#!/usr/bin/perl


use strict;
use constant SUCCESS => 0;
use constant ERROR => -1;
require XML::Simple;
use subs ;
use Data::Dumper;
use File::Basename;
use Benchmark;


my %parameters = getParams();
my $location;
my $file;
my $ext;

if ($parameters{file} ne "")
{
  my @rankings;
  my @final_ranking;

  ($file,$location) = fileparse($parameters{file});
  chop($location = `pwd`) if ($location eq "./");
#  @rankings = Rank_Probes($location,"hybridize_results.xml.zfish",@rankings);
#  @rankings = Rank_Probes($location,"hybridize_results.xml.tetraodon",@rankings);
  @rankings = Rank_Probes($location,"hybridize_results.xml.fugu",@rankings);
  @rankings = Rank_Probes($location,"hybridize_results.xml.chicken",@rankings);
  @rankings = Rank_Probes($location,"hybridize_results.xml.mouse",@rankings);
  
   @rankings = sort {$b->[2] <=> $a->[2]} @rankings;
   my @final_probes;
   my $last = "";
   for my $probe (@rankings)
   {
     if ($probe->[0] ne $last)
     { 
       push @final_ranking, [$probe->[0],$probe->[1],$probe->[2],$probe->[3],$probe->[4],$probe->[5],$probe->[6],$probe->[7]];
       $last = $probe->[0];
     }
   }
 
   print "<?xml version='1.0' encoding='ISO-8859-1'?>\n";
   print "<candidates>\n"; 
   for my $probe (@final_ranking)
   {
      print "  <probe>\n";
      print "    <cat1_score>$probe->[2]</cat1_score>\n";
      print "    <cat1_count>$probe->[3]</cat1_count>\n";
      print "    <cat2_score>$probe->[4]</cat2_score>\n";
      print "    <cat2_count>$probe->[5]</cat2_count>\n";
      print "    <cat3_score>$probe->[6]</cat3_score>\n";
      print "    <cat3_count>$probe->[7]</cat3_count>\n";
      print "    <species>$probe->[1]</species>\n";
      print "    <sequence>$probe->[0]</sequence>\n";
      print "  </probe>\n";
   }
   print "</candidates>\n"; 
}
Process_Target("Target$parameters{target}") if ($parameters{target} > 0);
Process_All() if ($parameters{all} == 1);

#########################################################################
#
# Process_All_Files
#
#########################################################################
sub Process_All
{
  debug_print("Start Processing all targets\n");
  opendir(TARGET,"/home/bcarlson/data");
  foreach my $target( grep /^Target\d$/, readdir(TARGET))
  {
    debug_print("\tCall Target [$target]\n");
    Process_Target($target);
  }
  debug_print("Stop Processing all targets\n");

}

#########################################################################
#
# Process_Target
#
#########################################################################
sub Process_Target
{
  my $target_dir = shift;
  my @rankings;
  $target_dir = "/home/bcarlson/data/$target_dir";
  debug_print("\tProcessing Target [$target_dir]\n");
  
  opendir(TARGET,$target_dir);
  foreach my $file ( grep /^human_.+verbose.probes.xml$/, readdir(TARGET))
  {
    debug_print("\t\tFile [$file]\n");
    @rankings = Rank_Probes($target_dir,$file,@rankings);
  }
   for (my $i=0; $i<$#rankings; $i++)
   {
     print "$rankings[$i][2]\t$rankings[$i][0]\n";
   }

}


#########################################################################
#
# Process_Probe_File
#
#########################################################################
sub Rank_Probes
{

   my @probe_list;
   my $infile = new XML::Simple();
   my $cat1_species_count;
   my $cat2_species_count;
   my $cat3_species_count;
   my $total_species;
   my $cat1_sum_success_rate;
   my $cat2_sum_success_rate;
   my $cat3_sum_success_rate;
   my $cat1_probe_success_rate;
   my $cat2_probe_success_rate;
   my $cat3_probe_success_rate;


   my ($location, $probe_file, @rankings) = (@_);
  # print "Processing: $probe_file  ($location)\n\n";
   debug_print("\t\t\tProcessing [$probe_file] at location [$location]\n"); 
   my $probes = $infile->XMLin("$location/$probe_file");
   for (my $i=0; $i<@{$probes->{hybridization}}; $i++)
   {
     $cat1_species_count=0;
     $cat2_species_count=0;
     $cat3_species_count=0;
     $total_species=0;
     $cat1_sum_success_rate=0;
     $cat2_sum_success_rate=0;
     $cat3_sum_success_rate=0;
     $cat1_probe_success_rate=0;
     $cat2_probe_success_rate=0;
     $cat3_probe_success_rate=0;

     debug_print("\t\t\t\t[".$probes->{hybridization}->[$i]->{probe}->{sequence}."]\n");
     #print "\t[".$probes->{hybridization}->[$i]->{probe}->{sequence}."]\n";
     #print "\t[".$probes->{hybridization}->[$i]->{probe}->{species}."]\n";
     for (my $j=0; $j<@{$probes->{hybridization}->[$i]->{result}}; $j++)
     {
       # print "\t\t[".$probes->{hybridization}->[$i]->{result}->[$j]->{species}."]\n";
       # print "\t\t[".$probes->{hybridization}->[$i]->{result}->[$j]->{success_rate}."]\n";
        $total_species++;
        if ($probes->{hybridization}->[$i]->{result}->[$j]->{success_rate} > 0)
        {

         if (($probes->{hybridization}->[$i]->{result}->[$j]->{species} eq "dog" ) ||
             ($probes->{hybridization}->[$i]->{result}->[$j]->{species} eq "cat" ) || 
             ($probes->{hybridization}->[$i]->{result}->[$j]->{species} eq "cow" ) || 
             ($probes->{hybridization}->[$i]->{result}->[$j]->{species} eq "pig" ))
        {
          $cat1_species_count++;
          $cat1_sum_success_rate += $probes->{hybridization}->[$i]->{result}->[$j]->{success_rate};
        }

         if (($probes->{hybridization}->[$i]->{result}->[$j]->{species} eq "mouse" ) ||
             ($probes->{hybridization}->[$i]->{result}->[$j]->{species} eq "rat" ))
        {
          $cat2_species_count++;
          $cat2_sum_success_rate += $probes->{hybridization}->[$i]->{result}->[$j]->{success_rate};
        }


         if (($probes->{hybridization}->[$i]->{result}->[$j]->{species} eq "zfish" ) ||
             ($probes->{hybridization}->[$i]->{result}->[$j]->{species} eq "tetraodon" ) ||
             ($probes->{hybridization}->[$i]->{result}->[$j]->{species} eq "chicken" ))
        {
          $cat3_species_count++;
          $cat3_sum_success_rate += $probes->{hybridization}->[$i]->{result}->[$j]->{success_rate};
        }
        }
       # print "Total Species [$total_species]\n";
       # print "Species Count [$species_count]\n";
        #print "Success Rate  [$sum_success_rate]\n\n"; 
     }
     #$probe_success_rate = sprintf("%.1f",$sum_success_rate / $species_count);
     $cat1_probe_success_rate = 0;
     $cat2_probe_success_rate = 0;
     $cat3_probe_success_rate = 0;
     if ($cat1_species_count > 0)
     {
       $cat1_probe_success_rate = sprintf("%.1f",$cat1_sum_success_rate / $cat1_species_count);
     }
    
     if ($cat2_species_count > 0)
     {
       $cat2_probe_success_rate = sprintf("%.1f",$cat2_sum_success_rate / $cat2_species_count);
     }

     if ($cat3_species_count > 0)
     {
       $cat3_probe_success_rate = sprintf("%.1f",$cat3_sum_success_rate / $cat3_species_count);
     }

     #print "OVER ALL PROBE RATE [$probe_success_rate] on $species_count species\n\n";
     #print "push $probes->{hybridization}->[$i]->{probe}->{sequence}\n"; 
     push @rankings, [$probes->{hybridization}->[$i]->{probe}->{sequence}, 
                      $probes->{hybridization}->[$i]->{probe}->{species},
                      $cat1_probe_success_rate,
                      $cat1_species_count,
                      $cat2_probe_success_rate,
                      $cat2_species_count,
                      $cat3_probe_success_rate,
                      $cat3_species_count];

   } # End Loop for probe
   #@rankings = sort {$b->[2] <=> $a->[2]} @rankings;
   return @rankings;
}





#use Data::Dumper;
#print Dumper(\%parameters);

sub Calculate_Score
{
   my $target_file;
   my @species = ("mouse","baboon","cat","chicken","chimp","cow","dog","fugu","pig","rat","tetra","zfish");
   my $species_dna = "";

   my ($probe, $target, $source_species) = (@_);
   
   for (my $i=0; $i<=$#species ; $i++)
   {
   }  # End Loop for Species
}


sub debug_print
{
   use vars qw(%parameters);
   if ($parameters{debug} ne "")
   {
     my $msg = shift;
     open(LOGFILE, ">>".$parameters{debug});  
     print LOGFILE $msg;
     close (LOGFILE);
   }
}


#########################################################################
# getParams - gets all options from the command line or keeps defaults
#             dies if required options are not included
#
# uses Getopt::Std and reads @ARGV
# defaults can be modified here
# returns hash of options and values
# 
# THIS LIST IS OUT OF DATE... JUST LOOK BELOW
# pick_from_query - 0 to make overgo from reference sequence 
#                   1 to make from query
# identity_cutoff - minimum proportion identical bases for pip
# ref_offset      - offset from begining of reference sequence at which to
#                   begin looking for pips
# gap_length      - optimal spacing between overgos
# overgo_length   - the length of the overgo probe to design
# overgo_overlap  - the overlap between primers for the probe
# overgo_target_gc- the target GC percentage to go for
# overgo_gc_wiggle- the maximum difference from the overgo_target_gc 
#                   for an overgo
# debug           - write debugging messages to stdout and logfile
# overgo_filename - file to write picked overgos to
# all_overgo_filename - file to write all possible overgos to (defaults
#                       to none)
#########################################################################
sub getParams {
    my %opts;
    my $errstr = "";
    use Getopt::Long;
    use Data::Dumper;
 
    GetOptions(\%opts, 'debug:s','all','target=i','file=s');
    #debug_print Dumper(\%opts);
    if (defined($opts{debug}))
    {
       $opts{debug} = "rank_probes.log" if ($opts{debug} eq "");
    } 

    if ((!defined($opts{target})) && (!defined($opts{all})) && (!defined($opts{file})))
    {
      $errstr .= "ERROR : No probes sources defined.\n";
    }

    if ((defined($opts{target})) && ($opts{target} =~ m/[^1-5]/))
    {
      $errstr .= "Target number invalid, only 1-5\n";
    }

    if ($errstr) {
        print <<"END_USAGE";
Description:
This script will count the number of probes for each species category.

Usage: rank_probes [options] 

Probe Sources:
-file   <filename>  Rank probes from a single probe file.
-target <num>       Rank probes from one target (1-5).
-all                Rank probes from all targets.
-debug  [logfile]   Print debug messages to log file (Default: rank_probes.log)

END_USAGE
        print "\n$errstr" if ($errstr);
        die("\n");
    }
    return (
        all    => $opts{all} || 0,
        file   => $opts{file} || "",
        target => $opts{target} || 0,
        debug  => $opts{debug} || "");
}