#!/acrm/usr/local/bin/perl 
#*************************************************************************
#
#   Program:    xml2flat
#   File:       
#   
#   Version:    V1.0
#   Date:       15.12.05
#   Function:   Dump the PDBSPROTEC XML file in flat file format
#   
#   Copyright:  (c) UCL / Dr. Andrew C. R. Martin 2005
#   Author:     Dr. Andrew C. R. Martin
#   Address:    Biomolecular Structure & Modelling Unit,
#               Department of Biochemistry & Molecular Biology,
#               University College,
#               Gower Street,
#               London.
#               WC1E 6BT.
#   Phone:      +44 (0)207 679 7034
#   EMail:      andrew@bioinf.org.uk
#               martin@biochem.ucl.ac.uk
#   Web:        http://www.bioinf.org.uk/
#               
#               
#*************************************************************************
#
#   This program is not in the public domain, but it may be copied
#   according to the conditions laid out in the accompanying file
#   COPYING.DOC
#
#   The code may be modified as required, but any modifications must be
#   documented so that the person responsible can be identified. If 
#   someone else breaks this code, I don't want to be blamed for code 
#   that does not work! 
#
#   The code may not be sold commercially or included as part of a 
#   commercial product except as described in the file COPYING.DOC.
#
#*************************************************************************
#
#   Description:
#   ============
#   This is really just a little demo program to allow you to convert
#   the XML format to the flat file format. It shows you how to use
#   Perl/DOM to parse the data. Warning, since it's DOM, everything
#   gets loaded into memory, so 1Gig of RAM is a good idea...
#
#*************************************************************************
#
#   Usage:
#   ======
#
#*************************************************************************
#
#   Revision History:
#   =================
#
#*************************************************************************
use strict;
use XML::DOM;
my($file, $parser, $doc, $pdbse_tag, $rowcount);

$file = shift @ARGV;   # Get the filename from the command line
$parser = XML::DOM::Parser->new();
$doc = $parser->parsefile($file);

print " pdbcode | chainid | res1 | res2 | sprot  |     ec\n";
print "---------+---------+------+------+--------+------------\n";
$rowcount = 0;

foreach $pdbse_tag ($doc->getElementsByTagName('pdb_sprot_ec'))
{
    my($pdb, $chain);
    my($chain_tag, $region_tag);

    $pdb = $pdbse_tag->getAttribute('pdb');
    foreach $chain_tag ($pdbse_tag->getElementsByTagName('chain'))
    {
        $chain = $chain_tag->getAttribute('id');
        foreach $region_tag ($chain_tag->getElementsByTagName('region'))
        {
            my ($ec, $res1, $res2, $sprot);
            my ($ec_tag);
            
            $res1  = $region_tag->getAttribute('res1');
            $res2  = $region_tag->getAttribute('res2');
            $sprot = $region_tag->getAttribute('sprot');

            foreach $ec_tag ($region_tag->getElementsByTagName('ec'))
            {
                my $ec1 = $ec_tag->getAttribute('ec1');
                my $ec2 = $ec_tag->getAttribute('ec2');
                my $ec3 = $ec_tag->getAttribute('ec3');
                my $ec4 = $ec_tag->getAttribute('ec4');
                $ec = "$ec1.$ec2.$ec3.$ec4";
                printf " %4s    | %s       | %-4s | %-4s | %s | %s\n",
                       $pdb, $chain, $res1, $res2, $sprot, $ec;
                $rowcount++;
            }
        }
    }
}
print "($rowcount rows)\n\n";