in reply to Find duplicate fields and merging data in a text file

This code should work. It creates a hash, the part number being the key. The value is another hash, containing two keys, num and des, whose values are the quantity and the list of designators.
#!/usr/bin/perl use strict; use warnings; my %parts; while(my $line = <DATA>){ chomp $line; $line =~ s/"//g; my ($part,$num,@des) = split /,/,$line; $parts{$part}{num} += $num; $parts{$part}{des} = {} unless exists $parts{$part}{des}; my %des = %{ $parts{$part}{des} }; @des{@des} = (); %{ $parts{$part}{des} } = %des; } foreach my $part_id (sort keys %parts){ my $part = $parts{$part_id}; my $des_string = join ',', sort keys %{ $part->{des} }; $des_string = '"'.$des_string.'"' if $des_string =~ /,/; print join ',',$part_id,$part->{num},$des_string; print "\n"; } __DATA__ 032-00751-0000,1,R383 032-00794-0000,6,"RP1,RP2,RP3,RP22,RP24,RP26" 032-00795-0000,8,"RP10,RP11,RP12,RP13,RP14,RP15,RP16,RP17" 032-00804-0000,7,"R7,R14,R21,R23,R41,R42,R49" 032-00807-0000,2,"RP18,RP19" 032-00807-0000,4,"RP8,RP9,RP200,RP201" 032-00808-0000,3,"RP21,RP23,RP25" 032-00820-0000,5,"R966,R970,R971,R1041,R1076" 032-00820-0000,1,R3000 032-00893-0000,1,R1164

Update: designators sorted.

Replies are listed 'Best First'.
Re^2: Find duplicate fields and merging data in a text file
by jwkrahn (Abbot) on Apr 30, 2010 at 02:31 UTC
    $parts{$part}{des} = {} unless exists $parts{$part}{des}; my %des = %{ $parts{$part}{des} }; @des{@des} = (); %{ $parts{$part}{des} } = %des;

    Or simply:

    @{ $parts{ $part }{ des } }{ @des } = ();