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) {
  54
  55         $siteinfo = array();
  56
  57         $url = trim($url, "'");
  58         $url = trim($url, '"');
  59         $siteinfo["url"] = $url;
  60         $siteinfo["type"] = "link";
  61
  62         $ch = curl_init();
  63         curl_setopt($ch, CURLOPT_URL, $url);
  64         curl_setopt($ch, CURLOPT_HEADER, 1);
  65         curl_setopt($ch, CURLOPT_NOBODY, 0);
  66         curl_setopt($ch, CURLOPT_TIMEOUT, 3);
  67         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
  68         //curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
  69         curl_setopt($ch,CURLOPT_USERAGENT,' Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:24.0) Gecko/20100101 Firefox/24.0');
  70
  71         $header = curl_exec($ch);
  72         $curl_info = @curl_getinfo($ch);
  73         $http_code = $curl_info['http_code'];
  74         curl_close($ch);
  75
  76         if ((($curl_info['http_code'] == "301") OR ($curl_info['http_code'] == "302") OR ($curl_info['http_code'] == "303") OR ($curl_info['http_code'] == "307"))
  77                 AND (($curl_info['redirect_url'] != "") OR ($curl_info['location'] != ""))) {
  78                 if ($curl_info['redirect_url'] != "")
  79                         $siteinfo = parseurl_getsiteinfo($curl_info['redirect_url']);
  80                 else
  81                         $siteinfo = parseurl_getsiteinfo($curl_info['location']);
  82                 return($siteinfo);
  83         }
  84
  85         if ($do_oembed) {
  86                 require_once("include/oembed.php");
  87
  88                 $oembed_data = oembed_fetch_url($url);
  89
  90                 if ($oembed_data->type != "error")
  91                         $siteinfo["type"] = $oembed_data->type;
  92         }
  93
  94         // Fetch the first mentioned charset. Can be in body or header
  95         $charset = "";
  96         if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
  97                 $charset = trim(array_pop($matches));
  98
  99         if ($charset == "")
 100                 $charset = "utf-8";
 101
 102         $pos = strpos($header, "\r\n\r\n");
 103
 104         if ($pos)
 105                 $body = trim(substr($header, $pos));
 106         else
 107                 $body = $header;
 108
 109         $body = mb_convert_encoding($body, "UTF-8", $charset);
 110         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
 111
 112         $doc = new DOMDocument();
 113         @$doc->loadHTML($body);
 114
 115         deletenode($doc, 'style');
 116         deletenode($doc, 'script');
 117         deletenode($doc, 'option');
 118         deletenode($doc, 'h1');
 119         deletenode($doc, 'h2');
 120         deletenode($doc, 'h3');
 121         deletenode($doc, 'h4');
 122         deletenode($doc, 'h5');
 123         deletenode($doc, 'h6');
 124         deletenode($doc, 'ol');
 125         deletenode($doc, 'ul');
 126
 127         $xpath = new DomXPath($doc);
 128
 129         $list = $xpath->query("//meta[@content]");
 130         foreach ($list as $node) {
 131                 $attr = array();
 132                 if ($node->attributes->length)
 133                         foreach ($node->attributes as $attribute)
 134                                 $attr[$attribute->name] = $attribute->value;
 135
 136                 if (@$attr["http-equiv"] == 'refresh') {
 137                         $path = $attr["content"];
 138                         $pathinfo = explode(";", $path);
 139                         $content = "";
 140                         foreach ($pathinfo AS $value) {
 141                                 if (substr(strtolower($value), 0, 4) == "url=")
 142                                         $content = substr($value, 4);
 143                         }
 144                         if ($content != "") {
 145                                 $siteinfo = parseurl_getsiteinfo($content);
 146                                 return($siteinfo);
 147                         }
 148                 }
 149         }
 150
 151         //$list = $xpath->query("head/title");
 152         $list = $xpath->query("//title");
 153         foreach ($list as $node)
 154                 $siteinfo["title"] =  html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
 155
 156         //$list = $xpath->query("head/meta[@name]");
 157         $list = $xpath->query("//meta[@name]");
 158         foreach ($list as $node) {
 159                 $attr = array();
 160                 if ($node->attributes->length)
 161                         foreach ($node->attributes as $attribute)
 162                                 $attr[$attribute->name] = $attribute->value;
 163
 164                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 165
 166                 if ($attr["content"] != "")
 167                         switch (strtolower($attr["name"])) {
 168                                 case "fulltitle":
 169                                         $siteinfo["title"] = $attr["content"];
 170                                         break;
 171                                 case "description":
 172                                         $siteinfo["text"] = $attr["content"];
 173                                         break;
 174                                 case "twitter:image":
 175                                         $siteinfo["image"] = $attr["content"];
 176                                         break;
 177                                 case "twitter:card":
 178                                         if (($siteinfo["type"] == "") OR ($attr["content"] == "photo"))
 179                                                 $siteinfo["type"] = $attr["content"];
 180                                         break;
 181                                 case "twitter:description":
 182                                         $siteinfo["text"] = $attr["content"];
 183                                         break;
 184                                 case "twitter:title":
 185                                         $siteinfo["title"] = $attr["content"];
 186                                         break;
 187                                 case "dc.title":
 188                                         $siteinfo["title"] = $attr["content"];
 189                                         break;
 190                                 case "dc.description":
 191                                         $siteinfo["text"] = $attr["content"];
 192                                         break;
 193                         }
 194                 if ($siteinfo["type"] == "summary")
 195                         $siteinfo["type"] = "link";
 196         }
 197
 198         //$list = $xpath->query("head/meta[@property]");
 199         $list = $xpath->query("//meta[@property]");
 200         foreach ($list as $node) {
 201                 $attr = array();
 202                 if ($node->attributes->length)
 203                         foreach ($node->attributes as $attribute)
 204                                 $attr[$attribute->name] = $attribute->value;
 205
 206                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 207
 208                 if ($attr["content"] != "")
 209                         switch (strtolower($attr["property"])) {
 210                                 case "og:image":
 211                                         $siteinfo["image"] = $attr["content"];
 212                                         break;
 213                                 case "og:title":
 214                                         $siteinfo["title"] = $attr["content"];
 215                                         break;
 216                                 case "og:description":
 217                                         $siteinfo["text"] = $attr["content"];
 218                                         break;
 219                         }
 220         }
 221
 222         if (isset($oembed_data) AND ($oembed_data->type == "link")) {
 223                 if (isset($oembed_data->title) AND (trim($oembed_data->title) != ""))
 224                         $siteinfo["title"] = $oembed_data->title;
 225                 if (isset($oembed_data->description) AND (trim($oembed_data->description) != ""))
 226                         $siteinfo["text"] = trim($oembed_data->description);
 227                 if (isset($oembed_data->thumbnail_url) AND (trim($oembed_data->thumbnail_url) != ""))
 228                         $siteinfo["image"] = $oembed_data->thumbnail_url;
 229         }
 230
 231         if ((@$siteinfo["image"] == "") AND !$no_guessing) {
 232             $list = $xpath->query("//img[@src]");
 233             foreach ($list as $node) {
 234                 $attr = array();
 235                 if ($node->attributes->length)
 236                     foreach ($node->attributes as $attribute)
 237                         $attr[$attribute->name] = $attribute->value;
 238
 239                         $src = completeurl($attr["src"], $url);
 240                         $photodata = @getimagesize($src);
 241
 242                         if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
 243                                 if ($photodata[0] > 300) {
 244                                         $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
 245                                         $photodata[0] = 300;
 246                                 }
 247                                 if ($photodata[1] > 300) {
 248                                         $photodata[0] = round($photodata[0] * (300 / $photodata[1]));
 249                                         $photodata[1] = 300;
 250                                 }
 251                                 $siteinfo["images"][] = array("src"=>$src,
 252                                                                 "width"=>$photodata[0],
 253                                                                 "height"=>$photodata[1]);
 254                         }
 255
 256                 }
 257     } else {
 258                 $src = completeurl($siteinfo["image"], $url);
 259
 260                 unset($siteinfo["image"]);
 261
 262                 $photodata = @getimagesize($src);
 263
 264                 if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
 265                         $siteinfo["images"][] = array("src"=>$src,
 266                                                         "width"=>$photodata[0],
 267                                                         "height"=>$photodata[1]);
 268         }
 269
 270         if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$no_guessing) {
 271                 $text = "";
 272
 273                 $list = $xpath->query("//div[@class='article']");
 274                 foreach ($list as $node)
 275                         if (strlen($node->nodeValue) > 40)
 276                                 $text .= " ".trim($node->nodeValue);
 277
 278                 if ($text == "") {
 279                         $list = $xpath->query("//div[@class='content']");
 280                         foreach ($list as $node)
 281                                 if (strlen($node->nodeValue) > 40)
 282                                         $text .= " ".trim($node->nodeValue);
 283                 }
 284
 285                 // If none text was found then take the paragraph content
 286                 if ($text == "") {
 287                         $list = $xpath->query("//p");
 288                         foreach ($list as $node)
 289                                 if (strlen($node->nodeValue) > 40)
 290                                         $text .= " ".trim($node->nodeValue);
 291                 }
 292
 293                 if ($text != "") {
 294                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 295
 296                         while (strpos($text, "  "))
 297                                 $text = trim(str_replace("  ", " ", $text));
 298
 299                         $siteinfo["text"] = trim(html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...');
 300                 }
 301         }
 302
 303         logger("parseurl_getsiteinfo: Siteinfo for ".$url." ".print_r($siteinfo, true), LOGGER_DEBUG);
 304
 305         return($siteinfo);
 306 }
 307
 308 function arr_add_hashes(&$item,$k) {
 309         $item = '#' . $item;
 310 }
 311
 312 function parse_url_content(&$a) {
 313
 314         $text = null;
 315         $str_tags = '';
 316
 317         $textmode = false;
 318
 319         if(local_user() && (! feature_enabled(local_user(),'richtext')))
 320                 $textmode = true;
 321
 322         //if($textmode)
 323         $br = (($textmode) ? "\n" : '<br />');
 324
 325         if(x($_GET,'binurl'))
 326                 $url = trim(hex2bin($_GET['binurl']));
 327         else
 328                 $url = trim($_GET['url']);
 329
 330         if($_GET['title'])
 331                 $title = strip_tags(trim($_GET['title']));
 332
 333         if($_GET['description'])
 334                 $text = strip_tags(trim($_GET['description']));
 335
 336         if($_GET['tags']) {
 337                 $arr_tags = str_getcsv($_GET['tags']);
 338                 if(count($arr_tags)) {
 339                         array_walk($arr_tags,'arr_add_hashes');
 340                         $str_tags = $br . implode(' ',$arr_tags) . $br;
 341                 }
 342         }
 343
 344         logger('parse_url: ' . $url);
 345
 346         if($textmode)
 347                 $template = '[bookmark=%s]%s[/bookmark]%s';
 348         else
 349                 $template = "<a class=\"bookmark\" href=\"%s\" >%s</a>%s";
 350
 351         $arr = array('url' => $url, 'text' => '');
 352
 353         call_hooks('parse_link', $arr);
 354
 355         if(strlen($arr['text'])) {
 356                 echo $arr['text'];
 357                 killme();
 358         }
 359
 360
 361         if($url && $title && $text) {
 362
 363                 $title = str_replace(array("\r","\n"),array('',''),$title);
 364
 365                 if($textmode)
 366                         $text = '[quote]' . trim($text) . '[/quote]' . $br;
 367                 else {
 368                         $text = '<blockquote>' . htmlspecialchars(trim($text)) . '</blockquote><br />';
 369                         $title = htmlspecialchars($title);
 370                 }
 371
 372                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 373
 374                 logger('parse_url (unparsed): returns: ' . $result);
 375
 376                 echo $result;
 377                 killme();
 378         }
 379
 380         $siteinfo = parseurl_getsiteinfo($url);
 381
 382         $url= $siteinfo["url"];
 383
 384         $sitedata = "";
 385
 386         if($siteinfo["title"] == "") {
 387                 $sitedata .= sprintf($template,$url,$url,'') . $str_tags;
 388                 killme();
 389         } else {
 390                 $text = $siteinfo["text"];
 391                 $title = $siteinfo["title"];
 392         }
 393
 394         $image = "";
 395
 396         if (($siteinfo["type"] != "video") AND (sizeof($siteinfo["images"]) > 0)){
 397                 /* Execute below code only if image is present in siteinfo */
 398
 399                 $total_images = 0;
 400                 $max_images = get_config('system','max_bookmark_images');
 401                 if($max_images === false)
 402                         $max_images = 2;
 403                 else
 404                         $max_images = intval($max_images);
 405
 406                 foreach ($siteinfo["images"] as $imagedata) {
 407                         if($textmode)
 408                                 $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]' . "\n";
 409                         else
 410                                 $image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" /><br />';
 411                         $total_images ++;
 412                         if($max_images && $max_images >= $total_images)
 413                                 break;
 414                 }
 415         }
 416
 417         if(strlen($text)) {
 418                 if($textmode)
 419                         $text = '[quote]'.trim($text).'[/quote]';
 420                 else
 421                         $text = '<blockquote>'.htmlspecialchars(trim($text)).'</blockquote>';
 422         }
 423
 424         if($image)
 425                 $text = $br.$br.$image.$text;
 426         else
 427                 $text = $br.$text;
 428
 429         $title = str_replace(array("\r","\n"),array('',''),$title);
 430
 431         $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 432
 433         logger('parse_url: returns: ' . $result);
 434
 435         $sitedata .=  trim($result);
 436
 437         if (($siteinfo["type"] == "video") AND ($url != ""))
 438                 echo "[class=type-video]".$sitedata."[/class]";
 439         elseif (($siteinfo["type"] != "photo"))
 440                 echo "[class=type-link]".$sitedata."[/class]";
 441         else
 442                 echo "[class=type-photo]".$title.$br.$image."[/class]";
 443
 444         killme();
 445 }
 446 ?>