Ruby script to **split a csv file** into separate files categorized by a category column.

categorize_csv.rb

#!/usr/bin/env ruby
#
# categorize_csv.rb
# Fiedlschuster, 2013-08-21
# 
# Usage: ruby categorize_csv.rb --column 2 data.csv
#   This splits the data.csv file into separate files, 
#   one for each category found in column 2 (numbered 1,2,3,...),
#   and outputs to data.category_name.csv
#

require 'csv'
require 'optparse'

import_file = ARGV.last
column_to_categorize_by = 1
data_by_category = {}

option_parser = OptionParser.new do |opts|
  opts.banner = "
    Usage: ruby categorize_csv.rb --column 2 data.csv
      This splits the data.csv file into separate files, 
      one for each category found in column 2 (numbered 1,2,3,...),
      and outputs to data.category_name.csv
  "
  opts.on("-c [number]", "--column [number]", Integer, "The column number specifying the column to categorize by. (Starting counting from 1.)") do |column_number|
    column_to_categorize_by = column_number 
  end
end
option_parser.parse!

if not import_file
  print option_parser.help
  exit 0
end

print "import #{import_file}\n"
CSV.foreach import_file, headers: false, col_sep: ";" do |row|
  category = row[column_to_categorize_by - 1]
  data_by_category[category] ||= []
  data_by_category[category] << row
end

data_by_category.each do |category, data|
  export_file = import_file.gsub(".csv", ".#{category}.csv")
  print "export #{export_file}\n"
  CSV.open(export_file, "wb", headers: false, col_sep: ";") do |csv|
    data.each do |row|
      csv << row
    end
  end
end