#!/usr/bin/perl


use strict;
require XML::Simple;
use subs ;
use Data::Dumper;
use File::Basename;
use Benchmark;

# This script will take a XML file of probes, and look up the location in the human
# the location of the probes.

my %opt = getParams();
my $in1= new XML::Simple();
my $all_probes = $in1->XMLin($opt{probe_file}, keyattr => {});
my $filecount = 1;
open (OUTPUT, ">".$opt{output_file} . "." . $filecount);
#print Dumper(\$all_probes);
my $count = 0;
for (my $i=0; $i<@{$all_probes->{probe}}; $i++)
{
  if ($count >= $opt{max_probes})
  {
    close (OUTPUT);
    $filecount++;
    $count = 0;
    open (OUTPUT, ">".$opt{output_file} . "." . $filecount);
  }
 my $def_line = "";
 my $idx = $opt{species_idx};

 $def_line .= $all_probes->{probe}->[$i]->{name} if (ref($all_probes->{probe}->[$i]->{name}) ne "HASH");
 $def_line .= $all_probes->{probe}->[$i]->{species}->[$idx]->{name} if (ref($all_probes->{probe}->[$i]->{species}->[$idx]->{name}) ne "HASH");
 $def_line .= "," . $all_probes->{probe}->[$i]->{species}->[$idx]->{chromosome} if (ref($all_probes->{probe}->[$i]->{species}->[$idx]->{chromosome}) ne "HASH");
 $def_line .= "," . $all_probes->{probe}->[$i]->{species}->[$idx]->{probe_start} if (ref($all_probes->{probe}->[$i]->{species}->[$idx]->{probe_start}) ne "HASH");
 $def_line .= "," . $all_probes->{probe}->[$i]->{species}->[$idx]->{probe_end} if (ref($all_probes->{probe}->[$i]->{species}->[$idx]->{probe_end}) ne "HASH");
 print OUTPUT ">$def_line \n";
 print OUTPUT $all_probes->{probe}->[$i]->{species}->[$idx]->{sequence} . "\n\n";
 $count++;
}


#########################################################################
# getParams - gets all options from the command line or keeps defaults
#             dies if required options are not included
#########################################################################
sub getParams {
    my %opts;
    my $errstr = "";
    use Getopt::Long;
    use Data::Dumper;
 
    GetOptions(\%opts, 'file=s','output=s', 'count=i', 'species_idx=i');
    if (!defined($opts{file}))
    {
      $errstr .= "ERROR : No probes sources defined.\n";
    }
    if (!defined($opts{file}))
    {
      $errstr .= "ERROR : No output file specified.\n";
    }

    if ($errstr) {
        print <<"END_USAGE";
Description:

This script will take a XML probe file, and create a FASTA formated
file that is ready for megablast.  The scripts outputs several files
with the same basename, each file containing N probes (default 3000)

Usage: create_megablast_file.pl [-count <number>] -file <input_file> -output <output_file> -species_idx <id>

Probe Sources:
-file    <filename>  Probe file to split in to megablast ready batches.
-output  <filename>  Basename for the output file  (extension will be .1-n)
-count   <number>    Number of probes per file
-species_idx <idx>       

END_USAGE
        print "\n$errstr" if ($errstr);
        die("\n");
    }
    return (
            probe_file => $opts{file},
            output_file => $opts{output},
            max_probes => $opts{count} || 99999999,
            species_idx => $opts{species_idx} || 0
           );
}
##############################################################################
# Start Description
# From one XML probe file prepare one or more FASTA format files ready for megablast.  Each FASTA file will contain maximum of N probes (default 3000).
# End Description
##############################################################################