[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 $table; ///< <b>string</b> done pings table name |
---|
| 24 | |
---|
| 25 | /** |
---|
| 26 | Object constructor |
---|
| 27 | |
---|
| 28 | @param core <b>dcCore</b> dcCore instance |
---|
| 29 | */ |
---|
| 30 | public function __construct($core) |
---|
| 31 | { |
---|
| 32 | $this->core =& $core; |
---|
| 33 | $this->con =& $this->core->con; |
---|
| 34 | $this->table = $this->core->prefix.'ping'; |
---|
| 35 | } |
---|
| 36 | |
---|
| 37 | /// @name Send trackbacks |
---|
| 38 | //@{ |
---|
| 39 | /** |
---|
| 40 | Get all pings sent for a given post. |
---|
| 41 | |
---|
| 42 | @param post_id <b>integer</b> Post ID |
---|
| 43 | @return <b>record</b> |
---|
| 44 | */ |
---|
| 45 | public function getPostPings($post_id) |
---|
| 46 | { |
---|
| 47 | $strReq = 'SELECT ping_url, ping_dt '. |
---|
| 48 | 'FROM '.$this->table.' '. |
---|
| 49 | 'WHERE post_id = '.(integer) $post_id; |
---|
| 50 | |
---|
| 51 | return $this->con->select($strReq); |
---|
| 52 | } |
---|
| 53 | |
---|
| 54 | /** |
---|
| 55 | Sends a ping to given <var>$url</var>. |
---|
| 56 | |
---|
| 57 | @param url <b>string</b> URL to ping |
---|
| 58 | @param post_id <b>integer</b> Post ID |
---|
| 59 | @param post_title <b>string</b> Post title |
---|
| 60 | @param post_excerpt <b>string</b> Post excerpt |
---|
| 61 | @param post_url <b>string</b> Post URL |
---|
| 62 | */ |
---|
| 63 | public function ping($url,$post_id,$post_title,$post_excerpt,$post_url) |
---|
| 64 | { |
---|
| 65 | if ($this->core->blog === null) { |
---|
| 66 | return false; |
---|
| 67 | } |
---|
| 68 | |
---|
| 69 | $post_id = (integer) $post_id; |
---|
| 70 | |
---|
| 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)."' "; |
---|
| 75 | |
---|
| 76 | $rs = $this->con->select($strReq); |
---|
| 77 | |
---|
| 78 | if (!$rs->isEmpty()) { |
---|
| 79 | throw new Exception(sprintf(__('%s has still been pinged'),$url)); |
---|
| 80 | } |
---|
| 81 | |
---|
[1671] | 82 | $ping_parts = explode('|',$url); |
---|
[0] | 83 | |
---|
[1671] | 84 | # Let's walk by the trackback way |
---|
| 85 | if (count($ping_parts) < 2) { |
---|
| 86 | $data = array( |
---|
| 87 | 'title' => $post_title, |
---|
| 88 | 'excerpt' => $post_excerpt, |
---|
| 89 | 'url' => $post_url, |
---|
| 90 | 'blog_name' => trim(html::escapeHTML(html::clean($this->core->blog->name))) |
---|
| 91 | //,'__debug' => false |
---|
| 92 | ); |
---|
| 93 | |
---|
| 94 | # Ping |
---|
| 95 | try |
---|
| 96 | { |
---|
| 97 | $http = self::initHttp($url,$path); |
---|
| 98 | $http->post($path,$data,'UTF-8'); |
---|
| 99 | $res = $http->getContent(); |
---|
| 100 | } |
---|
| 101 | catch (Exception $e) |
---|
| 102 | { |
---|
| 103 | throw new Exception(__('Unable to ping URL')); |
---|
| 104 | } |
---|
| 105 | |
---|
| 106 | $pattern = |
---|
| 107 | '|<response>.*<error>(.*)</error>(.*)'. |
---|
| 108 | '(<message>(.*)</message>(.*))?'. |
---|
| 109 | '</response>|msU'; |
---|
| 110 | |
---|
| 111 | if (!preg_match($pattern,$res,$match)) |
---|
| 112 | { |
---|
| 113 | throw new Exception(sprintf(__('%s is not a ping URL'),$url)); |
---|
| 114 | } |
---|
| 115 | |
---|
| 116 | $ping_error = trim($match[1]); |
---|
| 117 | $ping_msg = (!empty($match[4])) ? $match[4] : ''; |
---|
[0] | 118 | } |
---|
[1671] | 119 | # Damnit ! Let's play pingback |
---|
| 120 | else { |
---|
| 121 | try { |
---|
| 122 | $xmlrpc = new xmlrpcClient($ping_parts[0]); |
---|
| 123 | $res = $xmlrpc->query('pingback.ping', $post_url, $ping_parts[1]); |
---|
| 124 | $ping_error = '0'; |
---|
| 125 | } |
---|
| 126 | catch (xmlrpcException $e) { |
---|
| 127 | $ping_error = $e->getCode(); |
---|
| 128 | $ping_msg = $e->getMessage(); |
---|
| 129 | } |
---|
| 130 | catch (Exception $e) { |
---|
| 131 | throw new Exception(__('Unable to ping URL')); |
---|
| 132 | } |
---|
[0] | 133 | } |
---|
| 134 | |
---|
| 135 | if ($ping_error != '0') { |
---|
| 136 | throw new Exception(sprintf(__('%s, ping error:'),$url).' '.$ping_msg); |
---|
| 137 | } else { |
---|
| 138 | # Notify ping result in database |
---|
| 139 | $cur = $this->con->openCursor($this->table); |
---|
| 140 | $cur->post_id = $post_id; |
---|
| 141 | $cur->ping_url = $url; |
---|
| 142 | $cur->ping_dt = date('Y-m-d H:i:s'); |
---|
| 143 | |
---|
| 144 | $cur->insert(); |
---|
| 145 | } |
---|
| 146 | } |
---|
| 147 | //@} |
---|
| 148 | |
---|
[1689] | 149 | private function pingAlreadyDone($post_id, $from_url) |
---|
| 150 | { |
---|
| 151 | $params = array( |
---|
| 152 | 'post_id' => $post_id, |
---|
| 153 | 'comment_site' => $from_url, |
---|
| 154 | 'comment_trackback' => 1, |
---|
| 155 | ); |
---|
| 156 | |
---|
| 157 | $rs = $this->core->blog->getComments($params, true); |
---|
| 158 | if ($rs && !$rs->isEmpty()) { |
---|
| 159 | return ($rs->f(0)); |
---|
| 160 | } |
---|
| 161 | |
---|
| 162 | return false; |
---|
| 163 | } |
---|
| 164 | |
---|
[1684] | 165 | private function addBacklink($post_id, $url, $blog_name, $title, $excerpt, &$comment) |
---|
| 166 | { |
---|
| 167 | if (empty($blog_name)) { |
---|
| 168 | $blog_name = 'Anonymous blog'; |
---|
| 169 | } |
---|
| 170 | |
---|
| 171 | $comment = |
---|
| 172 | "<!-- TB -->\n". |
---|
| 173 | '<p><strong>'.($title ? $title : $blog_name)."</strong></p>\n". |
---|
| 174 | '<p>'.$excerpt.'</p>'; |
---|
| 175 | |
---|
| 176 | $cur = $this->core->con->openCursor($this->core->prefix.'comment'); |
---|
| 177 | $cur->comment_author = (string) $blog_name; |
---|
| 178 | $cur->comment_site = (string) $url; |
---|
| 179 | $cur->comment_content = (string) $comment; |
---|
| 180 | $cur->post_id = $post_id; |
---|
| 181 | $cur->comment_trackback = 1; |
---|
| 182 | $cur->comment_status = $this->core->blog->settings->system->trackbacks_pub ? 1 : -1; |
---|
| 183 | $cur->comment_ip = http::realIP(); |
---|
| 184 | |
---|
| 185 | # --BEHAVIOR-- publicBeforeTrackbackCreate |
---|
| 186 | $this->core->callBehavior('publicBeforeTrackbackCreate',$cur); |
---|
| 187 | if ($cur->post_id) { |
---|
| 188 | $comment_id = $this->core->blog->addComment($cur); |
---|
| 189 | |
---|
| 190 | # --BEHAVIOR-- publicAfterTrackbackCreate |
---|
| 191 | $this->core->callBehavior('publicAfterTrackbackCreate',$cur,$comment_id); |
---|
| 192 | } |
---|
| 193 | } |
---|
| 194 | |
---|
[0] | 195 | /// @name Receive trackbacks |
---|
| 196 | //@{ |
---|
| 197 | /** |
---|
| 198 | Receives a trackback and insert it as a comment of given post. |
---|
| 199 | |
---|
| 200 | @param post_id <b>integer</b> Post ID |
---|
| 201 | */ |
---|
| 202 | public function receive($post_id) |
---|
| 203 | { |
---|
| 204 | header('Content-Type: text/xml; charset=UTF-8'); |
---|
| 205 | if (empty($_POST)) { |
---|
| 206 | http::head(405,'Method Not Allowed'); |
---|
| 207 | echo |
---|
| 208 | '<?xml version="1.0" encoding="utf-8"?>'."\n". |
---|
| 209 | "<response>\n". |
---|
| 210 | " <error>1</error>\n". |
---|
| 211 | " <message>POST request needed</message>\n". |
---|
| 212 | "</response>"; |
---|
| 213 | return; |
---|
| 214 | } |
---|
| 215 | |
---|
| 216 | $post_id = (integer) $post_id; |
---|
| 217 | |
---|
| 218 | $title = !empty($_POST['title']) ? $_POST['title'] : ''; |
---|
| 219 | $excerpt = !empty($_POST['excerpt']) ? $_POST['excerpt'] : ''; |
---|
| 220 | $url = !empty($_POST['url']) ? $_POST['url'] : ''; |
---|
| 221 | $blog_name = !empty($_POST['blog_name']) ? $_POST['blog_name'] : ''; |
---|
| 222 | $charset = ''; |
---|
| 223 | $comment = ''; |
---|
[1689] | 224 | |
---|
[0] | 225 | $err = false; |
---|
| 226 | $msg = ''; |
---|
| 227 | |
---|
| 228 | if ($this->core->blog === null) |
---|
| 229 | { |
---|
| 230 | $err = true; |
---|
| 231 | $msg = 'No blog.'; |
---|
| 232 | } |
---|
| 233 | elseif ($url == '') |
---|
| 234 | { |
---|
| 235 | $err = true; |
---|
| 236 | $msg = 'URL parameter is required.'; |
---|
| 237 | } |
---|
| 238 | elseif ($blog_name == '') { |
---|
| 239 | $err = true; |
---|
| 240 | $msg = 'Blog name is required.'; |
---|
| 241 | } |
---|
| 242 | |
---|
| 243 | if (!$err) |
---|
| 244 | { |
---|
| 245 | $post = $this->core->blog->getPosts(array('post_id'=>$post_id,'post_type'=>'')); |
---|
| 246 | |
---|
| 247 | if ($post->isEmpty()) |
---|
| 248 | { |
---|
| 249 | $err = true; |
---|
| 250 | $msg = 'No such post.'; |
---|
| 251 | } |
---|
| 252 | elseif (!$post->trackbacksActive()) |
---|
| 253 | { |
---|
| 254 | $err = true; |
---|
| 255 | $msg = 'Trackbacks are not allowed for this post or weblog.'; |
---|
| 256 | } |
---|
[1689] | 257 | |
---|
| 258 | $url = trim(html::clean($url)); |
---|
| 259 | if ($this->pingAlreadyDone($post->post_id, $url)) { |
---|
| 260 | $err = true; |
---|
| 261 | $msg = 'The trackback has already been registered'; |
---|
| 262 | } |
---|
[0] | 263 | } |
---|
| 264 | |
---|
| 265 | if (!$err) |
---|
| 266 | { |
---|
| 267 | $charset = self::getCharsetFromRequest(); |
---|
| 268 | |
---|
| 269 | if (!$charset) { |
---|
[1684] | 270 | $charset = self::detectCharset($title.' '.$excerpt.' '.$blog_name); |
---|
[0] | 271 | } |
---|
| 272 | |
---|
| 273 | if (strtolower($charset) != 'utf-8') { |
---|
| 274 | $title = iconv($charset,'UTF-8',$title); |
---|
| 275 | $excerpt = iconv($charset,'UTF-8',$excerpt); |
---|
| 276 | $blog_name = iconv($charset,'UTF-8',$blog_name); |
---|
| 277 | } |
---|
| 278 | |
---|
| 279 | $title = trim(html::clean($title)); |
---|
| 280 | $title = html::decodeEntities($title); |
---|
| 281 | $title = html::escapeHTML($title); |
---|
| 282 | $title = text::cutString($title,60); |
---|
| 283 | |
---|
| 284 | $excerpt = trim(html::clean($excerpt)); |
---|
| 285 | $excerpt = html::decodeEntities($excerpt); |
---|
| 286 | $excerpt = preg_replace('/\s+/ms',' ',$excerpt); |
---|
| 287 | $excerpt = text::cutString($excerpt,252); |
---|
| 288 | $excerpt = html::escapeHTML($excerpt).'...'; |
---|
| 289 | |
---|
| 290 | $blog_name = trim(html::clean($blog_name)); |
---|
| 291 | $blog_name = html::decodeEntities($blog_name); |
---|
| 292 | $blog_name = html::escapeHTML($blog_name); |
---|
| 293 | $blog_name = text::cutString($blog_name,60); |
---|
| 294 | |
---|
| 295 | try |
---|
| 296 | { |
---|
[1684] | 297 | $this->addBacklink($post_id, $url, $blog_name, $title, $excerpt, $comment); |
---|
[0] | 298 | } |
---|
| 299 | catch (Exception $e) |
---|
| 300 | { |
---|
| 301 | $err = 1; |
---|
| 302 | $msg = 'Something went wrong : '.$e->getMessage(); |
---|
| 303 | } |
---|
| 304 | } |
---|
| 305 | |
---|
| 306 | $resp = |
---|
| 307 | '<?xml version="1.0" encoding="utf-8"?>'."\n". |
---|
| 308 | "<response>\n". |
---|
| 309 | ' <error>'.(integer) $err."</error>\n"; |
---|
| 310 | |
---|
| 311 | if ($msg) { |
---|
| 312 | $resp .= ' <message>'.$msg."</message>\n"; |
---|
| 313 | } |
---|
| 314 | |
---|
| 315 | if (!empty($_POST['__debug'])) { |
---|
[1684] | 316 | $resp .= |
---|
| 317 | " <debug>\n". |
---|
| 318 | ' <title>'.$title."</title>\n". |
---|
| 319 | ' <excerpt>'.$excerpt."</excerpt>\n". |
---|
| 320 | ' <url>'.$url."</url>\n". |
---|
| 321 | ' <blog_name>'.$blog_name."</blog_name>\n". |
---|
| 322 | ' <charset>'.$charset."</charset>\n". |
---|
| 323 | ' <comment>'.$comment."</comment>\n". |
---|
| 324 | " </debug>\n"; |
---|
[0] | 325 | } |
---|
| 326 | |
---|
| 327 | echo $resp."</response>"; |
---|
| 328 | } |
---|
| 329 | //@} |
---|
[1674] | 330 | |
---|
| 331 | /// @name Receive pingbacks |
---|
| 332 | //@{ |
---|
| 333 | /** |
---|
| 334 | Receives a pingback and insert it as a comment of given post. |
---|
| 335 | |
---|
| 336 | @param from_url <b>string</b> Source URL |
---|
| 337 | @param to_url <b>string</b> Target URL |
---|
| 338 | */ |
---|
| 339 | public function receive_pb($from_url, $to_url) |
---|
| 340 | { |
---|
| 341 | $reg = '!^'.preg_quote($this->core->blog->url).'(.*)!'; |
---|
| 342 | $type = $args = $next = ''; |
---|
| 343 | |
---|
| 344 | # Are you dumb? |
---|
| 345 | if (!preg_match($reg, $to_url, $m)) { |
---|
| 346 | throw new Exception(__('Any chance you ping one of my contents? No? Really?'), 0); |
---|
| 347 | } |
---|
| 348 | |
---|
| 349 | # Does the targeted URL look like a registered post type? |
---|
| 350 | $url_part = $m[1]; |
---|
| 351 | $p_type = ''; |
---|
| 352 | $post_types = $this->core->getPostTypes(); |
---|
| 353 | foreach ($post_types as $k => $v) { |
---|
| 354 | $reg = '!^'.preg_quote(str_replace('%s', '', $v['public_url'])).'(.*)!'; |
---|
| 355 | if (preg_match($reg, $url_part, $n)) { |
---|
| 356 | $p_type = $k; |
---|
| 357 | $post_url = $n[1]; |
---|
| 358 | break; |
---|
| 359 | } |
---|
| 360 | } |
---|
| 361 | |
---|
| 362 | if (empty($p_type)) { |
---|
| 363 | throw new Exception(__('Sorry but you can not ping this type of content.'), 33); |
---|
| 364 | } |
---|
| 365 | |
---|
| 366 | # Time to see if we've got a winner... |
---|
| 367 | $params = array( |
---|
| 368 | 'post_type' => $p_type, |
---|
| 369 | 'post_url' => $post_url, |
---|
| 370 | ); |
---|
| 371 | $posts = $this->core->blog->getPosts($params); |
---|
| 372 | |
---|
| 373 | # Missed! |
---|
| 374 | if ($posts->isEmpty()) { |
---|
| 375 | throw new Exception(__('Oops. Kinda "not found" stuff. Please check the target URL twice.'), 33); |
---|
| 376 | } |
---|
| 377 | |
---|
| 378 | # Nice try. But, sorry, no. |
---|
| 379 | if (!$posts->trackbacksActive()) { |
---|
| 380 | throw new Exception(__('Sorry, dude. This entry does not accept pingback at the moment.'), 33); |
---|
| 381 | } |
---|
| 382 | |
---|
[1689] | 383 | if ($this->pingAlreadyDone($posts->post_id, $from_url)) { |
---|
| 384 | throw new Exception(__('Don\'t repeat yourself, please.'), 48); |
---|
| 385 | } |
---|
| 386 | |
---|
[1674] | 387 | # OK. We've found our champion. Time to check the remote part. |
---|
| 388 | try { |
---|
| 389 | $http = self::initHttp($from_url, $from_path); |
---|
| 390 | |
---|
| 391 | # First round : just to be sure the ping comes from an acceptable resource type. |
---|
| 392 | $http->setHeadersOnly(true); |
---|
| 393 | $http->get($from_path); |
---|
| 394 | $c_type = explode(';', $http->getHeader('content-type')); |
---|
| 395 | |
---|
| 396 | # Bad luck. Bye, bye... |
---|
| 397 | if (!in_array($c_type[0],array('text/html', 'application/xhtml+xml'))) { |
---|
| 398 | throw new Exception(__('Your source URL does not look like a supported content type. Sorry. Bye, bye!'), 0); |
---|
| 399 | } |
---|
| 400 | |
---|
| 401 | # Second round : let's go fetch and parse the remote content |
---|
| 402 | $http->setHeadersOnly(false); |
---|
| 403 | $http->get($from_path); |
---|
| 404 | $remote_content = $http->getContent(); |
---|
| 405 | |
---|
[1684] | 406 | $charset = self::getCharsetFromRequest($http->getHeader('content-type')); |
---|
[1674] | 407 | |
---|
[1684] | 408 | if (!$charset) { |
---|
| 409 | $charset = self::detectCharset($remote_content); |
---|
| 410 | } |
---|
| 411 | |
---|
[1674] | 412 | if (strtolower($charset) != 'utf-8') { |
---|
| 413 | $remote_content = iconv($charset,'UTF-8',$remote_content); |
---|
| 414 | } |
---|
| 415 | |
---|
| 416 | # We want a title... |
---|
| 417 | if (!preg_match('!<title>([^<].*?)</title>!mis', $remote_content, $m)) { |
---|
| 418 | throw new Exception(__('Where\'s your title?'), 0); |
---|
| 419 | } |
---|
| 420 | $title = trim(html::clean($m[1])); |
---|
| 421 | $title = html::decodeEntities($title); |
---|
| 422 | $title = html::escapeHTML($title); |
---|
| 423 | $title = text::cutString($title,60); |
---|
| 424 | |
---|
| 425 | preg_match('!<body[^>]*?>(.*)?</body>!msi', $remote_content, $m); |
---|
| 426 | $source = $m[1]; |
---|
| 427 | $source = preg_replace('![\r\n\s]+!ms',' ',$source); |
---|
| 428 | $source = preg_replace( "/<\/*(h\d|p|th|td|li|dt|dd|pre|caption|input|textarea|button)[^>]*>/", "\n\n", $source ); |
---|
| 429 | $source = strip_tags($source, '<a>'); |
---|
| 430 | $source = explode("\n\n",$source); |
---|
| 431 | |
---|
| 432 | $excerpt = ''; |
---|
| 433 | foreach ($source as $line) { |
---|
| 434 | if (strpos($line, $to_url) !== false) { |
---|
| 435 | if (preg_match("!<a[^>]+?".$to_url."[^>]*>([^>]+?)</a>!", $line, $m)) { |
---|
| 436 | $excerpt = strip_tags($line); |
---|
| 437 | break; |
---|
| 438 | } |
---|
| 439 | } |
---|
| 440 | } |
---|
| 441 | if ($excerpt) { |
---|
[1684] | 442 | $excerpt = '(…) '.text::cutString(html::escapeHTML($excerpt),200).' (…)'; |
---|
[1674] | 443 | } |
---|
| 444 | else { |
---|
[1684] | 445 | $excerpt = '(…)'; |
---|
[1674] | 446 | } |
---|
| 447 | |
---|
[1684] | 448 | $this->addBacklink($posts->post_id, $from_url, '', $title, $excerpt, $comment); |
---|
[1674] | 449 | } |
---|
| 450 | catch (Exception $e) { |
---|
| 451 | throw new Exception(__('Sorry, an internal problem has occured.'), 0); |
---|
| 452 | } |
---|
| 453 | |
---|
| 454 | return __('Thanks, mate. It was a pleasure.'); |
---|
| 455 | } |
---|
| 456 | //@} |
---|
[0] | 457 | |
---|
| 458 | private static function initHttp($url,&$path) |
---|
| 459 | { |
---|
| 460 | $client = netHttp::initClient($url,$path); |
---|
| 461 | $client->setTimeout(5); |
---|
| 462 | $client->setUserAgent('Dotclear - http://www.dotclear.org/'); |
---|
| 463 | $client->useGzip(false); |
---|
| 464 | $client->setPersistReferers(false); |
---|
| 465 | |
---|
| 466 | return $client; |
---|
| 467 | } |
---|
| 468 | |
---|
[1684] | 469 | private static function getCharsetFromRequest($header = '') |
---|
[0] | 470 | { |
---|
[1684] | 471 | if (!$header && isset($_SERVER['CONTENT_TYPE'])) { |
---|
| 472 | $header = $_SERVER['CONTENT_TYPE']; |
---|
| 473 | } |
---|
| 474 | |
---|
| 475 | if ($header) { |
---|
| 476 | if (preg_match('|charset=([a-zA-Z0-9-]+)|',$header,$m)) { |
---|
[0] | 477 | return $m[1]; |
---|
| 478 | } |
---|
| 479 | } |
---|
| 480 | |
---|
| 481 | return null; |
---|
| 482 | } |
---|
[1684] | 483 | |
---|
| 484 | private static function detectCharset($string) |
---|
| 485 | { |
---|
| 486 | return mb_detect_encoding($remote_content, |
---|
| 487 | 'UTF-8,ISO-8859-1,ISO-8859-2,ISO-8859-3,'. |
---|
| 488 | 'ISO-8859-4,ISO-8859-5,ISO-8859-6,ISO-8859-7,ISO-8859-8,'. |
---|
| 489 | 'ISO-8859-9,ISO-8859-10,ISO-8859-13,ISO-8859-14,ISO-8859-15'); |
---|
| 490 | } |
---|
| 491 | |
---|
[0] | 492 | /// @name Trackbacks auto discovery |
---|
| 493 | //@{ |
---|
| 494 | /** |
---|
| 495 | Returns an array containing all discovered trackbacks URLs in |
---|
| 496 | <var>$text</var>. |
---|
| 497 | |
---|
| 498 | @param text <b>string</b> Input text |
---|
| 499 | @return <b>array</b> |
---|
| 500 | */ |
---|
| 501 | public function discover($text) |
---|
| 502 | { |
---|
| 503 | $res = array(); |
---|
| 504 | |
---|
| 505 | foreach ($this->getTextLinks($text) as $link) |
---|
| 506 | { |
---|
| 507 | if (($url = $this->getPingURL($link)) !== null) { |
---|
| 508 | $res[] = $url; |
---|
| 509 | } |
---|
| 510 | } |
---|
| 511 | |
---|
| 512 | return $res; |
---|
| 513 | } |
---|
| 514 | //@} |
---|
| 515 | |
---|
| 516 | private function getTextLinks($text) |
---|
| 517 | { |
---|
| 518 | $res = array(); |
---|
| 519 | |
---|
| 520 | # href attribute on "a" tags |
---|
| 521 | if (preg_match_all('/<a ([^>]+)>/ms', $text, $match, PREG_SET_ORDER)) |
---|
| 522 | { |
---|
| 523 | for ($i = 0; $i<count($match); $i++) |
---|
| 524 | { |
---|
[1672] | 525 | if (preg_match('/href="((https?:\/)?\/[^"]+)"/ms', $match[$i][1], $matches)) { |
---|
[0] | 526 | $res[$matches[1]] = 1; |
---|
| 527 | } |
---|
| 528 | } |
---|
| 529 | } |
---|
| 530 | unset($match); |
---|
| 531 | |
---|
| 532 | # cite attributes on "blockquote" and "q" tags |
---|
| 533 | if (preg_match_all('/<(blockquote|q) ([^>]+)>/ms', $text, $match, PREG_SET_ORDER)) |
---|
| 534 | { |
---|
| 535 | for ($i = 0; $i<count($match); $i++) |
---|
| 536 | { |
---|
[1672] | 537 | if (preg_match('/cite="((https?:\/)?\/[^"]+)"/ms', $match[$i][2], $matches)) { |
---|
[0] | 538 | $res[$matches[1]] = 1; |
---|
| 539 | } |
---|
| 540 | } |
---|
| 541 | } |
---|
| 542 | |
---|
| 543 | return array_keys($res); |
---|
| 544 | } |
---|
| 545 | |
---|
| 546 | private function getPingURL($url) |
---|
| 547 | { |
---|
[1672] | 548 | if (strpos($url,'/') === 0) { |
---|
| 549 | $url = http::getHost().$url; |
---|
| 550 | } |
---|
| 551 | |
---|
[0] | 552 | try |
---|
| 553 | { |
---|
| 554 | $http = self::initHttp($url,$path); |
---|
| 555 | $http->get($path); |
---|
| 556 | $page_content = $http->getContent(); |
---|
[1671] | 557 | $pb_url = $http->getHeader('x-pingback'); |
---|
[0] | 558 | } |
---|
| 559 | catch (Exception $e) |
---|
| 560 | { |
---|
| 561 | return false; |
---|
| 562 | } |
---|
| 563 | |
---|
[1671] | 564 | # If we've got a X-Pingback header and it's a valid URL, it will be enough |
---|
| 565 | if ($pb_url && filter_var($pb_url,FILTER_VALIDATE_URL) && preg_match('!^https?:!',$pb_url)) { |
---|
| 566 | return $pb_url.'|'.$url; |
---|
| 567 | } |
---|
| 568 | |
---|
[1673] | 569 | # No X-Pingback header. A link rel=pingback, maybe ? |
---|
| 570 | $pattern_pingback = '!<link rel="pingback" href="(.*?)"( /)?>!msi'; |
---|
| 571 | |
---|
| 572 | if (preg_match($pattern_pingback,$page_content,$m)) { |
---|
| 573 | $pb_url = $m[1]; |
---|
| 574 | if (filter_var($pb_url,FILTER_VALIDATE_URL) && preg_match('!^https?:!',$pb_url)) { |
---|
| 575 | return $pb_url.'|'.$url; |
---|
| 576 | } |
---|
| 577 | } |
---|
| 578 | |
---|
| 579 | # No pingback ? OK, let's check for a trackback data chunk... |
---|
[0] | 580 | $pattern_rdf = |
---|
| 581 | '/<rdf:RDF.*?>.*?'. |
---|
| 582 | '<rdf:Description\s+(.*?)\/>'. |
---|
| 583 | '.*?<\/rdf:RDF>'. |
---|
| 584 | '/msi'; |
---|
| 585 | |
---|
| 586 | preg_match_all($pattern_rdf,$page_content,$rdf_all,PREG_SET_ORDER); |
---|
| 587 | |
---|
[1671] | 588 | $url_path = parse_url($url, PHP_URL_PATH); |
---|
| 589 | $sanitized_url = str_replace($url_path, html::sanitizeURL($url_path), $url); |
---|
| 590 | |
---|
[0] | 591 | for ($i=0; $i<count($rdf_all); $i++) |
---|
| 592 | { |
---|
| 593 | $rdf = $rdf_all[$i][1]; |
---|
[1671] | 594 | if (preg_match('/dc:identifier="'.preg_quote($url,'/').'"/msi',$rdf) || |
---|
| 595 | preg_match('/dc:identifier="'.preg_quote($sanitized_url,'/').'"/msi',$rdf)) { |
---|
[0] | 596 | if (preg_match('/trackback:ping="(.*?)"/msi',$rdf,$tb_link)) { |
---|
| 597 | return $tb_link[1]; |
---|
| 598 | } |
---|
| 599 | } |
---|
| 600 | } |
---|
| 601 | |
---|
| 602 | return null; |
---|
| 603 | } |
---|
| 604 | } |
---|
| 605 | ?> |
---|