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