#!/usr/bin/perl

# Randal R. Ketchem, Ph.D.                619.784.9879 (voice1)
# Department of Molecular Biology, MB-9   619.784.8754 (voice2)
# The Scripps Research Institute          619.784.9985 (FAX)
# 10550 N. Torrey Pines Road              ketchemr@scripps.edu (email)
# La Jolla, CA 92037-1027                 http://www.scripps.edu/~ketchemr

# 7col.comb file
#   2 ALA MB      7 GLN HN    6.0 # met_cnstr_vols:12
#   2 ALA MB      7 GLN QPG   6.0 # met_cnstr_vols:144

$prgName = `basename $0`;
chop($prgName);
if($#ARGV != 0)
{
   die <<EOHELP;

This program will remove duplicate assignments in 7col.comb files. It keeps
the shorter restraint.

Usage: $prgName [peak file]
EOHELP
}

# Scan through the input file.
while(<>)
{
   next if(/^\s*#/);
   next if(/^\s*$/);

# First grab the atom specifications.
   ($resNum1, $resName1, $atomName1,
    $resNum2, $resName2, $atomName2) = split(' ');

# Check for atom to identical atom.
next if("$resNum1 $resName1 $atomName1" eq "$resNum2 $resName2 $atomName2");

# If residue one is less, it goes first.
   if($resNum1 < $resNum2)
   {
      $line = "$resNum1 $resName1 $atomName1";
      $line .= " $resNum2 $resName2 $atomName2";
   }
# If residue two is less, it goes first.
   elsif($resNum1 > $resNum2)
   {
      $line = "$resNum2 $resName2 $atomName2";
      $line .= " $resNum1 $resName1 $atomName1";
   }
# They must be on the same residue. Sort by atom name.
   else
   {
      $line1 = "$resNum1 $resName1 $atomName1";
      $line2 = "$resNum2 $resName2 $atomName2";
      @lines = sort($line1, $line2);
      $line = join(' ', @lines);
   }

# Each occurance of $line is counted.
   ++$dup{$line};
# And saved in its entirety.
   $save{$line} .= $_;
}

# For every %save entry... Duplicates are a single entry separated by '\n'.
foreach $key (keys %save)
{
# If the entry contains more than one line.
   if($dup{$key} > 1)
   {
# Place the separate lines in an array. Sort the lines by distance so that
# the line with the longer distance is first.
      @lines = sort by_distance split('\n', $save{$key});
# Print the first line.
      print "$lines[0]\n";
# Print the other lines in the duplicate group commented.
      for($count = 1; $count <= $#lines; ++$count)
      { print "#$lines[$count]\n"; }
   }
# If the line is not duplicated, print it out normally.
   else { print "$save{$key}"; }
}

# Sort by distance subroutine.
sub by_distance
{
   @a = split(' ', $a);
   @b = split(' ', $b);
# Column 7 is the distance.
   $a[6] <=> $b[6];
}