#!usr/bin/perl

use strict;
use warnings;
use IO::String;
use List::Util 'max';

my $FILENAME4 = "test_data.txt";
open(DATA, $FILENAME4);

#create arrays and hashes to store stuff
my (%data, %all, @keys);
while (<DATA>) {
# avoid \n on last field
    chomp;     
#split the data into chunks
    my @chunks = split(/\s{2,}/, <DATA>); ## make sure you don't split inside annotated brackets
    
#create keys for the chunks
    my $key = shift @chunks;
#store the keys in an array unless they already exist
    push @keys, $key unless exists $data{$key};
    foreach my $chunk (@chunks) {
#return references using hashes
        $data{$key}{$chunk}++;
#add all chunks to the hash '%all'
        $all{$chunk} = 1;
    }
}

#remove new_clusters2.txt if it exists
my $remove2 = "new_clusters2.txt";
if (unlink($remove2) == 1) {
                        print "Existing \"new_clusters2.txt\" file was removed\n";
                          }

#now make a file for the ouput
        my $outputfile = "new_clusters2.txt";
           if (! open(POS, ">>$outputfile") ) {
            print "Cannot open file \"$outputfile\" to write to!!\n\n";
                exit;
        }

#sort the fields/columns keys and save them as an array
#my @fields = sort {$a <=> $b} keys %all; ##<--this sorting didn't work
my @fields = sort {lc($a) cmp lc($b)} keys %all; 

#find the longest entry in an array

my @array2 = ();
foreach my $e (@fields){
                        
                        ####
     my $d = $e;
     $d =~ m/(\S+)\_/; 
     my $prefix = $1;
      #  print  "Prefix: ". $prefix."_"."\n\n"; ## prints the prefices

        push(@array2, $1);
      #  print "* @array2 \n";    ## prints the prefices_
        
        }
                        #### 

my $longest =  max map {length} @array2;

#organise the data
foreach my $key (@keys) {
    while (keys %{$data{$key}}) {
        print  POS $key, " "; 
        foreach my $field (@fields) {                
            if ($data{$key}{$field}){
                printf  POS "%${longest}s ", $field;
                delete $data{$key}{$field} unless --$data{$key}{$field};
            }
            else {
               printf POS "%${longest}s ", "-";
            }
        }
        print POS "\n";
    }
}