Re: join column 3 for unique values

I'd stash everything into a data structure (HoHoA, ie. hash of hashes of arrays), which makes sure everything gets aggregated properly, then print it out.

use warnings;
use strict;

my %hash;

while (<DATA>){
    chomp;
    my ($key, $val, $colour) = split;
    push @{ $hash{$key}{$val} }, $colour;
}

for my $k (keys %hash){
    for (keys %{ $hash{$k} }){
        my $colours = join ':', @{ $hash{$k}{$_} };
        print "$k $_ $colours\n";
    }
}

__DATA__
A1 abc yellow
B1 xyz green
A2 cde red
A1 abc green
A2 cde yellow
A1 abc blue
[download]

Output:

A1 abc yellow:green:blue
A2 cde red:yellow
B1 xyz green
[download]

Comment on Re: join column 3 for unique values Select or Download Code

Replies are listed 'Best First'.
Re^2: join column 3 for unique values by hello_beginner (Novice) on Sep 05, 2016 at 15:04 UTC
Great, this works. If there is duplicate value for A1, like below Yellow, it reapts. we need i to print only once </> __DATA__ A1 abc yellow A1 abc yellow A2 cde yellow A1 abc blue Output: A1 abc yellow:yellow:blue ---only yellow:blue A2 cde yellow B1 xyz green	[reply]
Re^3: join column 3 for unique values by stevieb (Canon) on Sep 05, 2016 at 15:30 UTC
You'll need to show a little effort here now. Hint: check the array if the colour already exists before pushing it onto itself, perldoc -f grep, or use a hash for the colours as well, if the order doesn't matter. update: Also, please see How do I post a question effectively? for the use of `<code></code>` tags (they are for code and data), and use the Preview button to ensure all of your tags are closed etc.	[reply] [d/l]