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
  55         $a = get_app();
  56
  57         $siteinfo = array();
  58
  59         if ($count > 10) {
  60                 logger("parseurl_getsiteinfo: Endless loop detected for ".$url, LOGGER_DEBUG);
  61                 return($siteinfo);
  62         }
  63
  64         $url = trim($url, "'");
  65         $url = trim($url, '"');
  66         $siteinfo["url"] = $url;
  67         $siteinfo["type"] = "link";
  68
  69         $ch = curl_init();
  70         curl_setopt($ch, CURLOPT_URL, $url);
  71         curl_setopt($ch, CURLOPT_HEADER, 1);
  72         curl_setopt($ch, CURLOPT_NOBODY, 0);
  73         curl_setopt($ch, CURLOPT_TIMEOUT, 3);
  74         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
  75         //curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
  76         curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
  77
  78         $header = curl_exec($ch);
  79         $curl_info = @curl_getinfo($ch);
  80         $http_code = $curl_info['http_code'];
  81         curl_close($ch);
  82
  83         if ((($curl_info['http_code'] == "301") OR ($curl_info['http_code'] == "302") OR ($curl_info['http_code'] == "303") OR ($curl_info['http_code'] == "307"))
  84                 AND (($curl_info['redirect_url'] != "") OR ($curl_info['location'] != ""))) {
  85                 if ($curl_info['redirect_url'] != "")
  86                         $siteinfo = parseurl_getsiteinfo($curl_info['redirect_url'], $no_guessing, $do_oembed, ++$count);
  87                 else
  88                         $siteinfo = parseurl_getsiteinfo($curl_info['location'], $no_guessing, $do_oembed, ++$count);
  89                 return($siteinfo);
  90         }
  91
  92         if ($do_oembed) {
  93                 require_once("include/oembed.php");
  94
  95                 $oembed_data = oembed_fetch_url($url);
  96
  97                 if ($oembed_data->type != "error")
  98                         $siteinfo["type"] = $oembed_data->type;
  99         }
 100
 101         // Fetch the first mentioned charset. Can be in body or header
 102         $charset = "";
 103         if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
 104                 $charset = trim(trim(trim(array_pop($matches)), ';,'));
 105
 106         if ($charset == "")
 107                 $charset = "utf-8";
 108
 109         $pos = strpos($header, "\r\n\r\n");
 110
 111         if ($pos)
 112                 $body = trim(substr($header, $pos));
 113         else
 114                 $body = $header;
 115
 116         if (($charset != '') AND (strtoupper($charset) != "UTF-8")) {
 117                 logger("parseurl_getsiteinfo: detected charset ".$charset, LOGGER_DEBUG);
 118                 //$body = mb_convert_encoding($body, "UTF-8", $charset);
 119                 $body = iconv($charset, "UTF-8//TRANSLIT", $body);
 120         }
 121
 122         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
 123
 124         $doc = new DOMDocument();
 125         @$doc->loadHTML($body);
 126
 127         deletenode($doc, 'style');
 128         deletenode($doc, 'script');
 129         deletenode($doc, 'option');
 130         deletenode($doc, 'h1');
 131         deletenode($doc, 'h2');
 132         deletenode($doc, 'h3');
 133         deletenode($doc, 'h4');
 134         deletenode($doc, 'h5');
 135         deletenode($doc, 'h6');
 136         deletenode($doc, 'ol');
 137         deletenode($doc, 'ul');
 138
 139         $xpath = new DomXPath($doc);
 140
 141         $list = $xpath->query("//meta[@content]");
 142         foreach ($list as $node) {
 143                 $attr = array();
 144                 if ($node->attributes->length)
 145                         foreach ($node->attributes as $attribute)
 146                                 $attr[$attribute->name] = $attribute->value;
 147
 148                 if (@$attr["http-equiv"] == 'refresh') {
 149                         $path = $attr["content"];
 150                         $pathinfo = explode(";", $path);
 151                         $content = "";
 152                         foreach ($pathinfo AS $value) {
 153                                 if (substr(strtolower($value), 0, 4) == "url=")
 154                                         $content = substr($value, 4);
 155                         }
 156                         if ($content != "") {
 157                                 $siteinfo = parseurl_getsiteinfo($content, $no_guessing, $do_oembed, ++$count);
 158                                 return($siteinfo);
 159                         }
 160                 }
 161         }
 162
 163         //$list = $xpath->query("head/title");
 164         $list = $xpath->query("//title");
 165         foreach ($list as $node)
 166                 $siteinfo["title"] =  html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
 167
 168         //$list = $xpath->query("head/meta[@name]");
 169         $list = $xpath->query("//meta[@name]");
 170         foreach ($list as $node) {
 171                 $attr = array();
 172                 if ($node->attributes->length)
 173                         foreach ($node->attributes as $attribute)
 174                                 $attr[$attribute->name] = $attribute->value;
 175
 176                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 177
 178                 if ($attr["content"] != "")
 179                         switch (strtolower($attr["name"])) {
 180                                 case "fulltitle":
 181                                         $siteinfo["title"] = $attr["content"];
 182                                         break;
 183                                 case "description":
 184                                         $siteinfo["text"] = $attr["content"];
 185                                         break;
 186                                 case "twitter:image":
 187                                         $siteinfo["image"] = $attr["content"];
 188                                         break;
 189                                 case "twitter:image:src":
 190                                         $siteinfo["image"] = $attr["content"];
 191                                         break;
 192                                 case "twitter:card":
 193                                         if (($siteinfo["type"] == "") OR ($attr["content"] == "photo"))
 194                                                 $siteinfo["type"] = $attr["content"];
 195                                         break;
 196                                 case "twitter:description":
 197                                         $siteinfo["text"] = $attr["content"];
 198                                         break;
 199                                 case "twitter:title":
 200                                         $siteinfo["title"] = $attr["content"];
 201                                         break;
 202                                 case "dc.title":
 203                                         $siteinfo["title"] = $attr["content"];
 204                                         break;
 205                                 case "dc.description":
 206                                         $siteinfo["text"] = $attr["content"];
 207                                         break;
 208                         }
 209                 if ($siteinfo["type"] == "summary")
 210                         $siteinfo["type"] = "link";
 211         }
 212
 213         //$list = $xpath->query("head/meta[@property]");
 214         $list = $xpath->query("//meta[@property]");
 215         foreach ($list as $node) {
 216                 $attr = array();
 217                 if ($node->attributes->length)
 218                         foreach ($node->attributes as $attribute)
 219                                 $attr[$attribute->name] = $attribute->value;
 220
 221                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 222
 223                 if ($attr["content"] != "")
 224                         switch (strtolower($attr["property"])) {
 225                                 case "og:image":
 226                                         $siteinfo["image"] = $attr["content"];
 227                                         break;
 228                                 case "og:title":
 229                                         $siteinfo["title"] = $attr["content"];
 230                                         break;
 231                                 case "og:description":
 232                                         $siteinfo["text"] = $attr["content"];
 233                                         break;
 234                         }
 235         }
 236
 237         if (isset($oembed_data) AND ($oembed_data->type == "link") AND ($siteinfo["type"] != "photo")) {
 238                 if (isset($oembed_data->title) AND (trim($oembed_data->title) != ""))
 239                         $siteinfo["title"] = $oembed_data->title;
 240                 if (isset($oembed_data->description) AND (trim($oembed_data->description) != ""))
 241                         $siteinfo["text"] = trim($oembed_data->description);
 242                 if (isset($oembed_data->thumbnail_url) AND (trim($oembed_data->thumbnail_url) != ""))
 243                         $siteinfo["image"] = $oembed_data->thumbnail_url;
 244         }
 245
 246         if ((@$siteinfo["image"] == "") AND !$no_guessing) {
 247             $list = $xpath->query("//img[@src]");
 248             foreach ($list as $node) {
 249                 $attr = array();
 250                 if ($node->attributes->length)
 251                     foreach ($node->attributes as $attribute)
 252                         $attr[$attribute->name] = $attribute->value;
 253
 254                         $src = completeurl($attr["src"], $url);
 255                         $photodata = @getimagesize($src);
 256
 257                         if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
 258                                 if ($photodata[0] > 300) {
 259                                         $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
 260                                         $photodata[0] = 300;
 261                                 }
 262                                 if ($photodata[1] > 300) {
 263                                         $photodata[0] = round($photodata[0] * (300 / $photodata[1]));
 264                                         $photodata[1] = 300;
 265                                 }
 266                                 $siteinfo["images"][] = array("src"=>$src,
 267                                                                 "width"=>$photodata[0],
 268                                                                 "height"=>$photodata[1]);
 269                         }
 270
 271                 }
 272     } else {
 273                 $src = completeurl($siteinfo["image"], $url);
 274
 275                 unset($siteinfo["image"]);
 276
 277                 $photodata = @getimagesize($src);
 278
 279                 if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
 280                         $siteinfo["images"][] = array("src"=>$src,
 281                                                         "width"=>$photodata[0],
 282                                                         "height"=>$photodata[1]);
 283         }
 284
 285         if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$no_guessing) {
 286                 $text = "";
 287
 288                 $list = $xpath->query("//div[@class='article']");
 289                 foreach ($list as $node)
 290                         if (strlen($node->nodeValue) > 40)
 291                                 $text .= " ".trim($node->nodeValue);
 292
 293                 if ($text == "") {
 294                         $list = $xpath->query("//div[@class='content']");
 295                         foreach ($list as $node)
 296                                 if (strlen($node->nodeValue) > 40)
 297                                         $text .= " ".trim($node->nodeValue);
 298                 }
 299
 300                 // If none text was found then take the paragraph content
 301                 if ($text == "") {
 302                         $list = $xpath->query("//p");
 303                         foreach ($list as $node)
 304                                 if (strlen($node->nodeValue) > 40)
 305                                         $text .= " ".trim($node->nodeValue);
 306                 }
 307
 308                 if ($text != "") {
 309                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 310
 311                         while (strpos($text, "  "))
 312                                 $text = trim(str_replace("  ", " ", $text));
 313
 314                         $siteinfo["text"] = trim(html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...');
 315                 }
 316         }
 317
 318         logger("parseurl_getsiteinfo: Siteinfo for ".$url." ".print_r($siteinfo, true), LOGGER_DEBUG);
 319
 320         return($siteinfo);
 321 }
 322
 323 function arr_add_hashes(&$item,$k) {
 324         $item = '#' . $item;
 325 }
 326
 327 function parse_url_content(&$a) {
 328
 329         $text = null;
 330         $str_tags = '';
 331
 332         $textmode = false;
 333
 334         if(local_user() && (! feature_enabled(local_user(),'richtext')))
 335                 $textmode = true;
 336
 337         //if($textmode)
 338         $br = (($textmode) ? "\n" : '<br />');
 339
 340         if(x($_GET,'binurl'))
 341                 $url = trim(hex2bin($_GET['binurl']));
 342         else
 343                 $url = trim($_GET['url']);
 344
 345         if($_GET['title'])
 346                 $title = strip_tags(trim($_GET['title']));
 347
 348         if($_GET['description'])
 349                 $text = strip_tags(trim($_GET['description']));
 350
 351         if($_GET['tags']) {
 352                 $arr_tags = str_getcsv($_GET['tags']);
 353                 if(count($arr_tags)) {
 354                         array_walk($arr_tags,'arr_add_hashes');
 355                         $str_tags = $br . implode(' ',$arr_tags) . $br;
 356                 }
 357         }
 358
 359         logger('parse_url: ' . $url);
 360
 361         if($textmode)
 362                 $template = '[bookmark=%s]%s[/bookmark]%s';
 363         else
 364                 $template = "<a class=\"bookmark\" href=\"%s\" >%s</a>%s";
 365
 366         $arr = array('url' => $url, 'text' => '');
 367
 368         call_hooks('parse_link', $arr);
 369
 370         if(strlen($arr['text'])) {
 371                 echo $arr['text'];
 372                 killme();
 373         }
 374
 375
 376         if($url && $title && $text) {
 377
 378                 $title = str_replace(array("\r","\n"),array('',''),$title);
 379
 380                 if($textmode)
 381                         $text = '[quote]' . trim($text) . '[/quote]' . $br;
 382                 else {
 383                         $text = '<blockquote>' . htmlspecialchars(trim($text)) . '</blockquote><br />';
 384                         $title = htmlspecialchars($title);
 385                 }
 386
 387                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 388
 389                 logger('parse_url (unparsed): returns: ' . $result);
 390
 391                 echo $result;
 392                 killme();
 393         }
 394
 395         $siteinfo = parseurl_getsiteinfo($url);
 396
 397         $url= $siteinfo["url"];
 398
 399         $sitedata = "";
 400
 401         if($siteinfo["title"] == "") {
 402                 $sitedata .= sprintf($template,$url,$url,'') . $str_tags;
 403                 killme();
 404         } else {
 405                 $text = $siteinfo["text"];
 406                 $title = $siteinfo["title"];
 407         }
 408
 409         $image = "";
 410
 411         if (($siteinfo["type"] != "video") AND (sizeof($siteinfo["images"]) > 0)){
 412                 /* Execute below code only if image is present in siteinfo */
 413
 414                 $total_images = 0;
 415                 $max_images = get_config('system','max_bookmark_images');
 416                 if($max_images === false)
 417                         $max_images = 2;
 418                 else
 419                         $max_images = intval($max_images);
 420
 421                 foreach ($siteinfo["images"] as $imagedata) {
 422                         if($textmode)
 423                                 $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]' . "\n";
 424                         else
 425                                 $image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" /><br />';
 426                         $total_images ++;
 427                         if($max_images && $max_images >= $total_images)
 428                                 break;
 429                 }
 430         }
 431
 432         if(strlen($text)) {
 433                 if($textmode)
 434                         $text = '[quote]'.trim($text).'[/quote]';
 435                 else
 436                         $text = '<blockquote>'.htmlspecialchars(trim($text)).'</blockquote>';
 437         }
 438
 439         if($image)
 440                 $text = $br.$br.$image.$text;
 441         else
 442                 $text = $br.$text;
 443
 444         $title = str_replace(array("\r","\n"),array('',''),$title);
 445
 446         $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 447
 448         logger('parse_url: returns: ' . $result);
 449
 450         $sitedata .=  trim($result);
 451
 452         if (($siteinfo["type"] == "video") AND ($url != ""))
 453                 echo "[class=type-video]".$sitedata."[/class]";
 454         elseif (($siteinfo["type"] != "photo"))
 455                 echo "[class=type-link]".$sitedata."[/class]";
 456         else
 457                 echo "[class=type-photo]".$title.$br.$image."[/class]";
 458
 459         killme();
 460 }
 461 ?>