use strict;
use warnings;

my $start_time = time;
my $input1 = shift @ARGV;
my $input2 = shift @ARGV;

my %dstrbtn_hash;
open(IN1, '<', $input1) or die "Can't read source file $input1 : $!\n";
while(<IN1>){
    chomp;
    $dstrbtn_hash{ length($_) }++ unless />/;
}
close IN1;

open(IN2, '<', $input2) or die "Can't read source file $input2 : $!\n";
my @source;
while(<IN2>){
    # ignore header
    my $seq = <IN2>;
    chomp $seq;
    my $len = length( $seq );
    push @source, [ $seq, $len ]; # keep length alongside with header
}
close IN2;

my $filename = $input1.$input2."_extracted_seqs.fasta";
open (OUT, '>', $filename) or die "Can't write to file $filename : $!\n";

my $header_count = 1;
while( my ($key, $freq) = each %dstrbtn_hash) {
    my $size = $key - 3; 
    for my $iteration (1..$freq) {
        my $temp_source_seq;
        my @cand;
      EXTRACT: 
        {
            @cand = map { $_->[0] } grep { $_->[1] >= $size } @source; # filter sequences long enough
            die "No long enough sequence found in $input2\n" unless @cand;
            my $chosen=int(rand(@cand));
            $temp_source_seq = $cand[$chosen];
        }
      START:
        {
            my $temp_source_seq_len = length ($temp_source_seq);
            my $random_start_coord = int(rand($temp_source_seq_len-$size)); # substract $size to avoid loop
            my $extracted_seq = substr($temp_source_seq, $random_start_coord, $size);
            print OUT ">".$header_count."extracted_seq", "\n";
            print OUT $extracted_seq, "\n";
            $header_count++;
        }
    }
}

close OUT;
my $end_time = time;
my $duration = ($end_time - $start_time)/60;
print "Thank you for your patience, this Perl script operation has completed, it took $duration minutes, good bye!", " \n";