Extracts a random subset from a CSV data file while preserving the header (unless you don't want that).

RcsvExtract counts the lines in a given CSV data file, extracts a subset of a chosen size and optionally saves the generated subset to a new file.

Calling RcsvExtract with a path as the only parameter will tell you the line count and quit:

./ExtractRandomSubSet.rb some/file.csv

Note: You may have to call ruby ExtractRandomSubSet.rb some/file.csv instead if RcsvExtract is not executable (chmod +x ExtractRandomSubSet.rb) on your system.

Adding a number as the second parameter will provide you with a random subset of the size you specified and flood your console with it:

./ExtractRandomSubSet.rb some/file.csv 3000

If you rather want to save your subset to a new CSV file, just add a third parameter specifying its name like this:

./ExtractRandomSubSet.rb some/file.csv 3000 my/subset.csv

RcsvExtract expects to find a CSV header in the first line and per default preserves this line in your subset. If your data does not have a header, add the no_header flag to have RcsvExtract ignore the first line in any case.

• no_header parameter set: generate_indices needs to include line 0 instead of ignoring it. Move no_header switch from generate_subset to generate_indices to make it happen.

Copyright (c) 2010 Julian Schrader, http://julianschrader.com

See LICENSE for details.