mod/parse_url.php

   1 <?php
   2 /* To-Do
   3 https://developers.google.com/+/plugins/snippet/
   4
   5 <meta itemprop="name" content="Toller Titel">
   6 <meta itemprop="description" content="Eine tolle Beschreibung">
   7 <meta itemprop="image" content="http://maple.libertreeproject.org/images/tree-icon.png">
   8
   9 <body itemscope itemtype="http://schema.org/Product">
  10   <h1 itemprop="name">Shiny Trinket</h1>
  11   <img itemprop="image" src="{image-url}" />
  12   <p itemprop="description">Shiny trinkets are shiny.</p>
  13 </body>
  14 */
  15
  16 if(!function_exists('deletenode')) {
  17         function deletenode(&$doc, $node)
  18         {
  19                 $xpath = new DomXPath($doc);
  20                 $list = $xpath->query("//".$node);
  21                 foreach ($list as $child)
  22                         $child->parentNode->removeChild($child);
  23         }
  24 }
  25
  26 function completeurl($url, $scheme) {
  27         $urlarr = parse_url($url);
  28
  29         if (isset($urlarr["scheme"]))
  30                 return($url);
  31
  32         $schemearr = parse_url($scheme);
  33
  34         $complete = $schemearr["scheme"]."://".$schemearr["host"];
  35
  36         if (@$schemearr["port"] != "")
  37                 $complete .= ":".$schemearr["port"];
  38
  39                 if(strpos($urlarr['path'],'/') !== 0)
  40                         $complete .= '/';
  41
  42         $complete .= $urlarr["path"];
  43
  44         if (@$urlarr["query"] != "")
  45                 $complete .= "?".$urlarr["query"];
  46
  47         if (@$urlarr["fragment"] != "")
  48                 $complete .= "#".$urlarr["fragment"];
  49
  50         return($complete);
  51 }
  52
  53 function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1) {
  54         require_once("include/network.php");
  55
  56         $a = get_app();
  57
  58         $siteinfo = array();
  59
  60         if ($count > 10) {
  61                 logger("parseurl_getsiteinfo: Endless loop detected for ".$url, LOGGER_DEBUG);
  62                 return($siteinfo);
  63         }
  64
  65         $url = trim($url, "'");
  66         $url = trim($url, '"');
  67
  68         $url = original_url($url);
  69
  70         $siteinfo["url"] = $url;
  71         $siteinfo["type"] = "link";
  72
  73         $stamp1 = microtime(true);
  74
  75         $ch = curl_init();
  76         curl_setopt($ch, CURLOPT_URL, $url);
  77         curl_setopt($ch, CURLOPT_HEADER, 1);
  78         curl_setopt($ch, CURLOPT_NOBODY, 1);
  79         curl_setopt($ch, CURLOPT_TIMEOUT, 3);
  80         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
  81         //curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
  82         curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
  83
  84         $header = curl_exec($ch);
  85         $curl_info = @curl_getinfo($ch);
  86         $http_code = $curl_info['http_code'];
  87         curl_close($ch);
  88
  89         $a->save_timestamp($stamp1, "network");
  90
  91         if ((($curl_info['http_code'] == "301") OR ($curl_info['http_code'] == "302") OR ($curl_info['http_code'] == "303") OR ($curl_info['http_code'] == "307"))
  92                 AND (($curl_info['redirect_url'] != "") OR ($curl_info['location'] != ""))) {
  93                 if ($curl_info['redirect_url'] != "")
  94                         $siteinfo = parseurl_getsiteinfo($curl_info['redirect_url'], $no_guessing, $do_oembed, ++$count);
  95                 else
  96                         $siteinfo = parseurl_getsiteinfo($curl_info['location'], $no_guessing, $do_oembed, ++$count);
  97                 return($siteinfo);
  98         }
  99
 100         if ($do_oembed) {
 101                 require_once("include/oembed.php");
 102
 103                 $oembed_data = oembed_fetch_url($url);
 104
 105                 if ($oembed_data->type != "error")
 106                         $siteinfo["type"] = $oembed_data->type;
 107         }
 108
 109         // if the file is too large then exit
 110         if ($curl_info["download_content_length"] > 1000000)
 111                 return($siteinfo);
 112
 113         // if it isn't a HTML file then exit
 114         if (($curl_info["content_type"] != "") AND !strstr(strtolower($curl_info["content_type"]),"html"))
 115                 return($siteinfo);
 116
 117         $stamp1 = microtime(true);
 118
 119         // Now fetch the body as well
 120         $ch = curl_init();
 121         curl_setopt($ch, CURLOPT_URL, $url);
 122         curl_setopt($ch, CURLOPT_HEADER, 1);
 123         curl_setopt($ch, CURLOPT_NOBODY, 0);
 124         curl_setopt($ch, CURLOPT_TIMEOUT, 10);
 125         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
 126         curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
 127
 128         $header = curl_exec($ch);
 129         $curl_info = @curl_getinfo($ch);
 130         $http_code = $curl_info['http_code'];
 131         curl_close($ch);
 132
 133         $a->save_timestamp($stamp1, "network");
 134
 135         // Fetch the first mentioned charset. Can be in body or header
 136         $charset = "";
 137         if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
 138                 $charset = trim(trim(trim(array_pop($matches)), ';,'));
 139
 140         if ($charset == "")
 141                 $charset = "utf-8";
 142
 143         $pos = strpos($header, "\r\n\r\n");
 144
 145         if ($pos)
 146                 $body = trim(substr($header, $pos));
 147         else
 148                 $body = $header;
 149
 150         if (($charset != '') AND (strtoupper($charset) != "UTF-8")) {
 151                 logger("parseurl_getsiteinfo: detected charset ".$charset, LOGGER_DEBUG);
 152                 //$body = mb_convert_encoding($body, "UTF-8", $charset);
 153                 $body = iconv($charset, "UTF-8//TRANSLIT", $body);
 154         }
 155
 156         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
 157
 158         $doc = new DOMDocument();
 159         @$doc->loadHTML($body);
 160
 161         deletenode($doc, 'style');
 162         deletenode($doc, 'script');
 163         deletenode($doc, 'option');
 164         deletenode($doc, 'h1');
 165         deletenode($doc, 'h2');
 166         deletenode($doc, 'h3');
 167         deletenode($doc, 'h4');
 168         deletenode($doc, 'h5');
 169         deletenode($doc, 'h6');
 170         deletenode($doc, 'ol');
 171         deletenode($doc, 'ul');
 172
 173         $xpath = new DomXPath($doc);
 174
 175         $list = $xpath->query("//meta[@content]");
 176         foreach ($list as $node) {
 177                 $attr = array();
 178                 if ($node->attributes->length)
 179                         foreach ($node->attributes as $attribute)
 180                                 $attr[$attribute->name] = $attribute->value;
 181
 182                 if (@$attr["http-equiv"] == 'refresh') {
 183                         $path = $attr["content"];
 184                         $pathinfo = explode(";", $path);
 185                         $content = "";
 186                         foreach ($pathinfo AS $value) {
 187                                 if (substr(strtolower($value), 0, 4) == "url=")
 188                                         $content = substr($value, 4);
 189                         }
 190                         if ($content != "") {
 191                                 $siteinfo = parseurl_getsiteinfo($content, $no_guessing, $do_oembed, ++$count);
 192                                 return($siteinfo);
 193                         }
 194                 }
 195         }
 196
 197         //$list = $xpath->query("head/title");
 198         $list = $xpath->query("//title");
 199         foreach ($list as $node)
 200                 $siteinfo["title"] =  html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
 201
 202         //$list = $xpath->query("head/meta[@name]");
 203         $list = $xpath->query("//meta[@name]");
 204         foreach ($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                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 211
 212                 if ($attr["content"] != "")
 213                         switch (strtolower($attr["name"])) {
 214                                 case "fulltitle":
 215                                         $siteinfo["title"] = $attr["content"];
 216                                         break;
 217                                 case "description":
 218                                         $siteinfo["text"] = $attr["content"];
 219                                         break;
 220                                 case "thumbnail":
 221                                         $siteinfo["image"] = $attr["content"];
 222                                         break;
 223                                 case "twitter:image":
 224                                         $siteinfo["image"] = $attr["content"];
 225                                         break;
 226                                 case "twitter:image:src":
 227                                         $siteinfo["image"] = $attr["content"];
 228                                         break;
 229                                 case "twitter:card":
 230                                         if (($siteinfo["type"] == "") OR ($attr["content"] == "photo"))
 231                                                 $siteinfo["type"] = $attr["content"];
 232                                         break;
 233                                 case "twitter:description":
 234                                         $siteinfo["text"] = $attr["content"];
 235                                         break;
 236                                 case "twitter:title":
 237                                         $siteinfo["title"] = $attr["content"];
 238                                         break;
 239                                 case "dc.title":
 240                                         $siteinfo["title"] = $attr["content"];
 241                                         break;
 242                                 case "dc.description":
 243                                         $siteinfo["text"] = $attr["content"];
 244                                         break;
 245                                 case "keywords":
 246                                         $keywords = explode(",", $attr["content"]);
 247                                         break;
 248                                 case "news_keywords":
 249                                         $keywords = explode(",", $attr["content"]);
 250                                         break;
 251                         }
 252                 if ($siteinfo["type"] == "summary")
 253                         $siteinfo["type"] = "link";
 254         }
 255
 256         if (isset($keywords)) {
 257                 $siteinfo["keywords"] = array();
 258                 foreach ($keywords as $keyword)
 259                         $siteinfo["keywords"][] = trim($keyword);
 260         }
 261
 262         //$list = $xpath->query("head/meta[@property]");
 263         $list = $xpath->query("//meta[@property]");
 264         foreach ($list as $node) {
 265                 $attr = array();
 266                 if ($node->attributes->length)
 267                         foreach ($node->attributes as $attribute)
 268                                 $attr[$attribute->name] = $attribute->value;
 269
 270                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 271
 272                 if ($attr["content"] != "")
 273                         switch (strtolower($attr["property"])) {
 274                                 case "og:image":
 275                                         $siteinfo["image"] = $attr["content"];
 276                                         break;
 277                                 case "og:title":
 278                                         $siteinfo["title"] = $attr["content"];
 279                                         break;
 280                                 case "og:description":
 281                                         $siteinfo["text"] = $attr["content"];
 282                                         break;
 283                         }
 284         }
 285
 286         if (isset($oembed_data) AND ($oembed_data->type == "link") AND ($siteinfo["type"] != "photo")) {
 287                 if (isset($oembed_data->title) AND (trim($oembed_data->title) != ""))
 288                         $siteinfo["title"] = $oembed_data->title;
 289                 if (isset($oembed_data->description) AND (trim($oembed_data->description) != ""))
 290                         $siteinfo["text"] = trim($oembed_data->description);
 291                 if (isset($oembed_data->thumbnail_url) AND (trim($oembed_data->thumbnail_url) != ""))
 292                         $siteinfo["image"] = $oembed_data->thumbnail_url;
 293         }
 294
 295         if ((@$siteinfo["image"] == "") AND !$no_guessing) {
 296             $list = $xpath->query("//img[@src]");
 297             foreach ($list as $node) {
 298                 $attr = array();
 299                 if ($node->attributes->length)
 300                     foreach ($node->attributes as $attribute)
 301                         $attr[$attribute->name] = $attribute->value;
 302
 303                         $src = completeurl($attr["src"], $url);
 304                         $photodata = @getimagesize($src);
 305
 306                         if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
 307                                 if ($photodata[0] > 300) {
 308                                         $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
 309                                         $photodata[0] = 300;
 310                                 }
 311                                 if ($photodata[1] > 300) {
 312                                         $photodata[0] = round($photodata[0] * (300 / $photodata[1]));
 313                                         $photodata[1] = 300;
 314                                 }
 315                                 $siteinfo["images"][] = array("src"=>$src,
 316                                                                 "width"=>$photodata[0],
 317                                                                 "height"=>$photodata[1]);
 318                         }
 319
 320                 }
 321     } else {
 322                 $src = completeurl($siteinfo["image"], $url);
 323
 324                 unset($siteinfo["image"]);
 325
 326                 $photodata = @getimagesize($src);
 327
 328                 if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
 329                         $siteinfo["images"][] = array("src"=>$src,
 330                                                         "width"=>$photodata[0],
 331                                                         "height"=>$photodata[1]);
 332         }
 333
 334         if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$no_guessing) {
 335                 $text = "";
 336
 337                 $list = $xpath->query("//div[@class='article']");
 338                 foreach ($list as $node)
 339                         if (strlen($node->nodeValue) > 40)
 340                                 $text .= " ".trim($node->nodeValue);
 341
 342                 if ($text == "") {
 343                         $list = $xpath->query("//div[@class='content']");
 344                         foreach ($list as $node)
 345                                 if (strlen($node->nodeValue) > 40)
 346                                         $text .= " ".trim($node->nodeValue);
 347                 }
 348
 349                 // If none text was found then take the paragraph content
 350                 if ($text == "") {
 351                         $list = $xpath->query("//p");
 352                         foreach ($list as $node)
 353                                 if (strlen($node->nodeValue) > 40)
 354                                         $text .= " ".trim($node->nodeValue);
 355                 }
 356
 357                 if ($text != "") {
 358                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 359
 360                         while (strpos($text, "  "))
 361                                 $text = trim(str_replace("  ", " ", $text));
 362
 363                         $siteinfo["text"] = trim(html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...');
 364                 }
 365         }
 366
 367         logger("parseurl_getsiteinfo: Siteinfo for ".$url." ".print_r($siteinfo, true), LOGGER_DEBUG);
 368
 369         call_hooks('getsiteinfo', $siteinfo);
 370
 371         return($siteinfo);
 372 }
 373
 374 function arr_add_hashes(&$item,$k) {
 375         $item = '#' . $item;
 376 }
 377
 378 function parse_url_content(&$a) {
 379
 380         $text = null;
 381         $str_tags = '';
 382
 383         $textmode = false;
 384
 385         if(local_user() && (! feature_enabled(local_user(),'richtext')))
 386                 $textmode = true;
 387
 388         //if($textmode)
 389         $br = (($textmode) ? "\n" : '<br />');
 390
 391         if(x($_GET,'binurl'))
 392                 $url = trim(hex2bin($_GET['binurl']));
 393         else
 394                 $url = trim($_GET['url']);
 395
 396         if($_GET['title'])
 397                 $title = strip_tags(trim($_GET['title']));
 398
 399         if($_GET['description'])
 400                 $text = strip_tags(trim($_GET['description']));
 401
 402         if($_GET['tags']) {
 403                 $arr_tags = str_getcsv($_GET['tags']);
 404                 if(count($arr_tags)) {
 405                         array_walk($arr_tags,'arr_add_hashes');
 406                         $str_tags = $br . implode(' ',$arr_tags) . $br;
 407                 }
 408         }
 409
 410         logger('parse_url: ' . $url);
 411
 412         if($textmode)
 413                 $template = '[bookmark=%s]%s[/bookmark]%s';
 414         else
 415                 $template = "<a class=\"bookmark\" href=\"%s\" >%s</a>%s";
 416
 417         $arr = array('url' => $url, 'text' => '');
 418
 419         call_hooks('parse_link', $arr);
 420
 421         if(strlen($arr['text'])) {
 422                 echo $arr['text'];
 423                 killme();
 424         }
 425
 426
 427         if($url && $title && $text) {
 428
 429                 $title = str_replace(array("\r","\n"),array('',''),$title);
 430
 431                 if($textmode)
 432                         $text = '[quote]' . trim($text) . '[/quote]' . $br;
 433                 else {
 434                         $text = '<blockquote>' . htmlspecialchars(trim($text)) . '</blockquote><br />';
 435                         $title = htmlspecialchars($title);
 436                 }
 437
 438                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 439
 440                 logger('parse_url (unparsed): returns: ' . $result);
 441
 442                 echo $result;
 443                 killme();
 444         }
 445
 446         $siteinfo = parseurl_getsiteinfo($url);
 447
 448 //      if ($textmode) {
 449 //              require_once("include/items.php");
 450 //
 451 //              echo add_page_info_data($siteinfo);
 452 //              killme();
 453 //      }
 454
 455         $url= $siteinfo["url"];
 456
 457         // If the link contains BBCode stuff, make a short link out of this to avoid parsing problems
 458         if (strpos($url, '[') OR strpos($url, ']')) {
 459                 require_once("include/network.php");
 460                 $url = short_link($url);
 461         }
 462
 463         $sitedata = "";
 464
 465         if($siteinfo["title"] == "") {
 466                 $sitedata .= sprintf($template,$url,$url,'') . $str_tags;
 467                 killme();
 468         } else {
 469                 $text = $siteinfo["text"];
 470                 $title = $siteinfo["title"];
 471         }
 472
 473         $image = "";
 474
 475         if (($siteinfo["type"] != "video") AND (sizeof($siteinfo["images"]) > 0)){
 476                 /* Execute below code only if image is present in siteinfo */
 477
 478                 $total_images = 0;
 479                 $max_images = get_config('system','max_bookmark_images');
 480                 if($max_images === false)
 481                         $max_images = 2;
 482                 else
 483                         $max_images = intval($max_images);
 484
 485                 foreach ($siteinfo["images"] as $imagedata) {
 486                         if($textmode)
 487                                 $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]' . "\n";
 488                         else
 489                                 $image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" /><br />';
 490                         $total_images ++;
 491                         if($max_images && $max_images >= $total_images)
 492                                 break;
 493                 }
 494         }
 495
 496         if(strlen($text)) {
 497                 if($textmode)
 498                         $text = '[quote]'.trim($text).'[/quote]';
 499                 else
 500                         $text = '<blockquote>'.htmlspecialchars(trim($text)).'</blockquote>';
 501         }
 502
 503         if($image)
 504                 $text = $br.$br.$image.$text;
 505         else
 506                 $text = $br.$text;
 507
 508         $title = str_replace(array("\r","\n"),array('',''),$title);
 509
 510         $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 511
 512         logger('parse_url: returns: ' . $result);
 513
 514         $sitedata .=  trim($result);
 515
 516         if (($siteinfo["type"] == "video") AND ($url != ""))
 517                 echo "[class=type-video]".$sitedata."[/class]";
 518         elseif (($siteinfo["type"] != "photo"))
 519                 echo "[class=type-link]".$sitedata."[/class]";
 520         else
 521                 echo "[class=type-photo]".$title.$br.$image."[/class]";
 522
 523         killme();
 524 }
 525 ?>