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         $data = Cache::get("parse_url:".$no_guessing.":".$do_oembed.":".$url);
  60         if (!is_null($data)) {
  61                 $data = unserialize($data);
  62                 return $data;
  63         }
  64
  65         $data = parseurl_getsiteinfo($url, $no_guessing, $do_oembed);
  66
  67         Cache::set("parse_url:".$no_guessing.":".$do_oembed.":".$url,serialize($data), CACHE_DAY);
  68
  69         return $data;
  70 }
  71
  72 function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1) {
  73         require_once("include/network.php");
  74
  75         $a = get_app();
  76
  77         $siteinfo = array();
  78
  79         if ($count > 10) {
  80                 logger("parseurl_getsiteinfo: Endless loop detected for ".$url, LOGGER_DEBUG);
  81                 return($siteinfo);
  82         }
  83
  84         $url = trim($url, "'");
  85         $url = trim($url, '"');
  86
  87         $url = original_url($url);
  88
  89         $siteinfo["url"] = $url;
  90         $siteinfo["type"] = "link";
  91
  92         $stamp1 = microtime(true);
  93
  94         $ch = curl_init();
  95         curl_setopt($ch, CURLOPT_URL, $url);
  96         curl_setopt($ch, CURLOPT_HEADER, 1);
  97         curl_setopt($ch, CURLOPT_NOBODY, 1);
  98         curl_setopt($ch, CURLOPT_TIMEOUT, 3);
  99         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
 100         //curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
 101         curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
 102
 103         $header = curl_exec($ch);
 104         $curl_info = @curl_getinfo($ch);
 105         $http_code = $curl_info['http_code'];
 106         curl_close($ch);
 107
 108         $a->save_timestamp($stamp1, "network");
 109
 110         if ((($curl_info['http_code'] == "301") OR ($curl_info['http_code'] == "302") OR ($curl_info['http_code'] == "303") OR ($curl_info['http_code'] == "307"))
 111                 AND (($curl_info['redirect_url'] != "") OR ($curl_info['location'] != ""))) {
 112                 if ($curl_info['redirect_url'] != "")
 113                         $siteinfo = parseurl_getsiteinfo($curl_info['redirect_url'], $no_guessing, $do_oembed, ++$count);
 114                 else
 115                         $siteinfo = parseurl_getsiteinfo($curl_info['location'], $no_guessing, $do_oembed, ++$count);
 116                 return($siteinfo);
 117         }
 118
 119         // if the file is too large then exit
 120         if ($curl_info["download_content_length"] > 1000000)
 121                 return($siteinfo);
 122
 123         // if it isn't a HTML file then exit
 124         if (($curl_info["content_type"] != "") AND !strstr(strtolower($curl_info["content_type"]),"html"))
 125                 return($siteinfo);
 126
 127         if ($do_oembed) {
 128                 require_once("include/oembed.php");
 129
 130                 $oembed_data = oembed_fetch_url($url);
 131
 132                 if ($oembed_data->type != "error")
 133                         $siteinfo["type"] = $oembed_data->type;
 134
 135                 if (($oembed_data->type == "link") AND ($siteinfo["type"] != "photo")) {
 136                         if (isset($oembed_data->title))
 137                                 $siteinfo["title"] = $oembed_data->title;
 138                         if (isset($oembed_data->description))
 139                                 $siteinfo["text"] = trim($oembed_data->description);
 140                         if (isset($oembed_data->thumbnail_url))
 141                                 $siteinfo["image"] = $oembed_data->thumbnail_url;
 142                 }
 143         }
 144
 145         $stamp1 = microtime(true);
 146
 147         // Now fetch the body as well
 148         $ch = curl_init();
 149         curl_setopt($ch, CURLOPT_URL, $url);
 150         curl_setopt($ch, CURLOPT_HEADER, 1);
 151         curl_setopt($ch, CURLOPT_NOBODY, 0);
 152         curl_setopt($ch, CURLOPT_TIMEOUT, 10);
 153         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
 154         curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
 155
 156         $header = curl_exec($ch);
 157         $curl_info = @curl_getinfo($ch);
 158         $http_code = $curl_info['http_code'];
 159         curl_close($ch);
 160
 161         $a->save_timestamp($stamp1, "network");
 162
 163         // Fetch the first mentioned charset. Can be in body or header
 164         $charset = "";
 165         if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
 166                 $charset = trim(trim(trim(array_pop($matches)), ';,'));
 167
 168         if ($charset == "")
 169                 $charset = "utf-8";
 170
 171         $pos = strpos($header, "\r\n\r\n");
 172
 173         if ($pos)
 174                 $body = trim(substr($header, $pos));
 175         else
 176                 $body = $header;
 177
 178         if (($charset != '') AND (strtoupper($charset) != "UTF-8")) {
 179                 logger("parseurl_getsiteinfo: detected charset ".$charset, LOGGER_DEBUG);
 180                 //$body = mb_convert_encoding($body, "UTF-8", $charset);
 181                 $body = iconv($charset, "UTF-8//TRANSLIT", $body);
 182         }
 183
 184         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
 185
 186         $doc = new DOMDocument();
 187         @$doc->loadHTML($body);
 188
 189         deletenode($doc, 'style');
 190         deletenode($doc, 'script');
 191         deletenode($doc, 'option');
 192         deletenode($doc, 'h1');
 193         deletenode($doc, 'h2');
 194         deletenode($doc, 'h3');
 195         deletenode($doc, 'h4');
 196         deletenode($doc, 'h5');
 197         deletenode($doc, 'h6');
 198         deletenode($doc, 'ol');
 199         deletenode($doc, 'ul');
 200
 201         $xpath = new DomXPath($doc);
 202
 203         $list = $xpath->query("//meta[@content]");
 204         foreach ($list as $node) {
 205                 $attr = array();
 206                 if ($node->attributes->length)
 207                         foreach ($node->attributes as $attribute)
 208                                 $attr[$attribute->name] = $attribute->value;
 209
 210                 if (@$attr["http-equiv"] == 'refresh') {
 211                         $path = $attr["content"];
 212                         $pathinfo = explode(";", $path);
 213                         $content = "";
 214                         foreach ($pathinfo AS $value) {
 215                                 if (substr(strtolower($value), 0, 4) == "url=")
 216                                         $content = substr($value, 4);
 217                         }
 218                         if ($content != "") {
 219                                 $siteinfo = parseurl_getsiteinfo($content, $no_guessing, $do_oembed, ++$count);
 220                                 return($siteinfo);
 221                         }
 222                 }
 223         }
 224
 225         //$list = $xpath->query("head/title");
 226         $list = $xpath->query("//title");
 227         foreach ($list as $node)
 228                 $siteinfo["title"] =  html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
 229
 230         //$list = $xpath->query("head/meta[@name]");
 231         $list = $xpath->query("//meta[@name]");
 232         foreach ($list as $node) {
 233                 $attr = array();
 234                 if ($node->attributes->length)
 235                         foreach ($node->attributes as $attribute)
 236                                 $attr[$attribute->name] = $attribute->value;
 237
 238                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 239
 240                 if ($attr["content"] != "")
 241                         switch (strtolower($attr["name"])) {
 242                                 case "fulltitle":
 243                                         $siteinfo["title"] = $attr["content"];
 244                                         break;
 245                                 case "description":
 246                                         $siteinfo["text"] = $attr["content"];
 247                                         break;
 248                                 case "thumbnail":
 249                                         $siteinfo["image"] = $attr["content"];
 250                                         break;
 251                                 case "twitter:image":
 252                                         $siteinfo["image"] = $attr["content"];
 253                                         break;
 254                                 case "twitter:image:src":
 255                                         $siteinfo["image"] = $attr["content"];
 256                                         break;
 257                                 case "twitter:card":
 258                                         if (($siteinfo["type"] == "") OR ($attr["content"] == "photo"))
 259                                                 $siteinfo["type"] = $attr["content"];
 260                                         break;
 261                                 case "twitter:description":
 262                                         $siteinfo["text"] = $attr["content"];
 263                                         break;
 264                                 case "twitter:title":
 265                                         $siteinfo["title"] = $attr["content"];
 266                                         break;
 267                                 case "dc.title":
 268                                         $siteinfo["title"] = $attr["content"];
 269                                         break;
 270                                 case "dc.description":
 271                                         $siteinfo["text"] = $attr["content"];
 272                                         break;
 273                                 case "keywords":
 274                                         $keywords = explode(",", $attr["content"]);
 275                                         break;
 276                                 case "news_keywords":
 277                                         $keywords = explode(",", $attr["content"]);
 278                                         break;
 279                         }
 280                 if ($siteinfo["type"] == "summary")
 281                         $siteinfo["type"] = "link";
 282         }
 283
 284         if (isset($keywords)) {
 285                 $siteinfo["keywords"] = array();
 286                 foreach ($keywords as $keyword)
 287                         if (!in_array(trim($keyword), $siteinfo["keywords"]))
 288                                 $siteinfo["keywords"][] = trim($keyword);
 289         }
 290
 291         //$list = $xpath->query("head/meta[@property]");
 292         $list = $xpath->query("//meta[@property]");
 293         foreach ($list as $node) {
 294                 $attr = array();
 295                 if ($node->attributes->length)
 296                         foreach ($node->attributes as $attribute)
 297                                 $attr[$attribute->name] = $attribute->value;
 298
 299                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 300
 301                 if ($attr["content"] != "")
 302                         switch (strtolower($attr["property"])) {
 303                                 case "og:image":
 304                                         $siteinfo["image"] = $attr["content"];
 305                                         break;
 306                                 case "og:title":
 307                                         $siteinfo["title"] = $attr["content"];
 308                                         break;
 309                                 case "og:description":
 310                                         $siteinfo["text"] = $attr["content"];
 311                                         break;
 312                         }
 313         }
 314
 315         if ((@$siteinfo["image"] == "") AND !$no_guessing) {
 316             $list = $xpath->query("//img[@src]");
 317             foreach ($list as $node) {
 318                 $attr = array();
 319                 if ($node->attributes->length)
 320                     foreach ($node->attributes as $attribute)
 321                         $attr[$attribute->name] = $attribute->value;
 322
 323                         $src = completeurl($attr["src"], $url);
 324                         $photodata = @getimagesize($src);
 325
 326                         if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
 327                                 if ($photodata[0] > 300) {
 328                                         $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
 329                                         $photodata[0] = 300;
 330                                 }
 331                                 if ($photodata[1] > 300) {
 332                                         $photodata[0] = round($photodata[0] * (300 / $photodata[1]));
 333                                         $photodata[1] = 300;
 334                                 }
 335                                 $siteinfo["images"][] = array("src"=>$src,
 336                                                                 "width"=>$photodata[0],
 337                                                                 "height"=>$photodata[1]);
 338                         }
 339
 340                 }
 341     } else {
 342                 $src = completeurl($siteinfo["image"], $url);
 343
 344                 unset($siteinfo["image"]);
 345
 346                 $photodata = @getimagesize($src);
 347
 348                 if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
 349                         $siteinfo["images"][] = array("src"=>$src,
 350                                                         "width"=>$photodata[0],
 351                                                         "height"=>$photodata[1]);
 352         }
 353
 354         if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$no_guessing) {
 355                 $text = "";
 356
 357                 $list = $xpath->query("//div[@class='article']");
 358                 foreach ($list as $node)
 359                         if (strlen($node->nodeValue) > 40)
 360                                 $text .= " ".trim($node->nodeValue);
 361
 362                 if ($text == "") {
 363                         $list = $xpath->query("//div[@class='content']");
 364                         foreach ($list as $node)
 365                                 if (strlen($node->nodeValue) > 40)
 366                                         $text .= " ".trim($node->nodeValue);
 367                 }
 368
 369                 // If none text was found then take the paragraph content
 370                 if ($text == "") {
 371                         $list = $xpath->query("//p");
 372                         foreach ($list as $node)
 373                                 if (strlen($node->nodeValue) > 40)
 374                                         $text .= " ".trim($node->nodeValue);
 375                 }
 376
 377                 if ($text != "") {
 378                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 379
 380                         while (strpos($text, "  "))
 381                                 $text = trim(str_replace("  ", " ", $text));
 382
 383                         $siteinfo["text"] = trim(html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...');
 384                 }
 385         }
 386
 387         logger("parseurl_getsiteinfo: Siteinfo for ".$url." ".print_r($siteinfo, true), LOGGER_DEBUG);
 388
 389         call_hooks('getsiteinfo', $siteinfo);
 390
 391         return($siteinfo);
 392 }
 393
 394 function arr_add_hashes(&$item,$k) {
 395         $item = '#' . $item;
 396 }
 397
 398 function parse_url_content(&$a) {
 399
 400         $text = null;
 401         $str_tags = '';
 402
 403         $textmode = false;
 404
 405         if(local_user() && (! feature_enabled(local_user(),'richtext')))
 406                 $textmode = true;
 407
 408         //if($textmode)
 409         $br = (($textmode) ? "\n" : '<br />');
 410
 411         if(x($_GET,'binurl'))
 412                 $url = trim(hex2bin($_GET['binurl']));
 413         else
 414                 $url = trim($_GET['url']);
 415
 416         if($_GET['title'])
 417                 $title = strip_tags(trim($_GET['title']));
 418
 419         if($_GET['description'])
 420                 $text = strip_tags(trim($_GET['description']));
 421
 422         if($_GET['tags']) {
 423                 $arr_tags = str_getcsv($_GET['tags']);
 424                 if(count($arr_tags)) {
 425                         array_walk($arr_tags,'arr_add_hashes');
 426                         $str_tags = $br . implode(' ',$arr_tags) . $br;
 427                 }
 428         }
 429
 430         // add url scheme if missing
 431         $arrurl = parse_url($url);
 432         if (!x($arrurl, 'scheme')) {
 433                 if (x($arrurl, 'host'))
 434                         $url = "http:".$url;
 435                 else
 436                         $url = "http://".$url;
 437         }
 438
 439         logger('parse_url: ' . $url);
 440
 441         if($textmode)
 442                 $template = '[bookmark=%s]%s[/bookmark]%s';
 443         else
 444                 $template = "<a class=\"bookmark\" href=\"%s\" >%s</a>%s";
 445
 446         $arr = array('url' => $url, 'text' => '');
 447
 448         call_hooks('parse_link', $arr);
 449
 450         if(strlen($arr['text'])) {
 451                 echo $arr['text'];
 452                 killme();
 453         }
 454
 455
 456         if($url && $title && $text) {
 457
 458                 $title = str_replace(array("\r","\n"),array('',''),$title);
 459
 460                 if($textmode)
 461                         $text = '[quote]' . trim($text) . '[/quote]' . $br;
 462                 else {
 463                         $text = '<blockquote>' . htmlspecialchars(trim($text)) . '</blockquote><br />';
 464                         $title = htmlspecialchars($title);
 465                 }
 466
 467                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 468
 469                 logger('parse_url (unparsed): returns: ' . $result);
 470
 471                 echo $result;
 472                 killme();
 473         }
 474
 475         $siteinfo = parseurl_getsiteinfo($url);
 476
 477 //      if ($textmode) {
 478 //              require_once("include/items.php");
 479 //
 480 //              echo add_page_info_data($siteinfo);
 481 //              killme();
 482 //      }
 483
 484         $url= $siteinfo["url"];
 485
 486         // If the link contains BBCode stuff, make a short link out of this to avoid parsing problems
 487         if (strpos($url, '[') OR strpos($url, ']')) {
 488                 require_once("include/network.php");
 489                 $url = short_link($url);
 490         }
 491
 492         $sitedata = "";
 493
 494         if($siteinfo["title"] != "") {
 495                 $text = $siteinfo["text"];
 496                 $title = $siteinfo["title"];
 497         }
 498
 499         $image = "";
 500
 501         if (($siteinfo["type"] != "video") AND (sizeof($siteinfo["images"]) > 0)){
 502                 /* Execute below code only if image is present in siteinfo */
 503
 504                 $total_images = 0;
 505                 $max_images = get_config('system','max_bookmark_images');
 506                 if($max_images === false)
 507                         $max_images = 2;
 508                 else
 509                         $max_images = intval($max_images);
 510
 511                 foreach ($siteinfo["images"] as $imagedata) {
 512                         if($textmode)
 513                                 $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]' . "\n";
 514                         else
 515                                 $image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" /><br />';
 516                         $total_images ++;
 517                         if($max_images && $max_images >= $total_images)
 518                                 break;
 519                 }
 520         }
 521
 522         if(strlen($text)) {
 523                 if($textmode)
 524                         $text = '[quote]'.trim($text).'[/quote]';
 525                 else
 526                         $text = '<blockquote>'.htmlspecialchars(trim($text)).'</blockquote>';
 527         }
 528
 529         if($image)
 530                 $text = $br.$br.$image.$text;
 531         else
 532                 $text = $br.$text;
 533
 534         $title = str_replace(array("\r","\n"),array('',''),$title);
 535
 536         $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 537
 538         logger('parse_url: returns: ' . $result);
 539
 540         $sitedata .=  trim($result);
 541
 542         if (($siteinfo["type"] == "video") AND ($url != ""))
 543                 echo "[class=type-video]".$sitedata."[/class]";
 544         elseif (($siteinfo["type"] != "photo"))
 545                 echo "[class=type-link]".$sitedata."[/class]";
 546         else
 547                 echo "[class=type-photo]".$title.$br.$image."[/class]";
 548
 549         killme();
 550 }
 551 ?>