#!/usr/bin/perl
use strict;
use warnings;

use Benchmark;

use Data::Dumper;

my $DNAlength= 6_500_000;
my $sequencelength= 3;
my $testfilename= 'DNA.tst';
my $bufsize= 1024;

# create testdata
open(my $testfile, ">$testfilename") or die "$testfilename: $!";
my $buffer = '';
for (my $i=$DNAlength; $i; --$i) {
        $buffer.= qw(A C G T)[rand 4];
        if (length($buffer) > $bufsize) {
                print $testfile $buffer;
                $buffer = '';
        }
}
print $testfile $buffer;
close $testfile;

a0: {
        my $a0_0=new Benchmark;
        open($testfile, $testfilename) or die "$testfilename: $!";
        my $allbuffered= <$testfile>;
        close $testfile;
        my %count;
        my $template = ('AXA2X2A3X2' x (length($allbuffered) - 2)) . '
+AXA2XA';
        $count{$_}++ foreach unpack $template, $allbuffered;
        my $a0_n=new Benchmark;

        print "the unpack code took:",timestr(timediff($a0_n, $a0_0)),
+"\n";
}


a1: {
        # Start collecting data
        my %DNA_sequence_count;

        my $a1_0=new Benchmark;
        open($testfile, $testfilename) or die "$testfilename: $!";
        my $kept=0;
        while (1) {
                my $read_bytes = read($testfile, $buffer, $bufsize-$ke
+pt);
                last unless $read_bytes;
                my $end_buffer = $read_bytes-$sequencelength;
                for (my $i=0; $i <= $end_buffer; ++$i) {
                        ++$DNA_sequence_count{substr($buffer, $i, $seq
+uencelength)};
                }
                $kept= $sequencelength-1;
                $buffer= substr($buffer, $end_buffer+1, $kept);
        }
        close $testfile;
        my $a1_n=new Benchmark;

        print "the file code took:",timestr(timediff($a1_n, $a1_0)),"\
+n";
}
[download]