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         $siteinfo = array();
  56
  57         if ($count > 10) {
  58                 logger("parseurl_getsiteinfo: Endless loop detected for ".$url, LOGGER_DEBUG);
  59                 return($siteinfo);
  60         }
  61
  62         $url = trim($url, "'");
  63         $url = trim($url, '"');
  64         $siteinfo["url"] = $url;
  65         $siteinfo["type"] = "link";
  66
  67         $ch = curl_init();
  68         curl_setopt($ch, CURLOPT_URL, $url);
  69         curl_setopt($ch, CURLOPT_HEADER, 1);
  70         curl_setopt($ch, CURLOPT_NOBODY, 0);
  71         curl_setopt($ch, CURLOPT_TIMEOUT, 3);
  72         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
  73         //curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
  74         //curl_setopt($ch,CURLOPT_USERAGENT,' Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:24.0) Gecko/20100101 Firefox/24.0');
  75         curl_setopt($ch,CURLOPT_USERAGENT, "Mozilla/5.0 (compatible; Friendica)");
  76
  77         $header = curl_exec($ch);
  78         $curl_info = @curl_getinfo($ch);
  79         $http_code = $curl_info['http_code'];
  80         curl_close($ch);
  81
  82         if ((($curl_info['http_code'] == "301") OR ($curl_info['http_code'] == "302") OR ($curl_info['http_code'] == "303") OR ($curl_info['http_code'] == "307"))
  83                 AND (($curl_info['redirect_url'] != "") OR ($curl_info['location'] != ""))) {
  84                 if ($curl_info['redirect_url'] != "")
  85                         $siteinfo = parseurl_getsiteinfo($curl_info['redirect_url'], $no_guessing, $do_oembed, ++$count);
  86                 else
  87                         $siteinfo = parseurl_getsiteinfo($curl_info['location'], $no_guessing, $do_oembed, ++$count);
  88                 return($siteinfo);
  89         }
  90
  91         if ($do_oembed) {
  92                 require_once("include/oembed.php");
  93
  94                 $oembed_data = oembed_fetch_url($url);
  95
  96                 if ($oembed_data->type != "error")
  97                         $siteinfo["type"] = $oembed_data->type;
  98         }
  99
 100         // Fetch the first mentioned charset. Can be in body or header
 101         $charset = "";
 102         if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
 103                 $charset = trim(trim(trim(array_pop($matches)), ';,'));
 104
 105         if ($charset == "")
 106                 $charset = "utf-8";
 107
 108         $pos = strpos($header, "\r\n\r\n");
 109
 110         if ($pos)
 111                 $body = trim(substr($header, $pos));
 112         else
 113                 $body = $header;
 114
 115         if (($charset != '') AND (strtoupper($charset) != "UTF-8")) {
 116                 logger("parseurl_getsiteinfo: detected charset ".$charset, LOGGER_DEBUG);
 117                 //$body = mb_convert_encoding($body, "UTF-8", $charset);
 118                 $body = iconv($charset, "UTF-8//TRANSLIT", $body);
 119         }
 120
 121         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
 122
 123         $doc = new DOMDocument();
 124         @$doc->loadHTML($body);
 125
 126         deletenode($doc, 'style');
 127         deletenode($doc, 'script');
 128         deletenode($doc, 'option');
 129         deletenode($doc, 'h1');
 130         deletenode($doc, 'h2');
 131         deletenode($doc, 'h3');
 132         deletenode($doc, 'h4');
 133         deletenode($doc, 'h5');
 134         deletenode($doc, 'h6');
 135         deletenode($doc, 'ol');
 136         deletenode($doc, 'ul');
 137
 138         $xpath = new DomXPath($doc);
 139
 140         $list = $xpath->query("//meta[@content]");
 141         foreach ($list as $node) {
 142                 $attr = array();
 143                 if ($node->attributes->length)
 144                         foreach ($node->attributes as $attribute)
 145                                 $attr[$attribute->name] = $attribute->value;
 146
 147                 if (@$attr["http-equiv"] == 'refresh') {
 148                         $path = $attr["content"];
 149                         $pathinfo = explode(";", $path);
 150                         $content = "";
 151                         foreach ($pathinfo AS $value) {
 152                                 if (substr(strtolower($value), 0, 4) == "url=")
 153                                         $content = substr($value, 4);
 154                         }
 155                         if ($content != "") {
 156                                 $siteinfo = parseurl_getsiteinfo($content, $no_guessing, $do_oembed, ++$count);
 157                                 return($siteinfo);
 158                         }
 159                 }
 160         }
 161
 162         //$list = $xpath->query("head/title");
 163         $list = $xpath->query("//title");
 164         foreach ($list as $node)
 165                 $siteinfo["title"] =  html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
 166
 167         //$list = $xpath->query("head/meta[@name]");
 168         $list = $xpath->query("//meta[@name]");
 169         foreach ($list as $node) {
 170                 $attr = array();
 171                 if ($node->attributes->length)
 172                         foreach ($node->attributes as $attribute)
 173                                 $attr[$attribute->name] = $attribute->value;
 174
 175                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 176
 177                 if ($attr["content"] != "")
 178                         switch (strtolower($attr["name"])) {
 179                                 case "fulltitle":
 180                                         $siteinfo["title"] = $attr["content"];
 181                                         break;
 182                                 case "description":
 183                                         $siteinfo["text"] = $attr["content"];
 184                                         break;
 185                                 case "twitter:image":
 186                                         $siteinfo["image"] = $attr["content"];
 187                                         break;
 188                                 case "twitter:card":
 189                                         if (($siteinfo["type"] == "") OR ($attr["content"] == "photo"))
 190                                                 $siteinfo["type"] = $attr["content"];
 191                                         break;
 192                                 case "twitter:description":
 193                                         $siteinfo["text"] = $attr["content"];
 194                                         break;
 195                                 case "twitter:title":
 196                                         $siteinfo["title"] = $attr["content"];
 197                                         break;
 198                                 case "dc.title":
 199                                         $siteinfo["title"] = $attr["content"];
 200                                         break;
 201                                 case "dc.description":
 202                                         $siteinfo["text"] = $attr["content"];
 203                                         break;
 204                         }
 205                 if ($siteinfo["type"] == "summary")
 206                         $siteinfo["type"] = "link";
 207         }
 208
 209         //$list = $xpath->query("head/meta[@property]");
 210         $list = $xpath->query("//meta[@property]");
 211         foreach ($list as $node) {
 212                 $attr = array();
 213                 if ($node->attributes->length)
 214                         foreach ($node->attributes as $attribute)
 215                                 $attr[$attribute->name] = $attribute->value;
 216
 217                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 218
 219                 if ($attr["content"] != "")
 220                         switch (strtolower($attr["property"])) {
 221                                 case "og:image":
 222                                         $siteinfo["image"] = $attr["content"];
 223                                         break;
 224                                 case "og:title":
 225                                         $siteinfo["title"] = $attr["content"];
 226                                         break;
 227                                 case "og:description":
 228                                         $siteinfo["text"] = $attr["content"];
 229                                         break;
 230                         }
 231         }
 232
 233         if (isset($oembed_data) AND ($oembed_data->type == "link") AND ($siteinfo["type"] != "photo")) {
 234                 if (isset($oembed_data->title) AND (trim($oembed_data->title) != ""))
 235                         $siteinfo["title"] = $oembed_data->title;
 236                 if (isset($oembed_data->description) AND (trim($oembed_data->description) != ""))
 237                         $siteinfo["text"] = trim($oembed_data->description);
 238                 if (isset($oembed_data->thumbnail_url) AND (trim($oembed_data->thumbnail_url) != ""))
 239                         $siteinfo["image"] = $oembed_data->thumbnail_url;
 240         }
 241
 242         if ((@$siteinfo["image"] == "") AND !$no_guessing) {
 243             $list = $xpath->query("//img[@src]");
 244             foreach ($list as $node) {
 245                 $attr = array();
 246                 if ($node->attributes->length)
 247                     foreach ($node->attributes as $attribute)
 248                         $attr[$attribute->name] = $attribute->value;
 249
 250                         $src = completeurl($attr["src"], $url);
 251                         $photodata = @getimagesize($src);
 252
 253                         if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
 254                                 if ($photodata[0] > 300) {
 255                                         $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
 256                                         $photodata[0] = 300;
 257                                 }
 258                                 if ($photodata[1] > 300) {
 259                                         $photodata[0] = round($photodata[0] * (300 / $photodata[1]));
 260                                         $photodata[1] = 300;
 261                                 }
 262                                 $siteinfo["images"][] = array("src"=>$src,
 263                                                                 "width"=>$photodata[0],
 264                                                                 "height"=>$photodata[1]);
 265                         }
 266
 267                 }
 268     } else {
 269                 $src = completeurl($siteinfo["image"], $url);
 270
 271                 unset($siteinfo["image"]);
 272
 273                 $photodata = @getimagesize($src);
 274
 275                 if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
 276                         $siteinfo["images"][] = array("src"=>$src,
 277                                                         "width"=>$photodata[0],
 278                                                         "height"=>$photodata[1]);
 279         }
 280
 281         if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$no_guessing) {
 282                 $text = "";
 283
 284                 $list = $xpath->query("//div[@class='article']");
 285                 foreach ($list as $node)
 286                         if (strlen($node->nodeValue) > 40)
 287                                 $text .= " ".trim($node->nodeValue);
 288
 289                 if ($text == "") {
 290                         $list = $xpath->query("//div[@class='content']");
 291                         foreach ($list as $node)
 292                                 if (strlen($node->nodeValue) > 40)
 293                                         $text .= " ".trim($node->nodeValue);
 294                 }
 295
 296                 // If none text was found then take the paragraph content
 297                 if ($text == "") {
 298                         $list = $xpath->query("//p");
 299                         foreach ($list as $node)
 300                                 if (strlen($node->nodeValue) > 40)
 301                                         $text .= " ".trim($node->nodeValue);
 302                 }
 303
 304                 if ($text != "") {
 305                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 306
 307                         while (strpos($text, "  "))
 308                                 $text = trim(str_replace("  ", " ", $text));
 309
 310                         $siteinfo["text"] = trim(html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...');
 311                 }
 312         }
 313
 314         logger("parseurl_getsiteinfo: Siteinfo for ".$url." ".print_r($siteinfo, true), LOGGER_DEBUG);
 315
 316         return($siteinfo);
 317 }
 318
 319 function arr_add_hashes(&$item,$k) {
 320         $item = '#' . $item;
 321 }
 322
 323 function parse_url_content(&$a) {
 324
 325         $text = null;
 326         $str_tags = '';
 327
 328         $textmode = false;
 329
 330         if(local_user() && (! feature_enabled(local_user(),'richtext')))
 331                 $textmode = true;
 332
 333         //if($textmode)
 334         $br = (($textmode) ? "\n" : '<br />');
 335
 336         if(x($_GET,'binurl'))
 337                 $url = trim(hex2bin($_GET['binurl']));
 338         else
 339                 $url = trim($_GET['url']);
 340
 341         if($_GET['title'])
 342                 $title = strip_tags(trim($_GET['title']));
 343
 344         if($_GET['description'])
 345                 $text = strip_tags(trim($_GET['description']));
 346
 347         if($_GET['tags']) {
 348                 $arr_tags = str_getcsv($_GET['tags']);
 349                 if(count($arr_tags)) {
 350                         array_walk($arr_tags,'arr_add_hashes');
 351                         $str_tags = $br . implode(' ',$arr_tags) . $br;
 352                 }
 353         }
 354
 355         logger('parse_url: ' . $url);
 356
 357         if($textmode)
 358                 $template = '[bookmark=%s]%s[/bookmark]%s';
 359         else
 360                 $template = "<a class=\"bookmark\" href=\"%s\" >%s</a>%s";
 361
 362         $arr = array('url' => $url, 'text' => '');
 363
 364         call_hooks('parse_link', $arr);
 365
 366         if(strlen($arr['text'])) {
 367                 echo $arr['text'];
 368                 killme();
 369         }
 370
 371
 372         if($url && $title && $text) {
 373
 374                 $title = str_replace(array("\r","\n"),array('',''),$title);
 375
 376                 if($textmode)
 377                         $text = '[quote]' . trim($text) . '[/quote]' . $br;
 378                 else {
 379                         $text = '<blockquote>' . htmlspecialchars(trim($text)) . '</blockquote><br />';
 380                         $title = htmlspecialchars($title);
 381                 }
 382
 383                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 384
 385                 logger('parse_url (unparsed): returns: ' . $result);
 386
 387                 echo $result;
 388                 killme();
 389         }
 390
 391         $siteinfo = parseurl_getsiteinfo($url);
 392
 393         $url= $siteinfo["url"];
 394
 395         $sitedata = "";
 396
 397         if($siteinfo["title"] == "") {
 398                 $sitedata .= sprintf($template,$url,$url,'') . $str_tags;
 399                 killme();
 400         } else {
 401                 $text = $siteinfo["text"];
 402                 $title = $siteinfo["title"];
 403         }
 404
 405         $image = "";
 406
 407         if (($siteinfo["type"] != "video") AND (sizeof($siteinfo["images"]) > 0)){
 408                 /* Execute below code only if image is present in siteinfo */
 409
 410                 $total_images = 0;
 411                 $max_images = get_config('system','max_bookmark_images');
 412                 if($max_images === false)
 413                         $max_images = 2;
 414                 else
 415                         $max_images = intval($max_images);
 416
 417                 foreach ($siteinfo["images"] as $imagedata) {
 418                         if($textmode)
 419                                 $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]' . "\n";
 420                         else
 421                                 $image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" /><br />';
 422                         $total_images ++;
 423                         if($max_images && $max_images >= $total_images)
 424                                 break;
 425                 }
 426         }
 427
 428         if(strlen($text)) {
 429                 if($textmode)
 430                         $text = '[quote]'.trim($text).'[/quote]';
 431                 else
 432                         $text = '<blockquote>'.htmlspecialchars(trim($text)).'</blockquote>';
 433         }
 434
 435         if($image)
 436                 $text = $br.$br.$image.$text;
 437         else
 438                 $text = $br.$text;
 439
 440         $title = str_replace(array("\r","\n"),array('',''),$title);
 441
 442         $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 443
 444         logger('parse_url: returns: ' . $result);
 445
 446         $sitedata .=  trim($result);
 447
 448         if (($siteinfo["type"] == "video") AND ($url != ""))
 449                 echo "[class=type-video]".$sitedata."[/class]";
 450         elseif (($siteinfo["type"] != "photo"))
 451                 echo "[class=type-link]".$sitedata."[/class]";
 452         else
 453                 echo "[class=type-photo]".$title.$br.$image."[/class]";
 454
 455         killme();
 456 }
 457 ?>