Retrieving values matching the same ID with perl - perl

Question

This is a simple problem but cannot find any working solution for it. I have 2 files and the first file holds all the ID that I am interested in, for example "tomato", "cucumber", but also the ones I am not interested in, which hold no value in the second file. The second file has the following data structure
tomato red
tomato round
tomato sweet
cucumber green
cucumber bitter
cucumber watery
What I need to get is a file containing all the IDs with all the matching values from the second file, everything tab-seperated, like this:
tomato red round sweet
cucumber green bitter watery
What I did so far is create a hash out of the IDs in the first file:
while (<FILE>) {
chomp;
#records = split "\t", $_;
{%hash = map { $records[0] => 1 } #records};
}
And this for the second file:
while (<FILE2>) {
chomp;
#records2 = split "\t", $_;
$key, $value = $records2[0], $records2[1];
$data{$key} = join("\t", $value);
}
close FILE;
foreach my $key ( keys %data )
{
print OUT "$key\t$data{$key}\n"
if exists $hash{$key}
}
Would be grateful for some simple solution for combining all the values matching the same ID! :)

Answer 1

for th first file:
while (<FILE>) {
chomp;
#records = split "\t", $_;
$hash{$records[0]} = 1;
}
and for the second:
while (<FILE2>) {
chomp;
#records2 = split "\t", $_;
($key,$value) = #records2;
$data{$key} = [] unless exists $data{$key};
push #{$data{$key}}, $value;
}
close FILE;
foreach my $key ( keys %data ) {
print OUT $key."\t".join("\t", #{$data{$key}})."\n" if exists $hash{$key};
}

Answer 2

This seems to do what is needed
use strict;
use warnings;
my %data;
open my $fh, '<', 'file1.txt' or die $!;
while (<$fh>) {
$data{$1} = {} if /([^\t]+)/;
}
open $fh, '<', 'file2.txt' or die $!;
while (<$fh>) {
$data{$1}{$2}++ if /^(.+?)\t(.+?)$/ and exists $data{$1};
}
while ( my ($key, $values) = each %data) {
print join("\t", $key, keys %$values), "\n";
}
output
tomato sweet round red
cucumber green watery bitter

Answer 3

It's easier if you read the data mapping first.
Also, if you are using Perl, you should consider from the get-go leveraging one its main strengths - CPAN libraries. For example, the reading in of the file is as simple as read_file() from File::Slurp; instead of having to open/close the file yourself and then run a while(<>) loop.
use File::Slurp;
my %data;
my #data_lines = File::Slurp::read_file($filename2);
chomp(#data_lines);
foreach my $line (#data_lines) { # Improved version from CyberDem0n's answer
my ($key, $value) = split("\t", $line);
$data{$key} ||= []; # Make sure it's an array reference if first time
push #{ $data{$key} }, $value;
}
my #id_lines = File::Slurp::read_file($filename1);
chomp(#id_lines);
foreach my $id (#id_lines) {
print join("\t", ( $id, #{ $data{$id} } ) )."\n";
}
A slightly more hacky but a bit shorter code adds the ID to the list of values in the data hash from the get go:
my #data_lines = File::Slurp::read_file($filename2);
chomp(#data_lines);
foreach my $line (#data_lines) { # Improved version from CyberDem0n's answer
my ($key, $value) = split("\t", $line);
$data{$key} ||= [ $id ]; # Add the ID for printing
push #{ $data{$key} }, $value;
}
my #id_lines = File::Slurp::read_file($filename1);
chomp(#id_lines);
foreach my $id (#id_lines) {
print join("\t", #{ $data{$id} } ) ."\n"; # ID already in %data!
}