parse_zips.rb

gistfile1.rb

# original version, took approximatley 27 minutes.

lines = []
File.open('zips-in.txt', 'r+') do |f|
  f.each_line { |l| lines << l unless lines.include? l }
end

File.open('zips-out.csv', 'w+') do |f|
  lines.each do |l|
    f.write "#{l.split(',').join(' - ')},#{l}"
  end
end


# Updated code, also using 1.9.2
require 'benchmark'

Benchmark.bm(1) do |b|
  b.report('test') do
    lines = []
    File.open('zips-in.txt', 'r+') do |f|
      f.each_line { |l| lines << l }
    end

    File.open('zips-out.csv', 'w+') do |f|
      lines.uniq.each do |l|
        f.write "#{l.gsub(',',' - ').gsub(/\n/, '')},#{l}"
      end
    end
  end
end

#        user     system      total        real
# test  1.960000   0.140000   2.100000 (  2.122407)



# jruby 1.6.2
#        user     system      total        real
# test  2.653000   0.000000   2.653000 (  2.395000)

# rubinius 2.0.0pre
#        user     system      total        real
# test  7.756975   0.138513   7.895488 (  8.820250)

After putting this up, I know I can easily replace the split and join with gsub. Do'h