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($complete);
  51 }
  52
  53 function parseurl_getsiteinfo_cached($url, $no_guessing = false, $do_oembed = true) {
  54
  55         $data = Cache::get("parse_url:".$no_guessing.":".$do_oembed.":".$url);
  56         if (!is_null($data)) {
  57                 $data = unserialize($data);
  58                 return $data;
  59         }
  60
  61         $data = parseurl_getsiteinfo($url, $no_guessing, $do_oembed);
  62
  63         Cache::set("parse_url:".$no_guessing.":".$do_oembed.":".$url,serialize($data), CACHE_DAY);
  64
  65         return $data;
  66 }
  67
  68 function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1) {
  69         require_once("include/network.php");
  70
  71         $a = get_app();
  72
  73         $siteinfo = array();
  74
  75         if ($count > 10) {
  76                 logger("parseurl_getsiteinfo: Endless loop detected for ".$url, LOGGER_DEBUG);
  77                 return($siteinfo);
  78         }
  79
  80         $url = trim($url, "'");
  81         $url = trim($url, '"');
  82
  83         $url = original_url($url);
  84
  85         $siteinfo["url"] = $url;
  86         $siteinfo["type"] = "link";
  87
  88         $stamp1 = microtime(true);
  89
  90         $ch = curl_init();
  91         curl_setopt($ch, CURLOPT_URL, $url);
  92         curl_setopt($ch, CURLOPT_HEADER, 1);
  93         curl_setopt($ch, CURLOPT_NOBODY, 1);
  94         curl_setopt($ch, CURLOPT_TIMEOUT, 3);
  95         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
  96         //curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
  97         curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
  98
  99         $header = curl_exec($ch);
 100         $curl_info = @curl_getinfo($ch);
 101         $http_code = $curl_info['http_code'];
 102         curl_close($ch);
 103
 104         $a->save_timestamp($stamp1, "network");
 105
 106         if ((($curl_info['http_code'] == "301") OR ($curl_info['http_code'] == "302") OR ($curl_info['http_code'] == "303") OR ($curl_info['http_code'] == "307"))
 107                 AND (($curl_info['redirect_url'] != "") OR ($curl_info['location'] != ""))) {
 108                 if ($curl_info['redirect_url'] != "")
 109                         $siteinfo = parseurl_getsiteinfo($curl_info['redirect_url'], $no_guessing, $do_oembed, ++$count);
 110                 else
 111                         $siteinfo = parseurl_getsiteinfo($curl_info['location'], $no_guessing, $do_oembed, ++$count);
 112                 return($siteinfo);
 113         }
 114
 115         // if the file is too large then exit
 116         if ($curl_info["download_content_length"] > 1000000)
 117                 return($siteinfo);
 118
 119         // if it isn't a HTML file then exit
 120         if (($curl_info["content_type"] != "") AND !strstr(strtolower($curl_info["content_type"]),"html"))
 121                 return($siteinfo);
 122
 123         if ($do_oembed) {
 124                 require_once("include/oembed.php");
 125
 126                 $oembed_data = oembed_fetch_url($url);
 127
 128                 if ($oembed_data->type != "error")
 129                         $siteinfo["type"] = $oembed_data->type;
 130
 131                 if (($oembed_data->type == "link") AND ($siteinfo["type"] != "photo")) {
 132                         if (isset($oembed_data->title))
 133                                 $siteinfo["title"] = $oembed_data->title;
 134                         if (isset($oembed_data->description))
 135                                 $siteinfo["text"] = trim($oembed_data->description);
 136                         if (isset($oembed_data->thumbnail_url))
 137                                 $siteinfo["image"] = $oembed_data->thumbnail_url;
 138                 }
 139         }
 140
 141         $stamp1 = microtime(true);
 142
 143         // Now fetch the body as well
 144         $ch = curl_init();
 145         curl_setopt($ch, CURLOPT_URL, $url);
 146         curl_setopt($ch, CURLOPT_HEADER, 1);
 147         curl_setopt($ch, CURLOPT_NOBODY, 0);
 148         curl_setopt($ch, CURLOPT_TIMEOUT, 10);
 149         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
 150         curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
 151
 152         $header = curl_exec($ch);
 153         $curl_info = @curl_getinfo($ch);
 154         $http_code = $curl_info['http_code'];
 155         curl_close($ch);
 156
 157         $a->save_timestamp($stamp1, "network");
 158
 159         // Fetch the first mentioned charset. Can be in body or header
 160         $charset = "";
 161         if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
 162                 $charset = trim(trim(trim(array_pop($matches)), ';,'));
 163
 164         if ($charset == "")
 165                 $charset = "utf-8";
 166
 167         $pos = strpos($header, "\r\n\r\n");
 168
 169         if ($pos)
 170                 $body = trim(substr($header, $pos));
 171         else
 172                 $body = $header;
 173
 174         if (($charset != '') AND (strtoupper($charset) != "UTF-8")) {
 175                 logger("parseurl_getsiteinfo: detected charset ".$charset, LOGGER_DEBUG);
 176                 //$body = mb_convert_encoding($body, "UTF-8", $charset);
 177                 $body = iconv($charset, "UTF-8//TRANSLIT", $body);
 178         }
 179
 180         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
 181
 182         $doc = new DOMDocument();
 183         @$doc->loadHTML($body);
 184
 185         deletenode($doc, 'style');
 186         deletenode($doc, 'script');
 187         deletenode($doc, 'option');
 188         deletenode($doc, 'h1');
 189         deletenode($doc, 'h2');
 190         deletenode($doc, 'h3');
 191         deletenode($doc, 'h4');
 192         deletenode($doc, 'h5');
 193         deletenode($doc, 'h6');
 194         deletenode($doc, 'ol');
 195         deletenode($doc, 'ul');
 196
 197         $xpath = new DomXPath($doc);
 198
 199         $list = $xpath->query("//meta[@content]");
 200         foreach ($list as $node) {
 201                 $attr = array();
 202                 if ($node->attributes->length)
 203                         foreach ($node->attributes as $attribute)
 204                                 $attr[$attribute->name] = $attribute->value;
 205
 206                 if (@$attr["http-equiv"] == 'refresh') {
 207                         $path = $attr["content"];
 208                         $pathinfo = explode(";", $path);
 209                         $content = "";
 210                         foreach ($pathinfo AS $value) {
 211                                 if (substr(strtolower($value), 0, 4) == "url=")
 212                                         $content = substr($value, 4);
 213                         }
 214                         if ($content != "") {
 215                                 $siteinfo = parseurl_getsiteinfo($content, $no_guessing, $do_oembed, ++$count);
 216                                 return($siteinfo);
 217                         }
 218                 }
 219         }
 220
 221         //$list = $xpath->query("head/title");
 222         $list = $xpath->query("//title");
 223         foreach ($list as $node)
 224                 $siteinfo["title"] =  html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
 225
 226         //$list = $xpath->query("head/meta[@name]");
 227         $list = $xpath->query("//meta[@name]");
 228         foreach ($list as $node) {
 229                 $attr = array();
 230                 if ($node->attributes->length)
 231                         foreach ($node->attributes as $attribute)
 232                                 $attr[$attribute->name] = $attribute->value;
 233
 234                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 235
 236                 if ($attr["content"] != "")
 237                         switch (strtolower($attr["name"])) {
 238                                 case "fulltitle":
 239                                         $siteinfo["title"] = $attr["content"];
 240                                         break;
 241                                 case "description":
 242                                         $siteinfo["text"] = $attr["content"];
 243                                         break;
 244                                 case "thumbnail":
 245                                         $siteinfo["image"] = $attr["content"];
 246                                         break;
 247                                 case "twitter:image":
 248                                         $siteinfo["image"] = $attr["content"];
 249                                         break;
 250                                 case "twitter:image:src":
 251                                         $siteinfo["image"] = $attr["content"];
 252                                         break;
 253                                 case "twitter:card":
 254                                         if (($siteinfo["type"] == "") OR ($attr["content"] == "photo"))
 255                                                 $siteinfo["type"] = $attr["content"];
 256                                         break;
 257                                 case "twitter:description":
 258                                         $siteinfo["text"] = $attr["content"];
 259                                         break;
 260                                 case "twitter:title":
 261                                         $siteinfo["title"] = $attr["content"];
 262                                         break;
 263                                 case "dc.title":
 264                                         $siteinfo["title"] = $attr["content"];
 265                                         break;
 266                                 case "dc.description":
 267                                         $siteinfo["text"] = $attr["content"];
 268                                         break;
 269                                 case "keywords":
 270                                         $keywords = explode(",", $attr["content"]);
 271                                         break;
 272                                 case "news_keywords":
 273                                         $keywords = explode(",", $attr["content"]);
 274                                         break;
 275                         }
 276                 if ($siteinfo["type"] == "summary")
 277                         $siteinfo["type"] = "link";
 278         }
 279
 280         if (isset($keywords)) {
 281                 $siteinfo["keywords"] = array();
 282                 foreach ($keywords as $keyword)
 283                         if (!in_array(trim($keyword), $siteinfo["keywords"]))
 284                                 $siteinfo["keywords"][] = trim($keyword);
 285         }
 286
 287         //$list = $xpath->query("head/meta[@property]");
 288         $list = $xpath->query("//meta[@property]");
 289         foreach ($list as $node) {
 290                 $attr = array();
 291                 if ($node->attributes->length)
 292                         foreach ($node->attributes as $attribute)
 293                                 $attr[$attribute->name] = $attribute->value;
 294
 295                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 296
 297                 if ($attr["content"] != "")
 298                         switch (strtolower($attr["property"])) {
 299                                 case "og:image":
 300                                         $siteinfo["image"] = $attr["content"];
 301                                         break;
 302                                 case "og:title":
 303                                         $siteinfo["title"] = $attr["content"];
 304                                         break;
 305                                 case "og:description":
 306                                         $siteinfo["text"] = $attr["content"];
 307                                         break;
 308                         }
 309         }
 310
 311         if ((@$siteinfo["image"] == "") AND !$no_guessing) {
 312             $list = $xpath->query("//img[@src]");
 313             foreach ($list as $node) {
 314                 $attr = array();
 315                 if ($node->attributes->length)
 316                     foreach ($node->attributes as $attribute)
 317                         $attr[$attribute->name] = $attribute->value;
 318
 319                         $src = completeurl($attr["src"], $url);
 320                         $photodata = @getimagesize($src);
 321
 322                         if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
 323                                 if ($photodata[0] > 300) {
 324                                         $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
 325                                         $photodata[0] = 300;
 326                                 }
 327                                 if ($photodata[1] > 300) {
 328                                         $photodata[0] = round($photodata[0] * (300 / $photodata[1]));
 329                                         $photodata[1] = 300;
 330                                 }
 331                                 $siteinfo["images"][] = array("src"=>$src,
 332                                                                 "width"=>$photodata[0],
 333                                                                 "height"=>$photodata[1]);
 334                         }
 335
 336                 }
 337     } else {
 338                 $src = completeurl($siteinfo["image"], $url);
 339
 340                 unset($siteinfo["image"]);
 341
 342                 $photodata = @getimagesize($src);
 343
 344                 if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
 345                         $siteinfo["images"][] = array("src"=>$src,
 346                                                         "width"=>$photodata[0],
 347                                                         "height"=>$photodata[1]);
 348         }
 349
 350         if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$no_guessing) {
 351                 $text = "";
 352
 353                 $list = $xpath->query("//div[@class='article']");
 354                 foreach ($list as $node)
 355                         if (strlen($node->nodeValue) > 40)
 356                                 $text .= " ".trim($node->nodeValue);
 357
 358                 if ($text == "") {
 359                         $list = $xpath->query("//div[@class='content']");
 360                         foreach ($list as $node)
 361                                 if (strlen($node->nodeValue) > 40)
 362                                         $text .= " ".trim($node->nodeValue);
 363                 }
 364
 365                 // If none text was found then take the paragraph content
 366                 if ($text == "") {
 367                         $list = $xpath->query("//p");
 368                         foreach ($list as $node)
 369                                 if (strlen($node->nodeValue) > 40)
 370                                         $text .= " ".trim($node->nodeValue);
 371                 }
 372
 373                 if ($text != "") {
 374                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 375
 376                         while (strpos($text, "  "))
 377                                 $text = trim(str_replace("  ", " ", $text));
 378
 379                         $siteinfo["text"] = trim(html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...');
 380                 }
 381         }
 382
 383         logger("parseurl_getsiteinfo: Siteinfo for ".$url." ".print_r($siteinfo, true), LOGGER_DEBUG);
 384
 385         call_hooks('getsiteinfo', $siteinfo);
 386
 387         return($siteinfo);
 388 }
 389
 390 function arr_add_hashes(&$item,$k) {
 391         $item = '#' . $item;
 392 }
 393
 394 function parse_url_content(&$a) {
 395
 396         $text = null;
 397         $str_tags = '';
 398
 399         $textmode = false;
 400
 401         if(local_user() && (! feature_enabled(local_user(),'richtext')))
 402                 $textmode = true;
 403
 404         //if($textmode)
 405         $br = (($textmode) ? "\n" : '<br />');
 406
 407         if(x($_GET,'binurl'))
 408                 $url = trim(hex2bin($_GET['binurl']));
 409         else
 410                 $url = trim($_GET['url']);
 411
 412         if($_GET['title'])
 413                 $title = strip_tags(trim($_GET['title']));
 414
 415         if($_GET['description'])
 416                 $text = strip_tags(trim($_GET['description']));
 417
 418         if($_GET['tags']) {
 419                 $arr_tags = str_getcsv($_GET['tags']);
 420                 if(count($arr_tags)) {
 421                         array_walk($arr_tags,'arr_add_hashes');
 422                         $str_tags = $br . implode(' ',$arr_tags) . $br;
 423                 }
 424         }
 425
 426         // add url scheme if missing
 427         $arrurl = parse_url($url);
 428         if (!x($arrurl, 'scheme')) {
 429                 if (x($arrurl, 'host'))
 430                         $url = "http:".$url;
 431                 else
 432                         $url = "http://".$url;
 433         }
 434
 435         logger('parse_url: ' . $url);
 436
 437         if($textmode)
 438                 $template = '[bookmark=%s]%s[/bookmark]%s';
 439         else
 440                 $template = "<a class=\"bookmark\" href=\"%s\" >%s</a>%s";
 441
 442         $arr = array('url' => $url, 'text' => '');
 443
 444         call_hooks('parse_link', $arr);
 445
 446         if(strlen($arr['text'])) {
 447                 echo $arr['text'];
 448                 killme();
 449         }
 450
 451
 452         if($url && $title && $text) {
 453
 454                 $title = str_replace(array("\r","\n"),array('',''),$title);
 455
 456                 if($textmode)
 457                         $text = '[quote]' . trim($text) . '[/quote]' . $br;
 458                 else {
 459                         $text = '<blockquote>' . htmlspecialchars(trim($text)) . '</blockquote><br />';
 460                         $title = htmlspecialchars($title);
 461                 }
 462
 463                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 464
 465                 logger('parse_url (unparsed): returns: ' . $result);
 466
 467                 echo $result;
 468                 killme();
 469         }
 470
 471         $siteinfo = parseurl_getsiteinfo($url);
 472
 473 //      if ($textmode) {
 474 //              require_once("include/items.php");
 475 //
 476 //              echo add_page_info_data($siteinfo);
 477 //              killme();
 478 //      }
 479
 480         $url= $siteinfo["url"];
 481
 482         // If the link contains BBCode stuff, make a short link out of this to avoid parsing problems
 483         if (strpos($url, '[') OR strpos($url, ']')) {
 484                 require_once("include/network.php");
 485                 $url = short_link($url);
 486         }
 487
 488         $sitedata = "";
 489
 490         if($siteinfo["title"] != "") {
 491                 $text = $siteinfo["text"];
 492                 $title = $siteinfo["title"];
 493         }
 494
 495         $image = "";
 496
 497         if (($siteinfo["type"] != "video") AND (sizeof($siteinfo["images"]) > 0)){
 498                 /* Execute below code only if image is present in siteinfo */
 499
 500                 $total_images = 0;
 501                 $max_images = get_config('system','max_bookmark_images');
 502                 if($max_images === false)
 503                         $max_images = 2;
 504                 else
 505                         $max_images = intval($max_images);
 506
 507                 foreach ($siteinfo["images"] as $imagedata) {
 508                         if($textmode)
 509                                 $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]' . "\n";
 510                         else
 511                                 $image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" /><br />';
 512                         $total_images ++;
 513                         if($max_images && $max_images >= $total_images)
 514                                 break;
 515                 }
 516         }
 517
 518         if(strlen($text)) {
 519                 if($textmode)
 520                         $text = '[quote]'.trim($text).'[/quote]';
 521                 else
 522                         $text = '<blockquote>'.htmlspecialchars(trim($text)).'</blockquote>';
 523         }
 524
 525         if($image)
 526                 $text = $br.$br.$image.$text;
 527         else
 528                 $text = $br.$text;
 529
 530         $title = str_replace(array("\r","\n"),array('',''),$title);
 531
 532         $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 533
 534         logger('parse_url: returns: ' . $result);
 535
 536         $sitedata .=  trim($result);
 537
 538         if (($siteinfo["type"] == "video") AND ($url != ""))
 539                 echo "[class=type-video]".$sitedata."[/class]";
 540         elseif (($siteinfo["type"] != "photo"))
 541                 echo "[class=type-link]".$sitedata."[/class]";
 542         else
 543                 echo "[class=type-photo]".$title.$br.$image."[/class]";
 544
 545         killme();
 546 }
 547 ?>