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