iProgrammer Solutions Private Limited
 
Research-Fundamentals-and-Topic-Development.pdf
ayesha butalia
 
GDG Cloud Munich - Intro - Luiz Carneiro - #BuildWithAI - July - Abdel.pdf
Luiz Carneiro
 
Get More from Fiori Automation - What’s New, What Works, and What’s Next.pdf
Precisely
 
OFFOFFBOX™ – A New Era for African Film | Startup Presentation
ambaicciwalkerbrian
 
CIFDAQ's Market Wrap : Bears Back in Control?
CIFDAQ
 
Data_Analytics_vs_Data_Science_vs_BI_by_CA_Suvidha_Chaplot.pdf
CA Suvidha Chaplot
 
Automating ArcGIS Content Discovery with FME: A Real World Use Case
Safe Software
 
cloud computing vai.pptx for the project
vaibhavdobariyal79
 
Presentation about Hardware and Software in Computer
snehamodhawadiya
 
How Open Source Changed My Career by abdelrahman ismail
a0m0rajab1
 
Applied-Statistics-Mastering-Data-Driven-Decisions.pptx
parmaryashparmaryash
 
OA presentation.pptx OA presentation.pptx
pateldhruv002338
 
Trying to figure out MCP by actually building an app from scratch with open s...
Julien SIMON
 
Brief History of Internet - Early Days of Internet
sutharharshit158
 
The Future of AI & Machine Learning.pptx
pritsen4700
 
Software Development Methodologies in 2025
KodekX
 
Peak of Data & AI Encore - Real-Time Insights & Scalable Editing with ArcGIS
Safe Software
 
Accelerating Oracle Database 23ai Troubleshooting with Oracle AHF Fleet Insig...
Sandesh Rao
 
AI-Cloud-Business-Management-Platforms-The-Key-to-Efficiency-Growth.pdf
Artjoker Software Development Company
 
The Future of Mobile Is Context-Aware—Are You Ready?
iProgrammer Solutions Private Limited
 

SCons an Introduction

  • 1. SCons: an introduction Dean Giberson <[email protected]> Dec 17, 2008 If you've tried putting a build system together for a modern game you know that you are facing a monumental, never ending task. Once you do get a basic pipeline working, modern games require source assets that are in the range of 100 Gb of raw data. Tracking this amount of data takes time and resources. SCons (http://www.scons.org) is a Python based make replacement; with it you can tame your dependencies and data sizes. What follows is quick art centric introduction. 1 The Canonical Example The SCons documentation1 would give an example like this: env = Environment() env.Program( 'HelloWorld', ['HelloWorld.cpp']) follow this and provide a compilable HelloWorld.cpp and you would have a program ready to go. But how would you extend this to use your own pipeline tools? 2 Adding Compressed Textures Lets assume that you want to compress textures for use in a game. NVidia Textures2 Tools can be used on your TGA/PNG/JPG/DDS images. For exam- ple you want to convert this image3 into a DXT5 compressed image for use on target: nvcompress -color -bc3 gunmap.png gunmap.dds 1 http://www.scons.org/doc/0.97/HTML/scons-user/book1.html 2 http://developer.nvidia.com/object/texture_tools.html 3 Images provide by G3D Engine Data Collection http://g3d-cpp.sourceforge.net/. 1
  • 2. Once we understand how to do this from the command line adding it to SCons is quite easy. import SCons env = Environment(ENV = os.environ) env['NVCOMPRESS'] = 'nvcompress' env['NVCOMPRESSTYPE'] = SCons.Util.CLVar('-color') env['NVCOMPRESSFLAGS'] = SCons.Util.CLVar('-bc3') env['NVCOMPRESSCOM'] = '$NVCOMPRESS $NVCOMPRESSFLAGS $SOURCE $TARGET' NvCompressAction = Action( '$NVCOMPRESSCOM') NvCompressBuilder = Builder( action=NvCompressAction, suffix='.dds') env['BUILDERS']['NvCompress'] = NvCompressBuilder env.NvCompress( 'image', 'image.png') 2
  • 3. 3 A Quick walk-through to understand what is happening import SCons We import the SCons python module into our script. Normally you don't have to do this, but I'm using the 'SCons.Util.CLVar' class provided by SCons so I import the module to gain access to it. env = Environment(ENV = os.environ) Construct a SCons Environment object. SCons does not copy the system environment by default, this is by design as a build environment should be as explicit as possible. For now I'll just copy the system environment, but please note that this is bad practice. env['NVCOMPRESS'] = 'nvcompress' env['NVCOMPRESSTYPE'] = SCons.Util.CLVar('-color') env['NVCOMPRESSFLAGS'] = SCons.Util.CLVar('-bc3') env['NVCOMPRESSCOM'] = '$NVCOMPRESS $NVCOMPRESSFLAGS $SOURCE $TARGET' Now begin populating the Environment with data. Environment objects are class instances that behave like dictionaries. We can add to them through assignment, and query the contents using normal Python dictionary functions. In our case I'm lling four slots with: • the name of the executable ('NVCOMPRESS'), • the default type ag ('NVCOMPRESSTYPE'), • the default compression ag('NVCOMPRESSFLAGS'), • a template command line ('NVCOMPRESSCOM'). SCons uses late binding of variables found within environment strings, any sub-string that starts with a '$' character is interpreted at the calling site for the current value within the current environment. This acts like a controlled dynamic scoping system. I'll come back to what this means in practical terms in a moment, but for now accept that our default command line evaluates to: 'nvcompress -color -bc3 $SOURCE $TARGET' . 3
  • 4. NvCompressAction = Action( '$NVCOMPRESSCOM') NvCompressBuilder = Builder( action=NvCompressAction, suffix='.dds') These three lines are the core of our SCons extension; an Action object is created, and the command line string is set. The evaluation of this command follows the same rules as for environment variables set earlier. Then a Builder is constructed using the Action object just created. We also set the default extension for all targets. env['BUILDERS']['NvCompress'] = NvCompressBuilder The Environment is then extended with this builder, and a name is given for calls to it. env.NvCompress( 'image', 'image.png') The only thing left is to construct a Node in the dependency tree for the target le. SCons is a three stage process: 1. Create or gather a collection of tools and set the environment, 2. Create a tree of dependencies with sources and targets, 3. The dependencies are scanned, then for any out of date data, the action is called. Each of these steps is discrete, and must be occur in this order. Step 1 happens during the setup phase of SCons (default tools are scanned for and basic environments are constructed) and to a lesser extent within the running SConstruct script itself (as I've just shown). Step 2 happens within the SConstruct script and continues until the end of the script. After control returns to the SCons system the Step 3 begins and the dependency tree is scanned and the actions are triggered. Always in this order. It's for this reason that I say that a dependency node is constructed from the last line of the SConstruct. It doesn't actually run 'nvcompress' at this point. Only an object, representing the potential to run 'nvcompress', is constructed and added to the system. Work is done later by an internal SCons class, 'Taskmaster. 4
  • 5. 4 More les, more controls One of the great things about SCons is that it's embedded with Python. A consequence of this choice is that you, the tool writer, have access to all of Pythons functions. I've shown you how to add one Builder and access that Builder for a single target texture. Games are not made from one texture, you're going to want access to many textures and that means several targets. Glob to the rescue. from glob import glob for tex in glob( r'./*/*.png'): target = tex.replace('.png','.dds') env.NvCompress( target, tex) This will nd all of the source textures in subdirectories, and add them to the dependency graph. Any textures that are added are found automatically. The same strategy can be applied to other types of data as well. Python has great libraries for XML, SQL, even direct in memory structure access or peeking in compressed les. You will not have to drop out of Python for very many reasons. Check the module documentation4 for details. 5 Other compression types The previous method of adding dependencies will add each texture into the pipeline with the same compression options. Normally you will want a range of compression methods for textures. I want to focus on the clearest method, changing the values in the Environment. I mentioned earlier that the evaluation of strings set for Builders and Actions acts like dynamic scoping for variables. This feature allows us to change the functionality of a call by changing values when the dependency node is built. env.NvCompress( 'image2', 'image.png', NVCOMPRESSFLAGS='-bc2') Which will result in this command: 'nvcompress -color -bc2 image.png image2.dds'. Adding a method to match lename patterns in a database (or text le) gives us a simple way to control the compression of individual textures. # Global texture compression options 4 http://docs.python.org/modindex.html 5
  • 6. # format 'glob,opts' .envmaps*.png,-bc1 # Does not include cubemaps .nmaps*.png,-bc3n This simple text le has a line for each le pattern, a comma (',') and the compression option for the command line. Comments start with a hash character ('#') and continue to the end of the line. A parser for this format is easy to write. from glob import glob from fnmatch import fnmatch gCompressionOptions = [] f = open('texture_options.txt','rt') try: for line in f: line = line.split('#')[0] if line != '': (pattern,options) = line.split(',') gCompressionOptions.append( (pattern,options)) finally: f.close() for tex in glob( r'.**.png'): hasCustomPattern = False target = tex.replace('.png','.dds') for pat,opt in gCompressionOptions: if fnmatch(tex,pat): opt = opt.strip() env.NvCompress( target, tex, NVCOMPRESSFLAGS=opt) hasCustomPattern = True if not hasCustomPattern: env.NvCompress( target, tex) Once we have the patterns into an array it's simple to check if any les found by the glob matches a given pattern. If there is a match, set the compression options for that texture. If not the default is used. 6 Exploring dependencies A core strength of SCons is it's dependency system. This system is not normally based on time stamps but on a one way hash of the le contents (MD5). Using hash values allows for stronger connections between assets. 6
  • 7. In order to follow what SCons is doing with dependency checking you can use the 'debug=explain' command line option. This option will print out information about dependency checks and commands being run. The interesting thing about using hash values for dependency check is that you can't use touch to force a recompile of an asset, you must change the contents of the le or force SCons to rebuild an asset. On the ip side of this, you get control of the derived le from both the contents of the source les and the contents of the command line used to build the derived le. SCons combines all of these values into a string that represents the derived target asset, if any sources have changed that targets action is invoked. To get a view of the dependency tree use the tree=derived option. scons: Reading SConscript files ... special .envmapsgunmap.dds .envmapsgunmap.png -bc1 special .nmapsgunmap.dds .nmapsgunmap.png -bc3n normal .texgunmap.dds .texgunmap.png scons: done reading SConscript files. scons: Building targets ... scons: `.' is up to date. +-. +-envmaps | +-envmapsgunmap.dds +-nmaps | +-nmapsgunmap.dds +-tex +-texgunmap.dds scons: done building targets. 7 Sharing the results Strong dependency systems are great for personal development, you can be sure that you only need to build the minimum following a change. This doesn't help on a large team; if an artist makes a change then once submitted every person on the team needs to build that same asset. In my case building a compressed texture takes 9.5 seconds, if you have 20 other people on your team, you team will spend 190 seconds for each texture change. The odd part of this result is that every person is trying to get the same result from the same source. SCons took the strong hash key info and took it to the next logical level. If everyone calculates the hash the same way, then it's possible to store copies of the result in a shared location under that name. Then when you go to build, do 7
  • 8. a quick check rst. If the asset exists under that name in this cache then just copy it; if not then build the asset and place it into the cache under it's hash name. The result is a distributed build cache. And you can take advantage of it out of the box. Create a location with a lot of available disk space, available for everyone on your team, some server. Next place this line into you SConstruct le. env.CacheDir(r'x:Locationofcachedir') Now your derived les are shared, and only one person, normally the origi- nator of the source art, needs to build the nal result. In most cases, this saves hours of cumulative time for a large team of people. You results will be better if you have a continuous integration server for art. 8 The script we have Taking all of these changes into account we get the following script. import SCons env = Environment(ENV = os.environ) env.CacheDir(r'x:Locationofcachedir') env['NVCOMPRESS'] = 'nvcompress' env['NVCOMPRESSTYPE'] = SCons.Util.CLVar('-color') env['NVCOMPRESSFLAGS'] = SCons.Util.CLVar('-bc3') env['NVCOMPRESSCOM'] = '$NVCOMPRESS $NVCOMPRESSFLAGS $SOURCE $TARGET' NvCompressAction = Action( '$NVCOMPRESSCOM') NvCompressBuilder = Builder( action=NvCompressAction, suffix='.dds') env['BUILDERS']['NvCompress'] = NvCompressBuilder from glob import glob from fnmatch import fnmatch gCompressionOptions = [] f = open('texture_options.txt','rt') try: 8
  • 9. for line in f: line = line.split('#')[0] if line != '': (pattern,options) = line.split(',') gCompressionOptions.append( (pattern,options)) finally: f.close() for tex in glob( r'.**.png'): hasCustomPattern = False target = tex.replace('.png','.dds') for pat,opt in gCompressionOptions: if fnmatch(tex,pat): opt = opt.strip() env.NvCompress( target, tex, NVCOMPRESSFLAGS=opt) hasCustomPattern = True if not hasCustomPattern: env.NvCompress( target, tex) This will build all of your textures found in sub directories, using the com- pression options found in the texture_options.txt le, and share the results with colleagues using a shared drive. I hope you see how easy it is to build a strong build system for art given the right tools. SCons is not just for code, but can be used for art builds as well. Coupled with a continuous integration server, having a quick, robust pipeline is within your grasp. 9