#!/usr/bin/perl -w

# print out urls based on publication year

use strict;

my $fn = "urls_by_article.txt";

# articles are not indented.  All urls indented under article are the article's urls

my @articles;
my %articles;

open(URLS, $fn) || die("Can't open $fn: $!\n");
my $line = <URLS>;
while ($line) {
	chomp($line);
	#print "line=$line\n";

	if ($line !~ /^\t/) {
		my $article = $line;
		#print "$article\n";

		$line = <URLS>;
		while ($line && $line =~ /^\t(.+)$/) {
			my $url = $1;
			chomp($url);

			# ignore dlib urls
			next if ($url =~ m|^http://www.dlib.org| || $url =~ m|http://dx.doi.org/10.1045/|);

			#print "\t$url\n";

			my ($year) = $article =~ /dlib\/\w+(\d\d)\//;
			if ($year > 4) {
				$year = "19$year";
			}
			else {
				$year = "20$year";
			}
			push(@{ $articles{$year} }, $url);	
			$line = <URLS>;

		}
	}
}

close URLS;

# Print urls per publication year

foreach my $year (sort keys %articles) {
	foreach my $url (@{ $articles{$year} }) {
		print "$year\t$url\n";
	}
}