mod/parse_url.php

   1 <?php
   2 /**
   3  * @file mod/parse_url.php
   4  *
   5  * @todo https://developers.google.com/+/plugins/snippet/
   6  *
   7  * @verbatim
   8  * <meta itemprop="name" content="Toller Titel">
   9  * <meta itemprop="description" content="Eine tolle Beschreibung">
  10  * <meta itemprop="image" content="http://maple.libertreeproject.org/images/tree-icon.png">
  11  *
  12  * <body itemscope itemtype="http://schema.org/Product">
  13  *   <h1 itemprop="name">Shiny Trinket</h1>
  14  *   <img itemprop="image" src="{image-url}" />
  15  *   <p itemprop="description">Shiny trinkets are shiny.</p>
  16  * </body>
  17  * @endverbatim
  18 */
  19
  20 if(!function_exists('deletenode')) {
  21         function deletenode(&$doc, $node)
  22         {
  23                 $xpath = new DomXPath($doc);
  24                 $list = $xpath->query("//".$node);
  25                 foreach ($list as $child)
  26                         $child->parentNode->removeChild($child);
  27         }
  28 }
  29
  30 function completeurl($url, $scheme) {
  31         $urlarr = parse_url($url);
  32
  33         if (isset($urlarr["scheme"]))
  34                 return($url);
  35
  36         $schemearr = parse_url($scheme);
  37
  38         $complete = $schemearr["scheme"]."://".$schemearr["host"];
  39
  40         if (@$schemearr["port"] != "")
  41                 $complete .= ":".$schemearr["port"];
  42
  43                 if(strpos($urlarr['path'],'/') !== 0)
  44                         $complete .= '/';
  45
  46         $complete .= $urlarr["path"];
  47
  48         if (@$urlarr["query"] != "")
  49                 $complete .= "?".$urlarr["query"];
  50
  51         if (@$urlarr["fragment"] != "")
  52                 $complete .= "#".$urlarr["fragment"];
  53
  54         return($complete);
  55 }
  56
  57 function parseurl_getsiteinfo_cached($url, $no_guessing = false, $do_oembed = true) {
  58
  59         if ($url == "")
  60                 return false;
  61
  62         $r = q("SELECT * FROM `parsed_url` WHERE `url` = '%s' AND `guessing` = %d AND `oembed` = %d",
  63                 dbesc(normalise_link($url)), intval(!$no_guessing), intval($do_oembed));
  64
  65         if ($r)
  66                 $data = $r[0]["content"];
  67
  68         if (!is_null($data)) {
  69                 $data = unserialize($data);
  70                 return $data;
  71         }
  72
  73         $data = parseurl_getsiteinfo($url, $no_guessing, $do_oembed);
  74
  75         q("INSERT INTO `parsed_url` (`url`, `guessing`, `oembed`, `content`, `created`) VALUES ('%s', %d, %d, '%s', '%s')
  76                  ON DUPLICATE KEY UPDATE `content` = '%s', `created` = '%s'",
  77                 dbesc(normalise_link($url)), intval(!$no_guessing), intval($do_oembed),
  78                 dbesc(serialize($data)), dbesc(datetime_convert()),
  79                 dbesc(serialize($data)), dbesc(datetime_convert()));
  80
  81         return $data;
  82 }
  83
  84 function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1) {
  85         require_once("include/network.php");
  86         require_once("include/Photo.php");
  87
  88         $a = get_app();
  89
  90         $siteinfo = array();
  91
  92         if ($count > 10) {
  93                 logger("parseurl_getsiteinfo: Endless loop detected for ".$url, LOGGER_DEBUG);
  94                 return($siteinfo);
  95         }
  96
  97         $url = trim($url, "'");
  98         $url = trim($url, '"');
  99
 100         $url = original_url($url);
 101
 102         $siteinfo["url"] = $url;
 103         $siteinfo["type"] = "link";
 104
 105         $stamp1 = microtime(true);
 106
 107         $ch = curl_init();
 108         curl_setopt($ch, CURLOPT_URL, $url);
 109         curl_setopt($ch, CURLOPT_HEADER, 1);
 110         curl_setopt($ch, CURLOPT_NOBODY, 1);
 111         curl_setopt($ch, CURLOPT_TIMEOUT, 3);
 112         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
 113         //curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
 114         curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
 115
 116         $header = curl_exec($ch);
 117         $curl_info = @curl_getinfo($ch);
 118         $http_code = $curl_info['http_code'];
 119         curl_close($ch);
 120
 121         $a->save_timestamp($stamp1, "network");
 122
 123         if ((($curl_info['http_code'] == "301") OR ($curl_info['http_code'] == "302") OR ($curl_info['http_code'] == "303") OR ($curl_info['http_code'] == "307"))
 124                 AND (($curl_info['redirect_url'] != "") OR ($curl_info['location'] != ""))) {
 125                 if ($curl_info['redirect_url'] != "")
 126                         $siteinfo = parseurl_getsiteinfo($curl_info['redirect_url'], $no_guessing, $do_oembed, ++$count);
 127                 else
 128                         $siteinfo = parseurl_getsiteinfo($curl_info['location'], $no_guessing, $do_oembed, ++$count);
 129                 return($siteinfo);
 130         }
 131
 132         // if the file is too large then exit
 133         if ($curl_info["download_content_length"] > 1000000)
 134                 return($siteinfo);
 135
 136         // if it isn't a HTML file then exit
 137         if (($curl_info["content_type"] != "") AND !strstr(strtolower($curl_info["content_type"]),"html"))
 138                 return($siteinfo);
 139
 140         if ($do_oembed) {
 141                 require_once("include/oembed.php");
 142
 143                 $oembed_data = oembed_fetch_url($url);
 144
 145                 if ($oembed_data->type != "error")
 146                         $siteinfo["type"] = $oembed_data->type;
 147
 148                 if (($oembed_data->type == "link") AND ($siteinfo["type"] != "photo")) {
 149                         if (isset($oembed_data->title))
 150                                 $siteinfo["title"] = $oembed_data->title;
 151                         if (isset($oembed_data->description))
 152                                 $siteinfo["text"] = trim($oembed_data->description);
 153                         if (isset($oembed_data->thumbnail_url))
 154                                 $siteinfo["image"] = $oembed_data->thumbnail_url;
 155                 }
 156         }
 157
 158         $stamp1 = microtime(true);
 159
 160         // Now fetch the body as well
 161         $ch = curl_init();
 162         curl_setopt($ch, CURLOPT_URL, $url);
 163         curl_setopt($ch, CURLOPT_HEADER, 1);
 164         curl_setopt($ch, CURLOPT_NOBODY, 0);
 165         curl_setopt($ch, CURLOPT_TIMEOUT, 10);
 166         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
 167         curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
 168
 169         $header = curl_exec($ch);
 170         $curl_info = @curl_getinfo($ch);
 171         $http_code = $curl_info['http_code'];
 172         curl_close($ch);
 173
 174         $a->save_timestamp($stamp1, "network");
 175
 176         // Fetch the first mentioned charset. Can be in body or header
 177         $charset = "";
 178         if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
 179                 $charset = trim(trim(trim(array_pop($matches)), ';,'));
 180
 181         if ($charset == "")
 182                 $charset = "utf-8";
 183
 184         $pos = strpos($header, "\r\n\r\n");
 185
 186         if ($pos)
 187                 $body = trim(substr($header, $pos));
 188         else
 189                 $body = $header;
 190
 191         if (($charset != '') AND (strtoupper($charset) != "UTF-8")) {
 192                 logger("parseurl_getsiteinfo: detected charset ".$charset, LOGGER_DEBUG);
 193                 //$body = mb_convert_encoding($body, "UTF-8", $charset);
 194                 $body = iconv($charset, "UTF-8//TRANSLIT", $body);
 195         }
 196
 197         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
 198
 199         $doc = new DOMDocument();
 200         @$doc->loadHTML($body);
 201
 202         deletenode($doc, 'style');
 203         deletenode($doc, 'script');
 204         deletenode($doc, 'option');
 205         deletenode($doc, 'h1');
 206         deletenode($doc, 'h2');
 207         deletenode($doc, 'h3');
 208         deletenode($doc, 'h4');
 209         deletenode($doc, 'h5');
 210         deletenode($doc, 'h6');
 211         deletenode($doc, 'ol');
 212         deletenode($doc, 'ul');
 213
 214         $xpath = new DomXPath($doc);
 215
 216         $list = $xpath->query("//meta[@content]");
 217         foreach ($list as $node) {
 218                 $attr = array();
 219                 if ($node->attributes->length)
 220                         foreach ($node->attributes as $attribute)
 221                                 $attr[$attribute->name] = $attribute->value;
 222
 223                 if (@$attr["http-equiv"] == 'refresh') {
 224                         $path = $attr["content"];
 225                         $pathinfo = explode(";", $path);
 226                         $content = "";
 227                         foreach ($pathinfo AS $value) {
 228                                 if (substr(strtolower($value), 0, 4) == "url=")
 229                                         $content = substr($value, 4);
 230                         }
 231                         if ($content != "") {
 232                                 $siteinfo = parseurl_getsiteinfo($content, $no_guessing, $do_oembed, ++$count);
 233                                 return($siteinfo);
 234                         }
 235                 }
 236         }
 237
 238         //$list = $xpath->query("head/title");
 239         $list = $xpath->query("//title");
 240         foreach ($list as $node)
 241                 $siteinfo["title"] =  html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
 242
 243         //$list = $xpath->query("head/meta[@name]");
 244         $list = $xpath->query("//meta[@name]");
 245         foreach ($list as $node) {
 246                 $attr = array();
 247                 if ($node->attributes->length)
 248                         foreach ($node->attributes as $attribute)
 249                                 $attr[$attribute->name] = $attribute->value;
 250
 251                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 252
 253                 if ($attr["content"] != "")
 254                         switch (strtolower($attr["name"])) {
 255                                 case "fulltitle":
 256                                         $siteinfo["title"] = $attr["content"];
 257                                         break;
 258                                 case "description":
 259                                         $siteinfo["text"] = $attr["content"];
 260                                         break;
 261                                 case "thumbnail":
 262                                         $siteinfo["image"] = $attr["content"];
 263                                         break;
 264                                 case "twitter:image":
 265                                         $siteinfo["image"] = $attr["content"];
 266                                         break;
 267                                 case "twitter:image:src":
 268                                         $siteinfo["image"] = $attr["content"];
 269                                         break;
 270                                 case "twitter:card":
 271                                         if (($siteinfo["type"] == "") OR ($attr["content"] == "photo"))
 272                                                 $siteinfo["type"] = $attr["content"];
 273                                         break;
 274                                 case "twitter:description":
 275                                         $siteinfo["text"] = $attr["content"];
 276                                         break;
 277                                 case "twitter:title":
 278                                         $siteinfo["title"] = $attr["content"];
 279                                         break;
 280                                 case "dc.title":
 281                                         $siteinfo["title"] = $attr["content"];
 282                                         break;
 283                                 case "dc.description":
 284                                         $siteinfo["text"] = $attr["content"];
 285                                         break;
 286                                 case "keywords":
 287                                         $keywords = explode(",", $attr["content"]);
 288                                         break;
 289                                 case "news_keywords":
 290                                         $keywords = explode(",", $attr["content"]);
 291                                         break;
 292                         }
 293                 if ($siteinfo["type"] == "summary")
 294                         $siteinfo["type"] = "link";
 295         }
 296
 297         if (isset($keywords)) {
 298                 $siteinfo["keywords"] = array();
 299                 foreach ($keywords as $keyword)
 300                         if (!in_array(trim($keyword), $siteinfo["keywords"]))
 301                                 $siteinfo["keywords"][] = trim($keyword);
 302         }
 303
 304         //$list = $xpath->query("head/meta[@property]");
 305         $list = $xpath->query("//meta[@property]");
 306         foreach ($list as $node) {
 307                 $attr = array();
 308                 if ($node->attributes->length)
 309                         foreach ($node->attributes as $attribute)
 310                                 $attr[$attribute->name] = $attribute->value;
 311
 312                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 313
 314                 if ($attr["content"] != "")
 315                         switch (strtolower($attr["property"])) {
 316                                 case "og:image":
 317                                         $siteinfo["image"] = $attr["content"];
 318                                         break;
 319                                 case "og:title":
 320                                         $siteinfo["title"] = $attr["content"];
 321                                         break;
 322                                 case "og:description":
 323                                         $siteinfo["text"] = $attr["content"];
 324                                         break;
 325                         }
 326         }
 327
 328         if ((@$siteinfo["image"] == "") AND !$no_guessing) {
 329             $list = $xpath->query("//img[@src]");
 330             foreach ($list as $node) {
 331                 $attr = array();
 332                 if ($node->attributes->length)
 333                     foreach ($node->attributes as $attribute)
 334                         $attr[$attribute->name] = $attribute->value;
 335
 336                         $src = completeurl($attr["src"], $url);
 337                         $photodata = get_photo_info($src);
 338
 339                         if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
 340                                 if ($photodata[0] > 300) {
 341                                         $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
 342                                         $photodata[0] = 300;
 343                                 }
 344                                 if ($photodata[1] > 300) {
 345                                         $photodata[0] = round($photodata[0] * (300 / $photodata[1]));
 346                                         $photodata[1] = 300;
 347                                 }
 348                                 $siteinfo["images"][] = array("src"=>$src,
 349                                                                 "width"=>$photodata[0],
 350                                                                 "height"=>$photodata[1]);
 351                         }
 352
 353                 }
 354     } elseif ($siteinfo["image"] != "") {
 355                 $src = completeurl($siteinfo["image"], $url);
 356
 357                 unset($siteinfo["image"]);
 358
 359                 $photodata = get_photo_info($src);
 360
 361                 if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
 362                         $siteinfo["images"][] = array("src"=>$src,
 363                                                         "width"=>$photodata[0],
 364                                                         "height"=>$photodata[1]);
 365         }
 366
 367         if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$no_guessing) {
 368                 $text = "";
 369
 370                 $list = $xpath->query("//div[@class='article']");
 371                 foreach ($list as $node)
 372                         if (strlen($node->nodeValue) > 40)
 373                                 $text .= " ".trim($node->nodeValue);
 374
 375                 if ($text == "") {
 376                         $list = $xpath->query("//div[@class='content']");
 377                         foreach ($list as $node)
 378                                 if (strlen($node->nodeValue) > 40)
 379                                         $text .= " ".trim($node->nodeValue);
 380                 }
 381
 382                 // If none text was found then take the paragraph content
 383                 if ($text == "") {
 384                         $list = $xpath->query("//p");
 385                         foreach ($list as $node)
 386                                 if (strlen($node->nodeValue) > 40)
 387                                         $text .= " ".trim($node->nodeValue);
 388                 }
 389
 390                 if ($text != "") {
 391                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 392
 393                         while (strpos($text, "  "))
 394                                 $text = trim(str_replace("  ", " ", $text));
 395
 396                         $siteinfo["text"] = trim(html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...');
 397                 }
 398         }
 399
 400         logger("parseurl_getsiteinfo: Siteinfo for ".$url." ".print_r($siteinfo, true), LOGGER_DEBUG);
 401
 402         call_hooks('getsiteinfo', $siteinfo);
 403
 404         return($siteinfo);
 405 }
 406
 407 function arr_add_hashes(&$item,$k) {
 408         $item = '#' . $item;
 409 }
 410
 411 function parse_url_content(&$a) {
 412
 413         require_once("include/items.php");
 414
 415         $text = null;
 416         $str_tags = '';
 417
 418         $textmode = false;
 419
 420         if(local_user() && (! feature_enabled(local_user(),'richtext')))
 421                 $textmode = true;
 422
 423         //if($textmode)
 424         $br = (($textmode) ? "\n" : '<br />');
 425
 426         if(x($_GET,'binurl'))
 427                 $url = trim(hex2bin($_GET['binurl']));
 428         else
 429                 $url = trim($_GET['url']);
 430
 431         if($_GET['title'])
 432                 $title = strip_tags(trim($_GET['title']));
 433
 434         if($_GET['description'])
 435                 $text = strip_tags(trim($_GET['description']));
 436
 437         if($_GET['tags']) {
 438                 $arr_tags = str_getcsv($_GET['tags']);
 439                 if(count($arr_tags)) {
 440                         array_walk($arr_tags,'arr_add_hashes');
 441                         $str_tags = $br . implode(' ',$arr_tags) . $br;
 442                 }
 443         }
 444
 445         // add url scheme if missing
 446         $arrurl = parse_url($url);
 447         if (!x($arrurl, 'scheme')) {
 448                 if (x($arrurl, 'host'))
 449                         $url = "http:".$url;
 450                 else
 451                         $url = "http://".$url;
 452         }
 453
 454         logger('parse_url: ' . $url);
 455
 456         if($textmode)
 457                 $template = '[bookmark=%s]%s[/bookmark]%s';
 458         else
 459                 $template = "<a class=\"bookmark\" href=\"%s\" >%s</a>%s";
 460
 461         $arr = array('url' => $url, 'text' => '');
 462
 463         call_hooks('parse_link', $arr);
 464
 465         if(strlen($arr['text'])) {
 466                 echo $arr['text'];
 467                 killme();
 468         }
 469
 470
 471         if($url && $title && $text) {
 472
 473                 $title = str_replace(array("\r","\n"),array('',''),$title);
 474
 475                 if($textmode)
 476                         $text = '[quote]' . trim($text) . '[/quote]' . $br;
 477                 else {
 478                         $text = '<blockquote>' . htmlspecialchars(trim($text)) . '</blockquote><br />';
 479                         $title = htmlspecialchars($title);
 480                 }
 481
 482                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 483
 484                 logger('parse_url (unparsed): returns: ' . $result);
 485
 486                 echo $result;
 487                 killme();
 488         }
 489
 490         $siteinfo = parseurl_getsiteinfo($url);
 491
 492         unset($siteinfo["keywords"]);
 493
 494         $info = add_page_info_data($siteinfo);
 495
 496         if (!$textmode)
 497                 // Replace ' with ’ - not perfect - but the richtext editor has problems otherwise
 498                 $info = str_replace(array("&#039;"), array("&#8217;"), $info);
 499
 500         echo $info;
 501
 502         killme();
 503 }
 504 ?>