[mirrors/Programs.git] / php / crawler / probably-more-old / old / crawler_5.5.phps

#!/usr/bin/php
<?php
//HSp33der Web Crawler 5.5
//<-Harvie 2oo7
/*
 * Description:
 * This script crawls the web and printing found URLs to STDOUT.
 *
 * Installation:
 * - Debian:
 *   # apt-get install php5-cli php5-curl curl
 *   $ chmod +x crawler.php
 *   $ ./crawler.php (or $ php crawler.php)
 * - Windows:
 *   c:/>path/to/php5/php.exe crawler.php
 *   (you can add php.exe to PATH or associate .php files with it)
 *
 * Notes:
 * - You can use EtherApe (or similar sniffer) to get graphical
 * illustration of connecions (spider with your machine in center).
 * - There are few bugs in PHP5,
 * so this crawler may freeze after few minutes,
 * but you can use $use_curlbin option (need some extra files).
 * Check this for more info: http://bugs.php.net/bug.php?id=43098
 *
 * Tips & Tricks:
 * - Make found URLs 100% unique:
 * cat urls.txt | sort -u > uurls.txt && mv -f uurls.txt urls.txt
 * - Show filesize and total urls count:
 * du -h urls.txt && cat urls.txt | wc -l
 * - Open another output pipe:
 * tail -f urls.txt
 * - Use some plugin (filter) to extract emails, etc...
 * tail -f urls.txt | ./example_mail_extractor.php
 *
 * Dedicated to: SooM.cz (This is SooM project)
 * Special thx2:
 * - PHP, cURL (technology)
 * - root.cz & php.net (help)
 * - Linus Torvalds (because I am crawling from Linux)
 */

///SETTINGS//////////////////
//Basic
$seed = 'http://www.secunia.com/'; //Start crawling from this page
$file = 'urls.txt'; //File to save crawled URLs (use sort to make URLs 100% exclusive)
$delete_url_db = false; //Delete file with saved URLs before crawling
$restore = true; //Use last url from $file instead of seed
//Advanced
$max_size = 10000; //How many bytes download from each page?
$max_urls = 30; //Size of URL buffer (prebuffered URLs to crawl, if full, no more URLs will be added to queue)
$buffer_increase = false; //Increase buffer at buffer underrun?
$random_url = false; //Select random urls from page? (This may override URL filter)
$history_max = 1000; //How many last URLs keep in history? (historied URLs will not be crawled again...)
//Curl binary plugin
$use_curlbin = false; //Use interface to cURL executable instead of file_get_contents() (More speed, less crashes)
$curlpath = 'curl'; //Path to cURL executable binary (you can download it from http://curl.haxx.se/)
//System/PHP
$socket_timeout = 1; //How long you want wait for a webserver? (seconds)
$time_limit = 0; //How long run? (seconds; 0==infinite)
$memory_limit = '128M'; //Maximum memory to eat (if exhausted, crawler will stop)
//Debug
$debug = true; //Use debuging mode? (Print errors and statistics)

//Filter
$eregi_url_blacklist = array( //Most important thing when crawling ;D
    '(W3\.org|W3C)', //Hell knows...
    '(shop|xxx|porn|lesbian|hot)', //Commercial sites
    '(google|209.85.135.104|yahoo.com|amazon.com|youtube.com)', //Big sites
    '(seznam.cz|centrum.cz|atlas.cz|zoznam.sk|quick.cz)', //Big local sites
    '.\.(css|ico|gif|jpg|png|bmp|cgi|js|vbs)', //Misc. webpage content
    '.\.(avi|mpg|mpeg|mov|wmv|wm|mp3|ogg|wma)', //Multimedia files
    '.\.(pdf|swf|flv|cfm)', //Other text files
    '.\.(exe|zip|rar|gz|bz|bz2|tar)' //Big and binary files
);


///FUNCTIONS/////////////////
function shell_curl_get($url, $timeout=0, $maxdata=0) { //I made this cURL interface cause file_get_contents() absolutely sux
	$curl = 'curl';
	if(isset($GLOBALS['curlpath'])) $curl = $GLOBALS['curlpath'];
	$args = "--get --location --connect-timeout $timeout --max-time $timeout --compressed --stderr curl.debug";
	$url = escapeshellcmd(trim($url));
	$cmd = "$curl --url \"$url\" $args";
	if($GLOBALS['debug']) echo("-Ex: $cmd\n");
	$fp = popen($cmd, 'r');
	stream_set_timeout($fp, $timeout);
	stream_set_blocking($fp, 0);
	//if($timeout>0) sleep($timeout);
	$data = '';
	while(!feof($fp)) {
		$in = fgets($fp, 100);
		$data .= $in;
		if($maxdata>0 && strlen($data)>=$maxdata) break;
		if($in == '') usleep(1000000/8);
	}
	@fclose($fp);
	return $data;
}

function check_crawl_url($url) { //Use this function to determine if url is wanted
    foreach($GLOBALS['eregi_url_blacklist'] as $black_url) {
        if(eregi($black_url, $url)) return(0);
    }
    if(in_array($url, $GLOBALS['urls'])) return(0);
    $file_url = explode('?', $url, 2); $file_url = $file_url[0];
    if(in_array($file_url, $GLOBALS['history'])) return(0);
    return(1);
}


///CODE//////////////////////
//Init
if($restore && is_file($file)) { //Load last url as seed
	$urls[1]=$seed; //original seed as backup (if no urls found at restored page)
	if($debug) echo("!Restoring: ");
	$rest = fopen($file, 'r');
	while(!feof($rest)) {
		$seed = $new;
		$new = fgets($rest);
	}
	fclose($rest);
	$seed = trim($seed);
	if($debug) echo($seed."\n");
	$urls[0]=$seed;
	$seed=$urls[1];
}

if($delete_url_db) @unlink($file);
ini_set('default_socket_timeout',$socket_timeout);
set_time_limit($time_limit);
ini_set('memory_limit', $memory_limit);

//Loop
if($debug) { $u = 0; $c = 0; $t = time(); }
$history[0] = '';
$fp = fopen($file, 'a+');
while(true) { //MAIN L00P
	if(sizeof($urls) <= 0) {
		$urls='';
		$urls[0]=$seed;
		if($buffer_increase) $max_urls++; //Increase size of buffer by one
		if($debug) echo("!Buffer underrun! !Buffer size is: $max_urls!\n");
	}
	
	//Get URL from buffer, download few bytes and write down URLs to array:
	$url = array_shift($urls);
	if($debug) echo("-Parsing: $url\n");
	if(!$use_curlbin) { $page = @file_get_contents($url, false, null, 0, $max_size); }
	else { $page = shell_curl_get($url, $socket_timeout, $max_size); }
	preg_match_all('(http:\/\/[_a-zA-Z0-9\.\-]+\.[a-zA-Z]{2,4}\/{1}[-_~&=\ ?\.a-z0-9\/]*)',htmlspecialchars_decode($page), $new_urls);
	$new_urls = $new_urls[0];
	
	foreach($new_urls as $new_url) { //Process grabed URLs
		if($debug) $c++;///Debug
		if(check_crawl_url($new_url)) {
			echo($new_url."\n");	
			fwrite($fp, $new_url."\n");
			$file_url = explode('?', $new_url, 2); $file_url = $file_url[0];
			array_push($history, $file_url);
			while(sizeof($history) > $history_max) @array_shift($history);
			if($random_url) { if(sizeof($urls) < $max_urls) array_push($urls, $new_urls[rand(0,sizeof($new_urls)-1)]); }
				else { if(sizeof($urls) < $max_urls) array_push($urls, $new_url); }
			if($debug) { ///Debug Block
				$u++;
				$uspeed = round($u/(time()-$t+1), 2);
				$cspeed = round($c/(time()-$t+1), 2);
				$time = round((time()-$t)/60, 1);
				$buffered = sizeof($urls);
				$historied = sizeof($history);
				$memory = round(memory_get_usage()/1000000, 2);
				echo("+$u URLs; $c Downloaded; UpTime: $time mins; Buffered: $buffered URLs; History: $historied URLs; Speed: $uspeed URLs/s, $cspeed Downloads/s; Memory used: $memory/$memory_limit\n");
			}
		}
	}
	$new_urls = ''; //Memory cleanup
}

##EOF
Commit	Line	Data
84aff5c8 H	1	#!/usr/bin/php
	2	<?php
	3	//HSp33der Web Crawler 5.5
	4	//<-Harvie 2oo7
	5	/*
	6	* Description:
	7	* This script crawls the web and printing found URLs to STDOUT.
	8	*
	9	* Installation:
	10	* - Debian:
	11	* # apt-get install php5-cli php5-curl curl
	12	* $ chmod +x crawler.php
	13	* $ ./crawler.php (or $ php crawler.php)
	14	* - Windows:
	15	* c:/>path/to/php5/php.exe crawler.php
	16	* (you can add php.exe to PATH or associate .php files with it)
	17	*
	18	* Notes:
	19	* - You can use EtherApe (or similar sniffer) to get graphical
	20	* illustration of connecions (spider with your machine in center).
	21	* - There are few bugs in PHP5,
	22	* so this crawler may freeze after few minutes,
	23	* but you can use $use_curlbin option (need some extra files).
	24	* Check this for more info: http://bugs.php.net/bug.php?id=43098
	25	*
	26	* Tips & Tricks:
	27	* - Make found URLs 100% unique:
	28	* cat urls.txt \| sort -u > uurls.txt && mv -f uurls.txt urls.txt
	29	* - Show filesize and total urls count:
	30	* du -h urls.txt && cat urls.txt \| wc -l
	31	* - Open another output pipe:
	32	* tail -f urls.txt
	33	* - Use some plugin (filter) to extract emails, etc...
	34	* tail -f urls.txt \| ./example_mail_extractor.php
	35	*
	36	* Dedicated to: SooM.cz (This is SooM project)
	37	* Special thx2:
	38	* - PHP, cURL (technology)
	39	* - root.cz & php.net (help)
	40	* - Linus Torvalds (because I am crawling from Linux)
	41	*/
	42
	43	///SETTINGS//////////////////
	44	//Basic
	45	$seed = 'http://www.secunia.com/'; //Start crawling from this page
	46	$file = 'urls.txt'; //File to save crawled URLs (use sort to make URLs 100% exclusive)
	47	$delete_url_db = false; //Delete file with saved URLs before crawling
	48	$restore = true; //Use last url from $file instead of seed
	49	//Advanced
	50	$max_size = 10000; //How many bytes download from each page?
	51	$max_urls = 30; //Size of URL buffer (prebuffered URLs to crawl, if full, no more URLs will be added to queue)
	52	$buffer_increase = false; //Increase buffer at buffer underrun?
	53	$random_url = false; //Select random urls from page? (This may override URL filter)
	54	$history_max = 1000; //How many last URLs keep in history? (historied URLs will not be crawled again...)
	55	//Curl binary plugin
	56	$use_curlbin = false; //Use interface to cURL executable instead of file_get_contents() (More speed, less crashes)
	57	$curlpath = 'curl'; //Path to cURL executable binary (you can download it from http://curl.haxx.se/)
	58	//System/PHP
	59	$socket_timeout = 1; //How long you want wait for a webserver? (seconds)
	60	$time_limit = 0; //How long run? (seconds; 0==infinite)
	61	$memory_limit = '128M'; //Maximum memory to eat (if exhausted, crawler will stop)
	62	//Debug
	63	$debug = true; //Use debuging mode? (Print errors and statistics)
	64
65	//Filter
66	$eregi_url_blacklist = array( //Most important thing when crawling ;D
67	'(W3\.org\|W3C)', //Hell knows...
68	'(shop\|xxx\|porn\|lesbian\|hot)', //Commercial sites
69	'(google\|209.85.135.104\|yahoo.com\|amazon.com\|youtube.com)', //Big sites
70	'(seznam.cz\|centrum.cz\|atlas.cz\|zoznam.sk\|quick.cz)', //Big local sites
71	'.\.(css\|ico\|gif\|jpg\|png\|bmp\|cgi\|js\|vbs)', //Misc. webpage content
72	'.\.(avi\|mpg\|mpeg\|mov\|wmv\|wm\|mp3\|ogg\|wma)', //Multimedia files
73	'.\.(pdf\|swf\|flv\|cfm)', //Other text files
74	'.\.(exe\|zip\|rar\|gz\|bz\|bz2\|tar)' //Big and binary files
75	);
76
77
78	///FUNCTIONS/////////////////
79	function shell_curl_get($url, $timeout=0, $maxdata=0) { //I made this cURL interface cause file_get_contents() absolutely sux
80	$curl = 'curl';
81	if(isset($GLOBALS['curlpath'])) $curl = $GLOBALS['curlpath'];
82	$args = "--get --location --connect-timeout $timeout --max-time $timeout --compressed --stderr curl.debug";
83	$url = escapeshellcmd(trim($url));
84	$cmd = "$curl --url \"$url\" $args";
85	if($GLOBALS['debug']) echo("-Ex: $cmd\n");
86	$fp = popen($cmd, 'r');
87	stream_set_timeout($fp, $timeout);
88	stream_set_blocking($fp, 0);
89	//if($timeout>0) sleep($timeout);
90	$data = '';
91	while(!feof($fp)) {
92	$in = fgets($fp, 100);
93	$data .= $in;
94	if($maxdata>0 && strlen($data)>=$maxdata) break;
95	if($in == '') usleep(1000000/8);
96	}
97	@fclose($fp);
98	return $data;
99	}
100
101	function check_crawl_url($url) { //Use this function to determine if url is wanted
102	foreach($GLOBALS['eregi_url_blacklist'] as $black_url) {
103	if(eregi($black_url, $url)) return(0);
104	}
105	if(in_array($url, $GLOBALS['urls'])) return(0);
106	$file_url = explode('?', $url, 2); $file_url = $file_url[0];
107	if(in_array($file_url, $GLOBALS['history'])) return(0);
108	return(1);
109	}
110
111
112	///CODE//////////////////////
113	//Init
114	if($restore && is_file($file)) { //Load last url as seed
115	$urls[1]=$seed; //original seed as backup (if no urls found at restored page)
116	if($debug) echo("!Restoring: ");
117	$rest = fopen($file, 'r');
118	while(!feof($rest)) {
119	$seed = $new;
120	$new = fgets($rest);
121	}
122	fclose($rest);
123	$seed = trim($seed);
124	if($debug) echo($seed."\n");
125	$urls[0]=$seed;
126	$seed=$urls[1];
127	}
128
129	if($delete_url_db) @unlink($file);
130	ini_set('default_socket_timeout',$socket_timeout);
131	set_time_limit($time_limit);
132	ini_set('memory_limit', $memory_limit);
133
134	//Loop
135	if($debug) { $u = 0; $c = 0; $t = time(); }
136	$history[0] = '';
137	$fp = fopen($file, 'a+');
138	while(true) { //MAIN L00P
139	if(sizeof($urls) <= 0) {
140	$urls='';
141	$urls[0]=$seed;
142	if($buffer_increase) $max_urls++; //Increase size of buffer by one
143	if($debug) echo("!Buffer underrun! !Buffer size is: $max_urls!\n");
144	}
145
146	//Get URL from buffer, download few bytes and write down URLs to array:
147	$url = array_shift($urls);
148	if($debug) echo("-Parsing: $url\n");
149	if(!$use_curlbin) { $page = @file_get_contents($url, false, null, 0, $max_size); }
150	else { $page = shell_curl_get($url, $socket_timeout, $max_size); }
151	preg_match_all('(http:\/\/[_a-zA-Z0-9\.\-]+\.[a-zA-Z]{2,4}\/{1}[-_~&=\ ?\.a-z0-9\/]*)',htmlspecialchars_decode($page), $new_urls);
152	$new_urls = $new_urls[0];
153
154	foreach($new_urls as $new_url) { //Process grabed URLs
155	if($debug) $c++;///Debug
156	if(check_crawl_url($new_url)) {
157	echo($new_url."\n");
158	fwrite($fp, $new_url."\n");
159	$file_url = explode('?', $new_url, 2); $file_url = $file_url[0];
160	array_push($history, $file_url);
161	while(sizeof($history) > $history_max) @array_shift($history);
162	if($random_url) { if(sizeof($urls) < $max_urls) array_push($urls, $new_urls[rand(0,sizeof($new_urls)-1)]); }
163	else { if(sizeof($urls) < $max_urls) array_push($urls, $new_url); }
164	if($debug) { ///Debug Block
165	$u++;
166	$uspeed = round($u/(time()-$t+1), 2);
167	$cspeed = round($c/(time()-$t+1), 2);
168	$time = round((time()-$t)/60, 1);
169	$buffered = sizeof($urls);
170	$historied = sizeof($history);
171	$memory = round(memory_get_usage()/1000000, 2);
172	echo("+$u URLs; $c Downloaded; UpTime: $time mins; Buffered: $buffered URLs; History: $historied URLs; Speed: $uspeed URLs/s, $cspeed Downloads/s; Memory used: $memory/$memory_limit\n");
173	}
174	}
175	}
176	$new_urls = ''; //Memory cleanup
177	}
178
179	##EOF