#!/usr/bin/perl -w

# Print availability on single line per each url
# Put pub date at beginning of line.
# If url is already shown, do not show again on subsequent lines.

# Example:
# 1995	http://abc.com	1	1	0	1 	...
# 1995	http://sss.edu	0	1	1	0	...

use strict;

# contains date, url, http response
my $fn = "http_responses_per_url.txt";

# contains pub years of urls sorted by years
my $urls_years = "urls_unique_with_pub_year.txt";   

# Build list of dates

my %dates;

open(FILE, $fn) || die("Can't open $fn: $!\n");
my $line = <FILE>;  # ignore first line (header)
while ($line = <FILE>) {
	chomp($line);
	my ($date, $url, $code) = split(/\t/, $line);
	$dates{$date} = 1;
}
close FILE;
	
# Build hash or hashes

my %seen;
my %urls;   # to make sure url is seen only once

open(FILE, $fn) || die("Can't open $fn: $!\n");
$line = <FILE>;  # ignore first line
while ($line = <FILE>) {
        chomp($line);
        my ($date, $url, $code) = split(/\t/, $line);
        $seen{$url}{$date} = $code;
}
close FILE;

# Print list

print "\t";  # for url
foreach my $date (sort keys (%dates)) {
	print "$date\t";
}
print "# times available\n";

open(FILE, $urls_years) || die("Unable to open $urls_years: $!");
while (my $line = <FILE>) {
	chomp($line);
	my ($year, $url) = split(/\t/, $line);

	print "$year\t$url\t";
	
	my $avail = 0;

	foreach my $date (sort keys (%dates)) {
		my $code = $seen{$url}{$date};
		if (defined $code) {
			if ($code eq '200' || $code eq '301' || $code eq '302' || $code eq '304') {
				print "1\t";
				$avail++;
			}
			else {
				print "0\t";
			}
		}
		else {
			print "N/A\t";
		}
	}
	print "$avail\n";
}