Sstrhsrtj

I have a csv file and like to split the file into smaller files based on column matching in the file using perl. I am working on Linux Rhel6.

example:

fruit1, fruit2,pricerate,quantity

orange, apple, 3,9

apple,lemon,8,1

orange, apple,3,8

pineapple,papaya,9,19

orange,apple,3,7

pineapple,papaya,9,10

Output is something like:

file1:

fruit1,fruit2,pricerate,quantity

orange,apple, 3,9

orange,apple,3,8

orange,apple,3,7

file2:

fruit1,fruit2,pricerate,quantity

pineapple,papaya,9,19

pineapple,papaya,9,10

the unmatched ones goes into a seperate file. Say file3.

One of the ways in which you can solve this is:

• Open the input file


• Store the first line of the input file (the header)


• 
  

  For every line in the input file after the header:

  
  
  
  
  
  • Read the first two columns
  
  
  • If we haven't opened an output file for the fields you want to match on yet, open a new output file and store its file handle in a hash. Write the header line to the new output file too.
  
  
  • Fetch the handle of the output file in which we should store this line from the file handle hash. Write the line to that file.
  
  
  
  

Here is some example code, which will match on the first two fields:

#!/usr/bin/perl



use strict;

use warnings;



my %filehandles=();

my $filenum=1;



open INPUT, "fruit.csv" 

    or die "Cannot open input file.";



my $header = <INPUT>;



while ( <INPUT> )

{   # Remove spaces from input 

    $_ =~ s/ //g;



    my @fields = split ',', $_;



    if ( ! $filehandles{$fields[0]}{$fields[1]} )

    {   open $filehandles{$fields[0]}{$fields[1]} , ">file$filenum"

            or die "Cannot open output file file$filenum.";

        print {$filehandles{$fields[0]}{$fields[1]}} $header;

        $filenum++;

    }

    print {$filehandles{$fields[0]}{$fields[1]}} $_;

}