class.dc.trackback.php @ 3521:b8a16a351d6d

Visit:

Revision 3521:b8a16a351d6d, 21.7 KB checked in by franck <carnet.franck.paul@…>, 9 years ago (diff)

backlinks:

retrieving ping URLs, let trackback first, then pingback, then finally webmention
get source post content to compose webmention excerpt and retrieve title
use source post title as blog name if this one is unknown (Anonymous blog is used if neither title nor blog name are known)

Rev	Line
[0]	1	<?php
	2	# -- BEGIN LICENSE BLOCK ---------------------------------------
	3	#
	4	# This file is part of Dotclear 2.
	5	#
[1179]	6	# Copyright (c) 2003-2013 Olivier Meunier & Association Dotclear
[0]	7	# Licensed under the GPL version 2.0 license.
	8	# See LICENSE file or
	9	# http://www.gnu.org/licenses/old-licenses/gpl-2.0.html
	10	#
	11	# -- END LICENSE BLOCK -----------------------------------------
	12	if (!defined('DC_RC_PATH')) { return; }
	13
	14	/**
	15	@ingroup DC_CORE
[1684]	16	@brief Trackbacks/Pingbacks sender and server
[0]	17
[1684]	18	Sends and receives trackbacks/pingbacks. Also handles trackbacks/pingbacks auto discovery.
[0]	19	*/
	20	class dcTrackback
	21	{
	22	public $core; ///< <b>dcCore</b> dcCore instance
	23	public $table; ///< <b>string</b> done pings table name
[2566]	24
[0]	25	/**
	26	Object constructor
[2566]	27
[0]	28	@param core <b>dcCore</b> dcCore instance
	29	*/
	30	public function __construct($core)
	31	{
	32	$this->core =& $core;
	33	$this->con =& $this->core->con;
	34	$this->table = $this->core->prefix.'ping';
	35	}
[2566]	36
[3415]	37	/// @name Send
[0]	38	//@{
	39	/**
	40	Get all pings sent for a given post.
[2566]	41
[0]	42	@param post_id <b>integer</b> Post ID
	43	@return <b>record</b>
	44	*/
	45	public function getPostPings($post_id)
	46	{
	47	$strReq = 'SELECT ping_url, ping_dt '.
	48	'FROM '.$this->table.' '.
	49	'WHERE post_id = '.(integer) $post_id;
[2566]	50
[0]	51	return $this->con->select($strReq);
	52	}
[2566]	53
[0]	54	/**
	55	Sends a ping to given <var>$url</var>.
[2566]	56
[0]	57	@param url <b>string</b> URL to ping
	58	@param post_id <b>integer</b> Post ID
	59	@param post_title <b>string</b> Post title
	60	@param post_excerpt <b>string</b> Post excerpt
	61	@param post_url <b>string</b> Post URL
	62	*/
	63	public function ping($url,$post_id,$post_title,$post_excerpt,$post_url)
	64	{
	65	if ($this->core->blog === null) {
	66	return false;
	67	}
[2566]	68
[0]	69	$post_id = (integer) $post_id;
[2566]	70
[0]	71	# Check for previously done trackback
	72	$strReq = 'SELECT post_id, ping_url FROM '.$this->table.' '.
	73	'WHERE post_id = '.$post_id.' '.
	74	"AND ping_url = '".$this->con->escape($url)."' ";
[2566]	75
[0]	76	$rs = $this->con->select($strReq);
[2566]	77
[0]	78	if (!$rs->isEmpty()) {
	79	throw new Exception(sprintf(__('%s has still been pinged'),$url));
	80	}
[2566]	81
[1671]	82	$ping_parts = explode('\|',$url);
[3415]	83	# Maybe a webmention
	84	if (count($ping_parts) == 3) {
	85	$payload = http_build_query(array(
	86	'source' => $post_url,
	87	'target' => $ping_parts[1]
	88	));
[2566]	89
[3415]	90	try {
	91	$http = self::initHttp($ping_parts[0],$path);
	92	$http->setMoreHeader('Content-Type: application/x-www-form-urlencoded');
	93	$http->post($path,$payload,'UTF-8');
	94
	95	# Read response status
	96	$status = $http->getStatus();
	97	$ping_error = '0';
	98	}
	99	catch (Exception $e) {
	100	throw new Exception(__('Unable to ping URL'));
	101	}
	102
	103	if (!in_array($status,array('200','201','202'))) {
	104	$ping_error = $http->getStatus();
	105	$ping_msg = __('Bad server response code');
	106	}
	107	}
	108	# No, let's walk by the trackback way
	109	elseif (count($ping_parts) < 2) {
[1671]	110	$data = array(
	111	'title' => $post_title,
	112	'excerpt' => $post_excerpt,
	113	'url' => $post_url,
	114	'blog_name' => trim(html::escapeHTML(html::clean($this->core->blog->name)))
	115	//,'__debug' => false
	116	);
[2566]	117
[1671]	118	# Ping
[3415]	119	try {
[1671]	120	$http = self::initHttp($url,$path);
	121	$http->post($path,$data,'UTF-8');
	122	$res = $http->getContent();
	123	}
[3415]	124	catch (Exception $e) {
[1671]	125	throw new Exception(__('Unable to ping URL'));
	126	}
[2566]	127
[1671]	128	$pattern =
	129	'\|<response>.<error>(.)</error>(.*)'.
	130	'(<message>(.)</message>(.))?'.
	131	'</response>\|msU';
[2566]	132
[3415]	133	if (!preg_match($pattern,$res,$match)) {
[1671]	134	throw new Exception(sprintf(__('%s is not a ping URL'),$url));
	135	}
[2566]	136
[1671]	137	$ping_error = trim($match[1]);
	138	$ping_msg = (!empty($match[4])) ? $match[4] : '';
[0]	139	}
[1671]	140	# Damnit ! Let's play pingback
	141	else {
	142	try {
	143	$xmlrpc = new xmlrpcClient($ping_parts[0]);
	144	$res = $xmlrpc->query('pingback.ping', $post_url, $ping_parts[1]);
	145	$ping_error = '0';
	146	}
	147	catch (xmlrpcException $e) {
	148	$ping_error = $e->getCode();
[2566]	149	$ping_msg = $e->getMessage();
[1671]	150	}
	151	catch (Exception $e) {
	152	throw new Exception(__('Unable to ping URL'));
	153	}
[0]	154	}
[2566]	155
[0]	156	if ($ping_error != '0') {
	157	throw new Exception(sprintf(__('%s, ping error:'),$url).' '.$ping_msg);
[3415]	158	}
	159	else {
[0]	160	# Notify ping result in database
	161	$cur = $this->con->openCursor($this->table);
	162	$cur->post_id = $post_id;
	163	$cur->ping_url = $url;
	164	$cur->ping_dt = date('Y-m-d H:i:s');
[2566]	165
[0]	166	$cur->insert();
	167	}
	168	}
	169	//@}
[2566]	170
[3415]	171	/// @name Receive
	172	//@{
	173	/**
	174	Receives a trackback and insert it as a comment of given post.
	175
	176	@param post_id <b>integer</b> Post ID
	177	*/
	178	public function receiveTrackback($post_id)
	179	{
	180	header('Content-Type: text/xml; charset=UTF-8');
	181	if (empty($_POST)) {
	182	http::head(405,'Method Not Allowed');
	183	echo
	184	'<?xml version="1.0" encoding="utf-8"?>'."\n".
	185	"<response>\n".
	186	" <error>1</error>\n".
	187	" <message>POST request needed</message>\n".
	188	"</response>";
	189	return;
	190	}
	191
	192	$post_id = (integer) $post_id;
	193
	194	$title = !empty($_POST['title']) ? $_POST['title'] : '';
	195	$excerpt = !empty($_POST['excerpt']) ? $_POST['excerpt'] : '';
	196	$url = !empty($_POST['url']) ? $_POST['url'] : '';
	197	$blog_name = !empty($_POST['blog_name']) ? $_POST['blog_name'] : '';
	198	$charset = '';
	199	$comment = '';
	200
	201	$err = false;
	202	$msg = '';
	203
	204	if ($this->core->blog === null) {
	205	$err = true;
	206	$msg = 'No blog.';
	207	}
	208	elseif ($url == '') {
	209	$err = true;
	210	$msg = 'URL parameter is required.';
	211	}
	212	elseif ($blog_name == '') {
	213	$err = true;
	214	$msg = 'Blog name is required.';
	215	}
	216
	217	if (!$err) {
	218	$post = $this->core->blog->getPosts(array('post_id'=>$post_id,'post_type'=>''));
	219
	220	if ($post->isEmpty()) {
	221	$err = true;
	222	$msg = 'No such post.';
	223	}
	224	elseif (!$post->trackbacksActive()) {
	225	$err = true;
	226	$msg = 'Trackbacks are not allowed for this post or weblog.';
	227	}
	228
	229	$url = trim(html::clean($url));
	230	if ($this->pingAlreadyDone($post->post_id,$url)) {
	231	$err = true;
	232	$msg = 'The trackback has already been registered';
	233	}
	234	}
	235
	236	if (!$err) {
	237	$charset = self::getCharsetFromRequest();
	238
	239	if (!$charset) {
	240	$charset = self::detectCharset($title.' '.$excerpt.' '.$blog_name);
	241	}
	242
	243	if (strtolower($charset) != 'utf-8') {
	244	$title = iconv($charset,'UTF-8',$title);
	245	$excerpt = iconv($charset,'UTF-8',$excerpt);
	246	$blog_name = iconv($charset,'UTF-8',$blog_name);
	247	}
	248
	249	$title = trim(html::clean($title));
	250	$title = html::decodeEntities($title);
	251	$title = html::escapeHTML($title);
	252	$title = text::cutString($title,60);
	253
	254	$excerpt = trim(html::clean($excerpt));
	255	$excerpt = html::decodeEntities($excerpt);
	256	$excerpt = preg_replace('/\s+/ms',' ',$excerpt);
	257	$excerpt = text::cutString($excerpt,252);
	258	$excerpt = html::escapeHTML($excerpt).'...';
	259
	260	$blog_name = trim(html::clean($blog_name));
	261	$blog_name = html::decodeEntities($blog_name);
	262	$blog_name = html::escapeHTML($blog_name);
	263	$blog_name = text::cutString($blog_name,60);
	264
	265	try {
	266	$this->addBacklink($post_id,$url,$blog_name,$title,$excerpt,$comment);
	267	}
	268	catch (Exception $e) {
	269	$err = 1;
	270	$msg = 'Something went wrong : '.$e->getMessage();
	271	}
	272	}
	273
	274	$resp =
	275	'<?xml version="1.0" encoding="utf-8"?>'."\n".
	276	"<response>\n".
	277	' <error>'.(integer) $err."</error>\n";
	278
	279	if ($msg) {
	280	$resp .= ' <message>'.$msg."</message>\n";
	281	}
	282
	283	if (!empty($_POST['__debug'])) {
	284	$resp .=
	285	" <debug>\n".
	286	' <title>'.$title."</title>\n".
	287	' <excerpt>'.$excerpt."</excerpt>\n".
	288	' <url>'.$url."</url>\n".
	289	' <blog_name>'.$blog_name."</blog_name>\n".
	290	' <charset>'.$charset."</charset>\n".
	291	' <comment>'.$comment."</comment>\n".
	292	" </debug>\n";
	293	}
	294
	295	echo $resp."</response>";
	296	}
	297
	298	/**
	299	Receives a pingback and insert it as a comment of given post.
	300
	301	@param from_url <b>string</b> Source URL
	302	@param to_url <b>string</b> Target URL
	303	*/
	304	public function receivePingback($from_url,$to_url)
	305	{
	306	try {
	307	$posts = $this->getTargetPost($to_url);
	308
	309	if ($this->pingAlreadyDone($posts->post_id,$from_url)) {
	310	throw new Exception(__('Don\'t repeat yourself, please.'), 48);
	311	}
	312
	313	$remote_content = $this->getRemoteContent($from_url);
	314
	315	# We want a title...
	316	if (!preg_match('!<title>([^<].*?)</title>!mis',$remote_content,$m)) {
	317	throw new Exception(__('Where\'s your title?'), 0);
	318	}
	319	$title = trim(html::clean($m[1]));
	320	$title = html::decodeEntities($title);
	321	$title = html::escapeHTML($title);
	322	$title = text::cutString($title,60);
	323
	324	preg_match('!<body[^>]?>(.)?</body>!msi',$remote_content,$m);
	325	$source = $m[1];
	326	$source = preg_replace('![\r\n\s]+!ms',' ',$source);
	327	$source = preg_replace( "/<\/(h\d\|p\|th\|td\|li\|dt\|dd\|pre\|caption\|input\|textarea\|button)[^>]>/","\n\n",$source );
	328	$source = strip_tags($source,'<a>');
	329	$source = explode("\n\n",$source);
	330
	331	$excerpt = '';
	332	foreach ($source as $line) {
	333	if (strpos($line, $to_url) !== false) {
	334	if (preg_match("!<a[^>]+?".$to_url."[^>]*>([^>]+?)</a>!",$line,$m)) {
	335	$excerpt = strip_tags($line);
	336	break;
	337	}
	338	}
	339	}
	340	if ($excerpt) {
	341	$excerpt = '(…) '.text::cutString(html::escapeHTML($excerpt),200).' (…)';
	342	}
	343	else {
	344	$excerpt = '(…)';
	345	}
	346
	347	$this->addBacklink($posts->post_id,$from_url,'',$title,$excerpt,$comment);
	348	}
	349	catch (Exception $e) {
	350	throw new Exception(__('Sorry, an internal problem has occured.'),0);
	351	}
	352
	353	return __('Thanks, mate. It was a pleasure.');
	354	}
	355
	356	/**
	357	Receives a webmention and insert it as a comment of given post.
[3521]	358
[3415]	359	NB: plugin Fair Trackback check source content to find url.
	360
	361	@return <b>null</b> Null on success, else throw an exception
	362	*/
	363	public function receiveWebmention()
	364	{
	365	$err = $post_id = false;
	366	header('Content-Type: text/html; charset=UTF-8');
	367
	368	try {
	369	# Check if post and target are valid URL
	370	if (empty($_POST['source']) \|\| empty($_POST['target'])) {
	371	throw new Exception('Source or target is not valid',0);
	372	}
	373
	374	$from_url = urldecode($_POST['source']);
	375	$to_url = urldecode($_POST['target']);
	376
	377	self::checkURLs($from_url,$to_url);
	378
	379	# Try to find post
	380	$posts = $this->getTargetPost($to_url);
	381	$post_id = $posts->post_id;
	382
	383	# Check if it's an updated mention
	384	if ($this->pingAlreadyDone($post_id,$from_url)) {
	385	$this->delBacklink($post_id,$from_url);
	386	}
	387
	388	# Create a comment for received webmention
[3521]	389	$remote_content = $this->getRemoteContent($from_url);
	390
	391	# We want a title...
	392	if (!preg_match('!<title>([^<].*?)</title>!mis',$remote_content,$m)) {
	393	throw new Exception(__('Where\'s your title?'), 0);
	394	}
	395	$title = trim(html::clean($m[1]));
	396	$title = html::decodeEntities($title);
	397	$title = html::escapeHTML($title);
	398	$title = text::cutString($title,60);
	399
	400	preg_match('!<body[^>]?>(.)?</body>!msi',$remote_content,$m);
	401	$source = $m[1];
	402	$source = preg_replace('![\r\n\s]+!ms',' ',$source);
	403	$source = preg_replace( "/<\/(h\d\|p\|th\|td\|li\|dt\|dd\|pre\|caption\|input\|textarea\|button)[^>]>/","\n\n",$source );
	404	$source = strip_tags($source,'<a>');
	405	$source = explode("\n\n",$source);
	406
	407	$excerpt = '';
	408	foreach ($source as $line) {
	409	if (strpos($line, $to_url) !== false) {
	410	if (preg_match("!<a[^>]+?".$to_url."[^>]*>([^>]+?)</a>!",$line,$m)) {
	411	$excerpt = strip_tags($line);
	412	break;
	413	}
	414	}
	415	}
	416	if ($excerpt) {
	417	$excerpt = '(…) '.text::cutString(html::escapeHTML($excerpt),200).' (…)';
	418	}
	419	else {
	420	$excerpt = '(…)';
	421	}
	422
[3415]	423	$this->addBacklink($post_id,$from_url,'',$title,$excerpt,$comment);
	424
	425	# All done, thanks
	426	$code = $this->core->blog->settings->system->trackbacks_pub ? 200 : 202;
	427	http::head($code);
	428	return;
	429	}
	430	catch (Exception $e) {
	431	$err = $e->getMessage();
	432	}
	433
	434	http::head(400);
	435	echo $err ?: 'Something went wrong.';
	436	return;
	437	}
	438
	439	/**
	440	Check if a post previously received a ping a from an URL.
	441
	442	@param post_id <b>integer</b> Post ID
	443	@param from_url <b>string</b> Source URL
	444	@return <b>boolean</b>
	445	*/
	446	private function pingAlreadyDone($post_id,$from_url)
[1689]	447	{
	448	$params = array(
	449	'post_id' => $post_id,
	450	'comment_site' => $from_url,
	451	'comment_trackback' => 1,
	452	);
[2566]	453
[3415]	454	$rs = $this->core->blog->getComments($params,true);
[1689]	455	if ($rs && !$rs->isEmpty()) {
	456	return ($rs->f(0));
	457	}
[2566]	458
[1689]	459	return false;
	460	}
[2566]	461
[3415]	462	/**
	463	Create a comment marked as trackback for a given post.
	464
	465	@param post_id <b>integer</b> Post ID
	466	@param url <b>string</b> Discovered URL
	467	@param blog name <b>string</b> Source blog name
	468	@param title <b>string</b> Comment title
	469	@param excerpt <b>string</b> Source excerpt
	470	@param comment <b>string</b> Comment content
	471	*/
	472	private function addBacklink($post_id,$url,$blog_name,$title,$excerpt,&$comment)
[1684]	473	{
	474	if (empty($blog_name)) {
[3521]	475	// Let use title as text link for this backlink
	476	$blog_name = ($title ?: 'Anonymous blog');
[1684]	477	}
[2566]	478
[1684]	479	$comment =
	480	"<!-- TB -->\n".
[3340]	481	'<p><strong>'.($title ?: $blog_name)."</strong></p>\n".
[1684]	482	'<p>'.$excerpt.'</p>';
[2566]	483
[1684]	484	$cur = $this->core->con->openCursor($this->core->prefix.'comment');
	485	$cur->comment_author = (string) $blog_name;
	486	$cur->comment_site = (string) $url;
	487	$cur->comment_content = (string) $comment;
	488	$cur->post_id = $post_id;
	489	$cur->comment_trackback = 1;
	490	$cur->comment_status = $this->core->blog->settings->system->trackbacks_pub ? 1 : -1;
	491	$cur->comment_ip = http::realIP();
[2566]	492
[1684]	493	# --BEHAVIOR-- publicBeforeTrackbackCreate
	494	$this->core->callBehavior('publicBeforeTrackbackCreate',$cur);
	495	if ($cur->post_id) {
	496	$comment_id = $this->core->blog->addComment($cur);
[2566]	497
[1684]	498	# --BEHAVIOR-- publicAfterTrackbackCreate
	499	$this->core->callBehavior('publicAfterTrackbackCreate',$cur,$comment_id);
	500	}
	501	}
[2566]	502
[0]	503	/**
[3415]	504	Delete previously received comment made from an URL for a given post.
[2566]	505
[3415]	506	@param post_id <b>integer</b> Post ID
	507	@param url <b>string</b> Source URL
[0]	508	*/
[3415]	509	private function delBacklink($post_id,$url)
[0]	510	{
[3415]	511	$this->con->execute(
	512	'DELETE FROM '.$this->core->prefix.'comment '.
	513	'WHERE post_id = '.((integer) $post_id).' '.
	514	"AND comment_site = '".$this->core->con->escape((string) $url)."' ".
	515	'AND comment_trackback = 1 '
[1674]	516	);
[0]	517	}
[2566]	518
[3415]	519	/**
	520	Find Charset from HTTP headers.
	521
	522	@param header <b>string</b> Source header
	523	@return <b>string</b>
	524	*/
[1684]	525	private static function getCharsetFromRequest($header = '')
[0]	526	{
[1684]	527	if (!$header && isset($_SERVER['CONTENT_TYPE'])) {
	528	$header = $_SERVER['CONTENT_TYPE'];
	529	}
[2566]	530
[1684]	531	if ($header) {
	532	if (preg_match('\|charset=([a-zA-Z0-9-]+)\|',$header,$m)) {
[0]	533	return $m[1];
	534	}
	535	}
[2566]	536
[0]	537	return null;
	538	}
[1684]	539
[3415]	540	/**
	541	Detect encoding.
	542
	543	@param content <b>string</b> Source URL
	544	@return <b>string</b>
	545	*/
[2674]	546	private static function detectCharset($content)
[1684]	547	{
[2674]	548	return mb_detect_encoding($content,
[1684]	549	'UTF-8,ISO-8859-1,ISO-8859-2,ISO-8859-3,'.
	550	'ISO-8859-4,ISO-8859-5,ISO-8859-6,ISO-8859-7,ISO-8859-8,'.
	551	'ISO-8859-9,ISO-8859-10,ISO-8859-13,ISO-8859-14,ISO-8859-15');
	552	}
[2566]	553
[3415]	554	/**
[3521]	555	Retreive local post from a given URL
[3415]	556
	557	@param to_url <b>string</b> Target URL
	558	@return <b>string</b>
	559	*/
	560	private function getTargetPost($to_url)
	561	{
	562	$reg = '!^'.preg_quote($this->core->blog->url).'(.*)!';
	563	$type = $args = $next = '';
	564
	565	# Are you dumb?
	566	if (!preg_match($reg,$to_url,$m)) {
	567	throw new Exception(__('Any chance you ping one of my contents? No? Really?'),0);
	568	}
	569
	570	# Does the targeted URL look like a registered post type?
	571	$url_part = $m[1];
	572	$p_type = '';
	573	$post_types = $this->core->getPostTypes();
	574	foreach ($post_types as $k => $v) {
	575	$reg = '!^'.preg_quote(str_replace('%s','',$v['public_url'])).'(.*)!';
	576	if (preg_match($reg,$url_part,$n)) {
	577	$p_type = $k;
	578	$post_url = $n[1];
	579	break;
	580	}
	581	}
	582
	583	if (empty($p_type)) {
	584	throw new Exception(__('Sorry but you can not ping this type of content.'),33);
	585	}
	586
	587	# Time to see if we've got a winner...
	588	$params = array(
	589	'post_type' => $p_type,
	590	'post_url' => $post_url,
	591	);
	592	$posts = $this->core->blog->getPosts($params);
	593
	594	# Missed!
	595	if ($posts->isEmpty()) {
	596	throw new Exception(__('Oops. Kinda "not found" stuff. Please check the target URL twice.'),33);
	597	}
	598
	599	# Nice try. But, sorry, no.
	600	if (!$posts->trackbacksActive()) {
	601	throw new Exception(__('Sorry, dude. This entry does not accept pingback at the moment.'),33);
	602	}
	603
	604	return $posts;
	605	}
	606
	607	/**
	608	Returns content of a distant page
	609
	610	@param from_url <b>string</b> Target URL
	611	@return <b>string</b>
	612	*/
	613	private function getRemoteContent($from_url)
	614	{
	615	$http = self::initHttp($from_url,$from_path);
	616
	617	# First round : just to be sure the ping comes from an acceptable resource type.
	618	$http->setHeadersOnly(true);
	619	$http->get($from_path);
	620	$c_type = explode(';',$http->getHeader('content-type'));
	621
	622	# Bad luck. Bye, bye...
	623	if (!in_array($c_type[0],array('text/html','application/xhtml+xml'))) {
	624	throw new Exception(__('Your source URL does not look like a supported content type. Sorry. Bye, bye!'),0);
	625	}
	626
	627	# Second round : let's go fetch and parse the remote content
	628	$http->setHeadersOnly(false);
	629	$http->get($from_path);
	630	$remote_content = $http->getContent();
	631
	632	# Convert content charset
	633	$charset = self::getCharsetFromRequest($http->getHeader('content-type'));
	634	if (!$charset) {
	635	$charset = self::detectCharset($remote_content);
	636	}
	637	if (strtolower($charset) != 'utf-8') {
	638	$remote_content = iconv($charset,'UTF-8',$remote_content);
	639	}
	640
	641	return $remote_content;
	642	}
	643	//@}
	644
	645	/// @name Discover
[0]	646	//@{
	647	/**
	648	Returns an array containing all discovered trackbacks URLs in
	649	<var>$text</var>.
[2566]	650
[0]	651	@param text <b>string</b> Input text
	652	@return <b>array</b>
	653	*/
	654	public function discover($text)
	655	{
	656	$res = array();
[2566]	657
[0]	658	foreach ($this->getTextLinks($text) as $link)
	659	{
	660	if (($url = $this->getPingURL($link)) !== null) {
	661	$res[] = $url;
	662	}
	663	}
[2566]	664
[0]	665	return $res;
	666	}
[2566]	667
[3415]	668	/**
	669	Find links into a text.
	670
	671	@param text <b>string</b> Text to scan
	672	@return <b>array</b>
	673	*/
[0]	674	private function getTextLinks($text)
	675	{
	676	$res = array();
[2566]	677
[0]	678	# href attribute on "a" tags
[3415]	679	if (preg_match_all('/<a ([^>]+)>/ms',$text,$match,PREG_SET_ORDER)) {
[0]	680	for ($i = 0; $i<count($match); $i++)
	681	{
[3415]	682	if (preg_match('/href="((https?:\/)?\/[^"]+)"/ms',$match[$i][1],$matches)) {
[0]	683	$res[$matches[1]] = 1;
	684	}
	685	}
	686	}
	687	unset($match);
[2566]	688
[0]	689	# cite attributes on "blockquote" and "q" tags
[3415]	690	if (preg_match_all('/<(blockquote\|q) ([^>]+)>/ms',$text,$match,PREG_SET_ORDER)) {
[0]	691	for ($i = 0; $i<count($match); $i++)
	692	{
[3415]	693	if (preg_match('/cite="((https?:\/)?\/[^"]+)"/ms',$match[$i][2],$matches)) {
[0]	694	$res[$matches[1]] = 1;
	695	}
	696	}
	697	}
[2566]	698
[0]	699	return array_keys($res);
	700	}
[2566]	701
[3415]	702	/**
	703	Check remote header/content to find api trace.
	704
	705	@param url <b>string</b> URL to scan
	706	@return <b>string</b>
	707	*/
[0]	708	private function getPingURL($url)
	709	{
[1672]	710	if (strpos($url,'/') === 0) {
	711	$url = http::getHost().$url;
	712	}
[2566]	713
[3415]	714	try {
[0]	715	$http = self::initHttp($url,$path);
	716	$http->get($path);
	717	$page_content = $http->getContent();
[1671]	718	$pb_url = $http->getHeader('x-pingback');
[3415]	719	$wm_url = $http->getHeader('link');
[0]	720	}
[3415]	721	catch (Exception $e) {
[0]	722	return false;
	723	}
[2566]	724
[3521]	725	# Let's check for an elderly trackback data chunk...
[0]	726	$pattern_rdf =
	727	'/<rdf:RDF.?>.?'.
	728	'<rdf:Description\s+(.*?)\/>'.
	729	'.*?<\/rdf:RDF>'.
	730	'/msi';
[2566]	731
[0]	732	preg_match_all($pattern_rdf,$page_content,$rdf_all,PREG_SET_ORDER);
[2566]	733
[1671]	734	$url_path = parse_url($url, PHP_URL_PATH);
[3415]	735	$sanitized_url = str_replace($url_path, html::sanitizeURL($url_path),$url);
[2566]	736
[0]	737	for ($i=0; $i<count($rdf_all); $i++)
	738	{
	739	$rdf = $rdf_all[$i][1];
[1671]	740	if (preg_match('/dc:identifier="'.preg_quote($url,'/').'"/msi',$rdf) \|\|
	741	preg_match('/dc:identifier="'.preg_quote($sanitized_url,'/').'"/msi',$rdf)) {
[0]	742	if (preg_match('/trackback:ping="(.*?)"/msi',$rdf,$tb_link)) {
	743	return $tb_link[1];
	744	}
	745	}
	746	}
[2566]	747
[3521]	748	# No trackback ? OK, let see if we've got a X-Pingback header and it's a valid URL, it will be enough
	749	if ($pb_url && filter_var($pb_url,FILTER_VALIDATE_URL) && preg_match('!^https?:!',$pb_url)) {
	750	return $pb_url.'\|'.$url;
	751	}
	752
	753	# No X-Pingback header. A link rel=pingback, maybe ?
	754	$pattern_pingback = '!<link rel="pingback" href="(.*?)"( /)?>!msi';
	755
	756	if (preg_match($pattern_pingback,$page_content,$m)) {
	757	$pb_url = $m[1];
	758	if (filter_var($pb_url,FILTER_VALIDATE_URL) && preg_match('!^https?:!',$pb_url)) {
	759	return $pb_url.'\|'.$url;
	760	}
	761	}
	762
[3415]	763	# Nothing, let's try webmention. Only support x/html content
	764	if ($wm_url) {
	765	$type = explode(';',$http->getHeader('content-type'));
	766	if (!in_array($type[0],array('text/html','application/xhtml+xml'))) {
	767	$wm_url = false;
	768	}
	769	}
	770
	771	# Check HTTP headers for a Link: <ENDPOINT_URL>; rel="webmention"
	772	$wm_api = false;
	773	if ($wm_url) {
	774	if(preg_match('~<((?:https?://)?[^>]+)>; rel="?(?:https?://webmention.org/?\|webmention)"?~',$wm_url,$match)) {
	775	if (filter_var($match[1],FILTER_VALIDATE_URL) && preg_match('!^https?:!',$match[1])) {
	776	$wm_api = $match[1];
	777	}
	778	}
	779	}
	780
	781	# Else check content for <link href="ENDPOINT_URL" rel="webmention" />
	782	if ($wm_url && !$wm_api) {
	783	$content = preg_replace('/<!--(.*)-->/Us','',$page_content);
	784	if (preg_match('/<(?:link\|a)[ ]+href="([^"])"[ ]+rel="[^" ] ?webmention ?[^" ]"[ ]\/?>/i',$content,$match)
	785	\|\| preg_match('/<(?:link\|a)[ ]+rel="[^" ]* ?webmention ?[^" ]"[ ]+href="([^"])"[ ]*\/?>/i',$content,$match)) {
	786	$wm_api = $match[1];
	787	}
	788	}
	789
	790	# We have a winner, let's add some tricks to make diference
	791	if ($wm_api) {
	792	return $wm_api.'\|'.$url.'\|webmention';
	793	}
	794
[0]	795	return null;
	796	}
[3415]	797	//@}
	798
	799	/**
	800	HTTP helper.
	801
	802	@param url <b>string</b> URL
	803	@param path <b>string</b> Path
	804	@return <b>object</b>
	805	*/
	806	private static function initHttp($url,&$path)
	807	{
	808	$client = netHttp::initClient($url,$path);
	809	$client->setTimeout(5);
	810	$client->setUserAgent('Dotclear - http://www.dotclear.org/');
	811	$client->useGzip(false);
	812	$client->setPersistReferers(false);
	813
	814	return $client;
	815	}
	816
	817	/**
	818	URL helper.
	819
	820	@param from_url <b>string</b> URL a
	821	@param to_url <b>string</b> URL b
	822	*/
	823	public static function checkURLs($from_url,$to_url)
	824	{
[3521]	825	if (!(filter_var($from_url,FILTER_VALIDATE_URL) && preg_match('!^https?://!',$from_url))) {
[3415]	826	throw new Exception(__('No valid source URL provided? Try again!'), 0);
	827	}
	828
[3521]	829	if (!(filter_var($to_url,FILTER_VALIDATE_URL) && preg_match('!^https?://!',$to_url))) {
[3415]	830	throw new Exception(__('No valid target URL provided? Try again!'), 0);
	831	}
	832
	833	if (html::sanitizeURL(urldecode($from_url)) == html::sanitizeURL(urldecode($to_url))) {
	834	throw new Exception(__('LOL!'), 0);
	835	}
	836	}
[3521]	837	}

Note: See TracBrowser for help on using the repository browser.

Dotclear

Context Navigation

source: inc/core/class.dc.trackback.php @ 3521:b8a16a351d6d

Download in other formats:

Sites map