#!/usr/bin/perl -w
####################
# Arpy Saunders, 9/27/06
#
# this program extracts information about the number of repeats for a given
# length from the output of the program cent_repeat_finder. The output file is
# in tab-delimited format so it can be easily input into Excel.

use strict;
use warnings;

################ Declare and initialize variables#################

my $filename = 'crf_repeats_out.txt'; # name of input, dropped reads file
my $record;
my @all_reads = '';
my $element;

my $junk1;
my $junk2;
my @junk = '';
my $repeat_length;
my @repeats = '';

# opens the fasta file containing the dropped reads and splits the single scalar
# into an array, where each element is an individual read.  split is done by the
# '>' marker

unless (open(FASTA, $filename)) {
	print "cannot open FASTA file \"$filename\"\n\n";
	exit;
	
}

# set input separator to "//\n" and read in a record to a scalar
$/ = "//\n";

# read the data into the scalar called "$record"
$record = <FASTA>;


@all_reads = split /(read_id:)/, $record;

foreach $element (@all_reads) # takes on one read at a time
	{
	
	if ($element =~ m/read_id/)
		{
		next;
		}
	
	#pull out scalar after sequence and then take out any newlines and spaces
		
	($junk1) = split (/\tsequence:/s, $element);
	
	$junk1=~ s/\n//g;
	$junk1=~ s/\s//g;
	
	# split the resulting element by :; since we are only interested in the last
	# string, the first two are stored in the scalars called crap1 and crap2
	
	(my $crap1, my $crap2, $repeat_length) = split (/:/s, $junk1);
	
	# add the repeat length to a list
	
	push (@repeats, $repeat_length);
	}
	
	# sort the repeats numerically
	
	my @sorted_repeats = sort {$a <=> $b} @repeats;
	
	count_unique (@sorted_repeats);
	
	
	#getting a hash list of numbers and thereby removing duplicates
	
########################################################################
#
#	Subroutines
#
########################################################################

######################
# count_unique
#
# this subroutine was taken from the faq section of a site on the internet.
# It functions to count the unique elements of an array and store them as a hash.
# This subroutine also functions to print the output file in the appropriate tab
# delimited format.

	sub count_unique {
    my @array = @_;
    my %count;
    my $hash;
	my %hash = ();
	my $number_total_reads;
	my $array;
	
	$number_total_reads = $#array - 1;
	
	
    map { $count{$_}++ } @array;

	my $outputfile = 'repeatcount_out.txt';
    unless (open (REPEATCOUNT_OUT, ">$outputfile"))
										{
											print "cannot open file \"$outputfile\" to write to!\n\n";
											exit;
										}
     
     
     
     #print them out:

	print REPEATCOUNT_OUT "length_of_repeat\t#_in_sample\ttotal_repeats=$number_total_reads";
	
    map {print REPEATCOUNT_OUT "$_\t${count{$_}}\n"} sort keys(%count);

      #or just return the hash:

    return %count;

	
	}