From: William James <w_a_x_man@...>
Date: 2006-12-19T09:50:04+09:00
Subject: Re: Output unique values in CSV columns to a text file

Drew Olson wrote:
> What I want to do is read in a CSV file and produce an output which
> lists the unique values from each column in the following format:
>
> Column: ColHeader1
>    UniqueVal1
>    UniqueVal2
> Column: ColHeader2
>    UniqueVal1
>    UniqueVal2
> ..
>
> What I'm currently getting is output that looks as follows:
>
> Column: ColHeader1
>
>    ColHeader1UniqueVal1
>    ColHeader1UniqueVal2
> Column: ColHeader2
>
>    ColHeader2UniqueVal1
>    ColHeader2UniqueVal2
> ..
>
> For some reason, it is appending the column header to each value and
> also printing a blank row to start each column. My code is below. Any
> help is much appreciated. Essentially I read the CSV into a hash where
> the key is the column header and the element is an array of values from
> that column. I then run .uniq! on each array in the hash and print the
> results to a file.
>
> require 'rubygems'
> require 'faster_csv'
>
> infile = "xyz.csv"
>
> uniques = {}
>
> FCSV.open(infile, :headers => true).each do |row|
>   row.each_with_index do |element,j|
>     uniques[row.headers[j]] ||= []
>     uniques[row.header[j]] << element
>   end
> end
>
> uniques.each do |key,element|
>   element.uniq!
> end
>
> File.open("unique_output.txt","w+") do |out|
>   uniques.each_key do |key|
>     out.write "Column: #{key}\n"
>     uniques[key].each do |element|
>       out.write "   #{element}\n"
>     end
>   end
> end
>
> --
> Posted via http://www.ruby-forum.com/.

data = DATA.readlines.map{|s| s.chomp.split(",")}
header = data.shift.map{|s| "Column: " + s}

data = data.transpose.map{|ary| ary.uniq.map{|s| "   " + s} }

puts header.zip(data)


__END__
It's,so,simple!
a,b,c
a,b,c
d,e,f