#!/usr/bin/perl -w

# print unique urls per article

use strict;

open(URLS, "url_id_list.txt") || die($!);
my @lines = <URLS>;
close URLS;

# Grab all article headings.
# 1.1 article url
# 1.2 article url

my @articles;

foreach (@lines) {
	chomp($_);
	if (/^\d+\.\d+ (.+)$/) {
		push(@articles, $_);
	}
}

my $article_count = @articles;

foreach my $article (@articles) {
	if ($article =~ /^(\d+)\.(\d+) (.+)$/) {
		my $sec1 = $1;
		my $sec2 = $2;

		# print article url
		print "$3\n";
			

		my %seen = ();

		# Now find all urls for this article
		foreach (@lines) {
			if (/^$sec1\.$sec2\.\d+ (.+)$/) {
				my $url = $1;
				if ($url !~ m|^http://www.dlib.org| &&
 				    $url !~ m|^http://dx.doi.org/10.1045/|) {
	
					if (!defined $seen{$url}) {
						$seen{$url} = 1;
						print "\t$url\n";
					}
				}
			}
		}
	}
}