mod/parse_url.php

   1 <?php
   2 /**
   3  * @file mod/parse_url.php
   4  *
   5  * @todo https://developers.google.com/+/plugins/snippet/
   6  *
   7  * @verbatim
   8  * <meta itemprop="name" content="Toller Titel">
   9  * <meta itemprop="description" content="Eine tolle Beschreibung">
  10  * <meta itemprop="image" content="http://maple.libertreeproject.org/images/tree-icon.png">
  11  *
  12  * <body itemscope itemtype="http://schema.org/Product">
  13  *   <h1 itemprop="name">Shiny Trinket</h1>
  14  *   <img itemprop="image" src="{image-url}" />
  15  *   <p itemprop="description">Shiny trinkets are shiny.</p>
  16  * </body>
  17  * @endverbatim
  18 */
  19
  20 if(!function_exists('deletenode')) {
  21         function deletenode(&$doc, $node)
  22         {
  23                 $xpath = new DomXPath($doc);
  24                 $list = $xpath->query("//".$node);
  25                 foreach ($list as $child)
  26                         $child->parentNode->removeChild($child);
  27         }
  28 }
  29
  30 if(! function_exists('completeurl')) {
  31 function completeurl($url, $scheme) {
  32         $urlarr = parse_url($url);
  33
  34         if (isset($urlarr["scheme"]))
  35                 return($url);
  36
  37         $schemearr = parse_url($scheme);
  38
  39         $complete = $schemearr["scheme"]."://".$schemearr["host"];
  40
  41         if (@$schemearr["port"] != "")
  42                 $complete .= ":".$schemearr["port"];
  43
  44                 if(strpos($urlarr['path'],'/') !== 0)
  45                         $complete .= '/';
  46
  47         $complete .= $urlarr["path"];
  48
  49         if (@$urlarr["query"] != "")
  50                 $complete .= "?".$urlarr["query"];
  51
  52         if (@$urlarr["fragment"] != "")
  53                 $complete .= "#".$urlarr["fragment"];
  54
  55         return($complete);
  56 }
  57 }
  58
  59 if(! function_exists('parseurl_getsiteinfo_cached')) {
  60 function parseurl_getsiteinfo_cached($url, $no_guessing = false, $do_oembed = true) {
  61
  62         if ($url == "")
  63                 return false;
  64
  65         $r = q("SELECT * FROM `parsed_url` WHERE `url` = '%s' AND `guessing` = %d AND `oembed` = %d",
  66                 dbesc(normalise_link($url)), intval(!$no_guessing), intval($do_oembed));
  67
  68         if ($r)
  69                 $data = $r[0]["content"];
  70
  71         if (!is_null($data)) {
  72                 $data = unserialize($data);
  73                 return $data;
  74         }
  75
  76         $data = parseurl_getsiteinfo($url, $no_guessing, $do_oembed);
  77
  78         q("INSERT INTO `parsed_url` (`url`, `guessing`, `oembed`, `content`, `created`) VALUES ('%s', %d, %d, '%s', '%s')",
  79                 dbesc(normalise_link($url)), intval(!$no_guessing), intval($do_oembed), dbesc(serialize($data)), dbesc(datetime_convert()));
  80
  81         return $data;
  82 }
  83 }
  84
  85 if(! function_exists('parseurl_getsiteinfo')) {
  86 function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1) {
  87         require_once("include/network.php");
  88         require_once("include/Photo.php");
  89
  90         $a = get_app();
  91
  92         $siteinfo = array();
  93
  94         if ($count > 10) {
  95                 logger("parseurl_getsiteinfo: Endless loop detected for ".$url, LOGGER_DEBUG);
  96                 return($siteinfo);
  97         }
  98
  99         $url = trim($url, "'");
 100         $url = trim($url, '"');
 101
 102         $url = original_url($url);
 103
 104         $siteinfo["url"] = $url;
 105         $siteinfo["type"] = "link";
 106
 107         $stamp1 = microtime(true);
 108
 109         $ch = curl_init();
 110         curl_setopt($ch, CURLOPT_URL, $url);
 111         curl_setopt($ch, CURLOPT_HEADER, 1);
 112         curl_setopt($ch, CURLOPT_NOBODY, 1);
 113         curl_setopt($ch, CURLOPT_TIMEOUT, 3);
 114         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
 115         //curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
 116         curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
 117
 118         $header = curl_exec($ch);
 119         $curl_info = @curl_getinfo($ch);
 120         $http_code = $curl_info['http_code'];
 121         curl_close($ch);
 122
 123         $a->save_timestamp($stamp1, "network");
 124
 125         if ((($curl_info['http_code'] == "301") OR ($curl_info['http_code'] == "302") OR ($curl_info['http_code'] == "303") OR ($curl_info['http_code'] == "307"))
 126                 AND (($curl_info['redirect_url'] != "") OR ($curl_info['location'] != ""))) {
 127                 if ($curl_info['redirect_url'] != "")
 128                         $siteinfo = parseurl_getsiteinfo($curl_info['redirect_url'], $no_guessing, $do_oembed, ++$count);
 129                 else
 130                         $siteinfo = parseurl_getsiteinfo($curl_info['location'], $no_guessing, $do_oembed, ++$count);
 131                 return($siteinfo);
 132         }
 133
 134         // if the file is too large then exit
 135         if ($curl_info["download_content_length"] > 1000000)
 136                 return($siteinfo);
 137
 138         // if it isn't a HTML file then exit
 139         if (($curl_info["content_type"] != "") AND !strstr(strtolower($curl_info["content_type"]),"html"))
 140                 return($siteinfo);
 141
 142         if ($do_oembed) {
 143                 require_once("include/oembed.php");
 144
 145                 $oembed_data = oembed_fetch_url($url);
 146
 147                 if ($oembed_data->type != "error")
 148                         $siteinfo["type"] = $oembed_data->type;
 149
 150                 if (($oembed_data->type == "link") AND ($siteinfo["type"] != "photo")) {
 151                         if (isset($oembed_data->title))
 152                                 $siteinfo["title"] = $oembed_data->title;
 153                         if (isset($oembed_data->description))
 154                                 $siteinfo["text"] = trim($oembed_data->description);
 155                         if (isset($oembed_data->thumbnail_url))
 156                                 $siteinfo["image"] = $oembed_data->thumbnail_url;
 157                 }
 158         }
 159
 160         $stamp1 = microtime(true);
 161
 162         // Now fetch the body as well
 163         $ch = curl_init();
 164         curl_setopt($ch, CURLOPT_URL, $url);
 165         curl_setopt($ch, CURLOPT_HEADER, 1);
 166         curl_setopt($ch, CURLOPT_NOBODY, 0);
 167         curl_setopt($ch, CURLOPT_TIMEOUT, 10);
 168         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
 169         curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
 170
 171         $header = curl_exec($ch);
 172         $curl_info = @curl_getinfo($ch);
 173         $http_code = $curl_info['http_code'];
 174         curl_close($ch);
 175
 176         $a->save_timestamp($stamp1, "network");
 177
 178         // Fetch the first mentioned charset. Can be in body or header
 179         $charset = "";
 180         if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
 181                 $charset = trim(trim(trim(array_pop($matches)), ';,'));
 182
 183         if ($charset == "")
 184                 $charset = "utf-8";
 185
 186         $pos = strpos($header, "\r\n\r\n");
 187
 188         if ($pos)
 189                 $body = trim(substr($header, $pos));
 190         else
 191                 $body = $header;
 192
 193         if (($charset != '') AND (strtoupper($charset) != "UTF-8")) {
 194                 logger("parseurl_getsiteinfo: detected charset ".$charset, LOGGER_DEBUG);
 195                 //$body = mb_convert_encoding($body, "UTF-8", $charset);
 196                 $body = iconv($charset, "UTF-8//TRANSLIT", $body);
 197         }
 198
 199         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
 200
 201         $doc = new DOMDocument();
 202         @$doc->loadHTML($body);
 203
 204         deletenode($doc, 'style');
 205         deletenode($doc, 'script');
 206         deletenode($doc, 'option');
 207         deletenode($doc, 'h1');
 208         deletenode($doc, 'h2');
 209         deletenode($doc, 'h3');
 210         deletenode($doc, 'h4');
 211         deletenode($doc, 'h5');
 212         deletenode($doc, 'h6');
 213         deletenode($doc, 'ol');
 214         deletenode($doc, 'ul');
 215
 216         $xpath = new DomXPath($doc);
 217
 218         $list = $xpath->query("//meta[@content]");
 219         foreach ($list as $node) {
 220                 $attr = array();
 221                 if ($node->attributes->length)
 222                         foreach ($node->attributes as $attribute)
 223                                 $attr[$attribute->name] = $attribute->value;
 224
 225                 if (@$attr["http-equiv"] == 'refresh') {
 226                         $path = $attr["content"];
 227                         $pathinfo = explode(";", $path);
 228                         $content = "";
 229                         foreach ($pathinfo AS $value) {
 230                                 if (substr(strtolower($value), 0, 4) == "url=")
 231                                         $content = substr($value, 4);
 232                         }
 233                         if ($content != "") {
 234                                 $siteinfo = parseurl_getsiteinfo($content, $no_guessing, $do_oembed, ++$count);
 235                                 return($siteinfo);
 236                         }
 237                 }
 238         }
 239
 240         //$list = $xpath->query("head/title");
 241         $list = $xpath->query("//title");
 242         foreach ($list as $node)
 243                 $siteinfo["title"] =  html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
 244
 245         //$list = $xpath->query("head/meta[@name]");
 246         $list = $xpath->query("//meta[@name]");
 247         foreach ($list as $node) {
 248                 $attr = array();
 249                 if ($node->attributes->length)
 250                         foreach ($node->attributes as $attribute)
 251                                 $attr[$attribute->name] = $attribute->value;
 252
 253                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 254
 255                 if ($attr["content"] != "")
 256                         switch (strtolower($attr["name"])) {
 257                                 case "fulltitle":
 258                                         $siteinfo["title"] = $attr["content"];
 259                                         break;
 260                                 case "description":
 261                                         $siteinfo["text"] = $attr["content"];
 262                                         break;
 263                                 case "thumbnail":
 264                                         $siteinfo["image"] = $attr["content"];
 265                                         break;
 266                                 case "twitter:image":
 267                                         $siteinfo["image"] = $attr["content"];
 268                                         break;
 269                                 case "twitter:image:src":
 270                                         $siteinfo["image"] = $attr["content"];
 271                                         break;
 272                                 case "twitter:card":
 273                                         if (($siteinfo["type"] == "") OR ($attr["content"] == "photo"))
 274                                                 $siteinfo["type"] = $attr["content"];
 275                                         break;
 276                                 case "twitter:description":
 277                                         $siteinfo["text"] = $attr["content"];
 278                                         break;
 279                                 case "twitter:title":
 280                                         $siteinfo["title"] = $attr["content"];
 281                                         break;
 282                                 case "dc.title":
 283                                         $siteinfo["title"] = $attr["content"];
 284                                         break;
 285                                 case "dc.description":
 286                                         $siteinfo["text"] = $attr["content"];
 287                                         break;
 288                                 case "keywords":
 289                                         $keywords = explode(",", $attr["content"]);
 290                                         break;
 291                                 case "news_keywords":
 292                                         $keywords = explode(",", $attr["content"]);
 293                                         break;
 294                         }
 295                 if ($siteinfo["type"] == "summary")
 296                         $siteinfo["type"] = "link";
 297         }
 298
 299         if (isset($keywords)) {
 300                 $siteinfo["keywords"] = array();
 301                 foreach ($keywords as $keyword)
 302                         if (!in_array(trim($keyword), $siteinfo["keywords"]))
 303                                 $siteinfo["keywords"][] = trim($keyword);
 304         }
 305
 306         //$list = $xpath->query("head/meta[@property]");
 307         $list = $xpath->query("//meta[@property]");
 308         foreach ($list as $node) {
 309                 $attr = array();
 310                 if ($node->attributes->length)
 311                         foreach ($node->attributes as $attribute)
 312                                 $attr[$attribute->name] = $attribute->value;
 313
 314                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 315
 316                 if ($attr["content"] != "")
 317                         switch (strtolower($attr["property"])) {
 318                                 case "og:image":
 319                                         $siteinfo["image"] = $attr["content"];
 320                                         break;
 321                                 case "og:title":
 322                                         $siteinfo["title"] = $attr["content"];
 323                                         break;
 324                                 case "og:description":
 325                                         $siteinfo["text"] = $attr["content"];
 326                                         break;
 327                         }
 328         }
 329
 330         if ((@$siteinfo["image"] == "") AND !$no_guessing) {
 331             $list = $xpath->query("//img[@src]");
 332             foreach ($list as $node) {
 333                 $attr = array();
 334                 if ($node->attributes->length)
 335                     foreach ($node->attributes as $attribute)
 336                         $attr[$attribute->name] = $attribute->value;
 337
 338                         $src = completeurl($attr["src"], $url);
 339                         $photodata = get_photo_info($src);
 340
 341                         if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
 342                                 if ($photodata[0] > 300) {
 343                                         $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
 344                                         $photodata[0] = 300;
 345                                 }
 346                                 if ($photodata[1] > 300) {
 347                                         $photodata[0] = round($photodata[0] * (300 / $photodata[1]));
 348                                         $photodata[1] = 300;
 349                                 }
 350                                 $siteinfo["images"][] = array("src"=>$src,
 351                                                                 "width"=>$photodata[0],
 352                                                                 "height"=>$photodata[1]);
 353                         }
 354
 355                 }
 356     } elseif ($siteinfo["image"] != "") {
 357                 $src = completeurl($siteinfo["image"], $url);
 358
 359                 unset($siteinfo["image"]);
 360
 361                 $photodata = get_photo_info($src);
 362
 363                 if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
 364                         $siteinfo["images"][] = array("src"=>$src,
 365                                                         "width"=>$photodata[0],
 366                                                         "height"=>$photodata[1]);
 367         }
 368
 369         if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$no_guessing) {
 370                 $text = "";
 371
 372                 $list = $xpath->query("//div[@class='article']");
 373                 foreach ($list as $node)
 374                         if (strlen($node->nodeValue) > 40)
 375                                 $text .= " ".trim($node->nodeValue);
 376
 377                 if ($text == "") {
 378                         $list = $xpath->query("//div[@class='content']");
 379                         foreach ($list as $node)
 380                                 if (strlen($node->nodeValue) > 40)
 381                                         $text .= " ".trim($node->nodeValue);
 382                 }
 383
 384                 // If none text was found then take the paragraph content
 385                 if ($text == "") {
 386                         $list = $xpath->query("//p");
 387                         foreach ($list as $node)
 388                                 if (strlen($node->nodeValue) > 40)
 389                                         $text .= " ".trim($node->nodeValue);
 390                 }
 391
 392                 if ($text != "") {
 393                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 394
 395                         while (strpos($text, "  "))
 396                                 $text = trim(str_replace("  ", " ", $text));
 397
 398                         $siteinfo["text"] = trim(html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...');
 399                 }
 400         }
 401
 402         logger("parseurl_getsiteinfo: Siteinfo for ".$url." ".print_r($siteinfo, true), LOGGER_DEBUG);
 403
 404         call_hooks('getsiteinfo', $siteinfo);
 405
 406         return($siteinfo);
 407 }
 408 }
 409
 410 if(! function_exists('arr_add_hashes')) {
 411 function arr_add_hashes(&$item,$k) {
 412         $item = '#' . $item;
 413 }
 414 }
 415
 416 if(! function_exists('parse_url_content')) {
 417 function parse_url_content(&$a) {
 418
 419         $text = null;
 420         $str_tags = '';
 421
 422         $textmode = false;
 423
 424         if(local_user() && (! feature_enabled(local_user(),'richtext')))
 425                 $textmode = true;
 426
 427         //if($textmode)
 428         $br = (($textmode) ? "\n" : '<br />');
 429
 430         if(x($_GET,'binurl'))
 431                 $url = trim(hex2bin($_GET['binurl']));
 432         else
 433                 $url = trim($_GET['url']);
 434
 435         if($_GET['title'])
 436                 $title = strip_tags(trim($_GET['title']));
 437
 438         if($_GET['description'])
 439                 $text = strip_tags(trim($_GET['description']));
 440
 441         if($_GET['tags']) {
 442                 $arr_tags = str_getcsv($_GET['tags']);
 443                 if(count($arr_tags)) {
 444                         array_walk($arr_tags,'arr_add_hashes');
 445                         $str_tags = $br . implode(' ',$arr_tags) . $br;
 446                 }
 447         }
 448
 449         // add url scheme if missing
 450         $arrurl = parse_url($url);
 451         if (!x($arrurl, 'scheme')) {
 452                 if (x($arrurl, 'host'))
 453                         $url = "http:".$url;
 454                 else
 455                         $url = "http://".$url;
 456         }
 457
 458         logger('parse_url: ' . $url);
 459
 460         if($textmode)
 461                 $template = '[bookmark=%s]%s[/bookmark]%s';
 462         else
 463                 $template = "<a class=\"bookmark\" href=\"%s\" >%s</a>%s";
 464
 465         $arr = array('url' => $url, 'text' => '');
 466
 467         call_hooks('parse_link', $arr);
 468
 469         if(strlen($arr['text'])) {
 470                 echo $arr['text'];
 471                 killme();
 472         }
 473
 474
 475         if($url && $title && $text) {
 476
 477                 $title = str_replace(array("\r","\n"),array('',''),$title);
 478
 479                 if($textmode)
 480                         $text = '[quote]' . trim($text) . '[/quote]' . $br;
 481                 else {
 482                         $text = '<blockquote>' . htmlspecialchars(trim($text)) . '</blockquote><br />';
 483                         $title = htmlspecialchars($title);
 484                 }
 485
 486                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 487
 488                 logger('parse_url (unparsed): returns: ' . $result);
 489
 490                 echo $result;
 491                 killme();
 492         }
 493
 494         $siteinfo = parseurl_getsiteinfo($url);
 495
 496 //      if ($textmode) {
 497 //              require_once("include/items.php");
 498 //
 499 //              echo add_page_info_data($siteinfo);
 500 //              killme();
 501 //      }
 502
 503         $url= $siteinfo["url"];
 504
 505         // If the link contains BBCode stuff, make a short link out of this to avoid parsing problems
 506         if (strpos($url, '[') OR strpos($url, ']')) {
 507                 require_once("include/network.php");
 508                 $url = short_link($url);
 509         }
 510
 511         $sitedata = "";
 512
 513         if($siteinfo["title"] != "") {
 514                 $text = $siteinfo["text"];
 515                 $title = $siteinfo["title"];
 516         }
 517
 518         $image = "";
 519
 520         if (($siteinfo["type"] != "video") AND (sizeof($siteinfo["images"]) > 0)){
 521                 /* Execute below code only if image is present in siteinfo */
 522
 523                 $total_images = 0;
 524                 $max_images = get_config('system','max_bookmark_images');
 525                 if($max_images === false)
 526                         $max_images = 2;
 527                 else
 528                         $max_images = intval($max_images);
 529
 530                 foreach ($siteinfo["images"] as $imagedata) {
 531                         if($textmode)
 532                                 $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]' . "\n";
 533                         else
 534                                 $image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" /><br />';
 535                         $total_images ++;
 536                         if($max_images && $max_images >= $total_images)
 537                                 break;
 538                 }
 539         }
 540
 541         if(strlen($text)) {
 542                 if($textmode)
 543                         $text = '[quote]'.trim($text).'[/quote]';
 544                 else
 545                         $text = '<blockquote>'.htmlspecialchars(trim($text)).'</blockquote>';
 546         }
 547
 548         if($image)
 549                 $text = $br.$br.$image.$text;
 550         else
 551                 $text = $br.$text;
 552
 553         $title = str_replace(array("\r","\n"),array('',''),$title);
 554
 555         $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 556
 557         logger('parse_url: returns: ' . $result);
 558
 559         $sitedata .=  trim($result);
 560
 561         if (($siteinfo["type"] == "video") AND ($url != ""))
 562                 echo "[class=type-video]".$sitedata."[/class]";
 563         elseif (($siteinfo["type"] != "photo"))
 564                 echo "[class=type-link]".$sitedata."[/class]";
 565         else
 566                 echo "[class=type-photo]".$title.$br.$image."[/class]";
 567
 568         killme();
 569 }
 570 }
 571 ?>