hosts-file-generator.php

<?php
/**
 * @author Elouan Martinet <[email protected]>
 * @copyright Copyright (c) 2016-2017, Elouan Martinet
 * @license AGPL-3.0+
 *
 * This program is free software: you can redistribute it and/or modify
 * it under the terms of the GNU Affero General Public License as
 * published by the Free Software Foundation, either version 3 of the
 * License, or (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
 * GNU Affero General Public License for more details.
 *
 * You should have received a copy of the GNU Affero General Public License
 * along with this program. If not, see <http://www.gnu.org/licenses/>.
 *
 * @version 0.3.1
 */

$target = '0.0.0.0';
$n = PHP_EOL;

$header = '# Generated on ' . date('r') . $n
	. '127.0.0.1 localhost.localdomain localhost' . $n
	. '::1 localhost.localdomain localhost';

// Warning: the script assumes the format of the following array is correct!
$sources = [ // 0: hosts file (allow whitespaces and comments), 1: one domaine per line (strict), array[start_chars, end_chars]: take the middle (strict)
	['http://winhelp2002.mvps.org/hosts.txt', 0],
	['https://adaway.org/hosts.txt', 0],
	['https://hosts-file.net/ad_servers.txt', 0],
	['https://pgl.yoyo.org/adservers/serverlist.php?hostformat=nohtml&showintro=0&startdate[day]=&startdate[month]=&startdate[year]=&mimetype=plaintext', 1],
	['https://easylist-downloads.adblockplus.org/malwaredomains_full.txt', ['||', '^']],
	['https://easylist-downloads.adblockplus.org/adwarefilters.txt', ['||', '^']],
	['https://raw.githubusercontent.com/Dawsey21/Lists/master/adblock-list.txt', ['||', '^']],
];

$whitelist = [
	'po.st',
];

$blacklist = [
	'coin-hive.com',
	'open.oneplus.net',
	'suplah.xss.ht',
];

function read_file_callback($file, $callable, &$result, $whitelist, $args = null) {
	$line = strtok($file, "\n");
	while($line !== false) {
		$line = trim($line);
		if(strlen($line) > 3) {
			if($args === null)
				$r = call_user_func($callable, $line);
			else
				$r = call_user_func_array($callable, array_merge(array($line), $args));

			if($r && !in_array($r, $whitelist) && !in_array($r, $result))
				$result[] = $r;
		}

		$line = strtok("\n");
	}
}

function filter_host($host) {
	return preg_match('#^([\p{L}0-9\-]+\.)+[\p{L}0-9\-]{2,}$#u', $host) ? $host : false;
}

function parse_hosts_line($line) { // split line by whitespaces and take second value
	if($line[0] === '#')
		return false;

	$e = preg_split('/\s+/', $line, 3);
	if(!$e || count($e) < 2)
		return false;

	return filter_host($e[1]);
}

function parse_middle($line, $begin, $end) {
	if(substr($line, 0, strlen($begin)) !== $begin
		|| substr($line, - strlen($end)) !== $end)
		return false;

	if(strlen($end) === 0)
		return filter_host(substr($line, strlen($begin)));
	else
		return filter_host(substr($line, strlen($begin), - strlen($end)));
}

$hosts = $blacklist;

foreach($sources as $src) {
	$file = file_get_contents($src[0]);
	if(!$file)
		continue;

	switch($src[1]) {
		case 0:
			// hosts file
			read_file_callback($file, 'parse_hosts_line', $hosts, $whitelist);
		break;
		case 1:
			// one domain per line
			read_file_callback($file, 'filter_host', $hosts, $whitelist);
		break;
		default:
			if(is_array($src[1])) {
				// take middle
				read_file_callback($file, 'parse_middle', $hosts, $whitelist, array($src[1][0], $src[1][1]));
			}
	}
}

echo $header . $n . $n;

foreach($hosts as $r) {
	echo $target . ' ' . $r . $n;
}
# vim: noet sw=4 ts=4