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 function completeurl($url, $scheme) {
  31         $urlarr = parse_url($url);
  32
  33         if (isset($urlarr["scheme"]))
  34                 return($url);
  35
  36         $schemearr = parse_url($scheme);
  37
  38         $complete = $schemearr["scheme"]."://".$schemearr["host"];
  39
  40         if (@$schemearr["port"] != "")
  41                 $complete .= ":".$schemearr["port"];
  42
  43                 if(strpos($urlarr['path'],'/') !== 0)
  44                         $complete .= '/';
  45
  46         $complete .= $urlarr["path"];
  47
  48         if (@$urlarr["query"] != "")
  49                 $complete .= "?".$urlarr["query"];
  50
  51         if (@$urlarr["fragment"] != "")
  52                 $complete .= "#".$urlarr["fragment"];
  53
  54         return($complete);
  55 }
  56
  57 function parseurl_getsiteinfo_cached($url, $no_guessing = false, $do_oembed = true) {
  58
  59         if ($url == "")
  60                 return false;
  61
  62         $r = q("SELECT * FROM `parsed_url` WHERE `url` = '%s' AND `guessing` = %d AND `oembed` = %d",
  63                 dbesc(normalise_link($url)), intval(!$no_guessing), intval($do_oembed));
  64
  65         if ($r)
  66                 $data = $r[0]["content"];
  67
  68         if (!is_null($data)) {
  69                 $data = unserialize($data);
  70                 return $data;
  71         }
  72
  73         $data = parseurl_getsiteinfo($url, $no_guessing, $do_oembed);
  74
  75         q("INSERT INTO `parsed_url` (`url`, `guessing`, `oembed`, `content`, `created`) VALUES ('%s', %d, %d, '%s', '%s')
  76                  ON DUPLICATE KEY UPDATE `content` = '%s', `created` = '%s'",
  77                 dbesc(normalise_link($url)), intval(!$no_guessing), intval($do_oembed),
  78                 dbesc(serialize($data)), dbesc(datetime_convert()),
  79                 dbesc(serialize($data)), dbesc(datetime_convert()));
  80
  81         return $data;
  82 }
  83
  84 function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1) {
  85         require_once("include/network.php");
  86         require_once("include/Photo.php");
  87
  88         $a = get_app();
  89
  90         $siteinfo = array();
  91
  92         if ($count > 10) {
  93                 logger("parseurl_getsiteinfo: Endless loop detected for ".$url, LOGGER_DEBUG);
  94                 return($siteinfo);
  95         }
  96
  97         $url = trim($url, "'");
  98         $url = trim($url, '"');
  99
 100         $url = original_url($url);
 101
 102         $siteinfo["url"] = $url;
 103         $siteinfo["type"] = "link";
 104
 105         $stamp1 = microtime(true);
 106
 107         $ch = curl_init();
 108         curl_setopt($ch, CURLOPT_URL, $url);
 109         curl_setopt($ch, CURLOPT_HEADER, 1);
 110         curl_setopt($ch, CURLOPT_NOBODY, 1);
 111         curl_setopt($ch, CURLOPT_TIMEOUT, 3);
 112         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
 113         //curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
 114         curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
 115
 116         $header = curl_exec($ch);
 117         $curl_info = @curl_getinfo($ch);
 118         $http_code = $curl_info['http_code'];
 119         curl_close($ch);
 120
 121         $a->save_timestamp($stamp1, "network");
 122
 123         if ((($curl_info['http_code'] == "301") OR ($curl_info['http_code'] == "302") OR ($curl_info['http_code'] == "303") OR ($curl_info['http_code'] == "307"))
 124                 AND (($curl_info['redirect_url'] != "") OR ($curl_info['location'] != ""))) {
 125                 if ($curl_info['redirect_url'] != "")
 126                         $siteinfo = parseurl_getsiteinfo($curl_info['redirect_url'], $no_guessing, $do_oembed, ++$count);
 127                 else
 128                         $siteinfo = parseurl_getsiteinfo($curl_info['location'], $no_guessing, $do_oembed, ++$count);
 129                 return($siteinfo);
 130         }
 131
 132         // if the file is too large then exit
 133         if ($curl_info["download_content_length"] > 1000000)
 134                 return($siteinfo);
 135
 136         // if it isn't a HTML file then exit
 137         if (($curl_info["content_type"] != "") AND !strstr(strtolower($curl_info["content_type"]),"html"))
 138                 return($siteinfo);
 139
 140         if ($do_oembed) {
 141                 require_once("include/oembed.php");
 142
 143                 $oembed_data = oembed_fetch_url($url);
 144
 145                 if ($oembed_data->type != "error")
 146                         $siteinfo["type"] = $oembed_data->type;
 147
 148                 if (($oembed_data->type == "link") AND ($siteinfo["type"] != "photo")) {
 149                         if (isset($oembed_data->title))
 150                                 $siteinfo["title"] = $oembed_data->title;
 151                         if (isset($oembed_data->description))
 152                                 $siteinfo["text"] = trim($oembed_data->description);
 153                         if (isset($oembed_data->thumbnail_url))
 154                                 $siteinfo["image"] = $oembed_data->thumbnail_url;
 155                 }
 156         }
 157
 158         $stamp1 = microtime(true);
 159
 160         // Now fetch the body as well
 161         $ch = curl_init();
 162         curl_setopt($ch, CURLOPT_URL, $url);
 163         curl_setopt($ch, CURLOPT_HEADER, 1);
 164         curl_setopt($ch, CURLOPT_NOBODY, 0);
 165         curl_setopt($ch, CURLOPT_TIMEOUT, 10);
 166         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
 167         curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
 168
 169         $header = curl_exec($ch);
 170         $curl_info = @curl_getinfo($ch);
 171         $http_code = $curl_info['http_code'];
 172         curl_close($ch);
 173
 174         $a->save_timestamp($stamp1, "network");
 175
 176         // Fetch the first mentioned charset. Can be in body or header
 177         $charset = "";
 178         if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
 179                 $charset = trim(trim(trim(array_pop($matches)), ';,'));
 180
 181         if ($charset == "")
 182                 $charset = "utf-8";
 183
 184         $pos = strpos($header, "\r\n\r\n");
 185
 186         if ($pos)
 187                 $body = trim(substr($header, $pos));
 188         else
 189                 $body = $header;
 190
 191         if (($charset != '') AND (strtoupper($charset) != "UTF-8")) {
 192                 logger("parseurl_getsiteinfo: detected charset ".$charset, LOGGER_DEBUG);
 193                 //$body = mb_convert_encoding($body, "UTF-8", $charset);
 194                 $body = iconv($charset, "UTF-8//TRANSLIT", $body);
 195         }
 196
 197         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
 198
 199         $doc = new DOMDocument();
 200         @$doc->loadHTML($body);
 201
 202         deletenode($doc, 'style');
 203         deletenode($doc, 'script');
 204         deletenode($doc, 'option');
 205         deletenode($doc, 'h1');
 206         deletenode($doc, 'h2');
 207         deletenode($doc, 'h3');
 208         deletenode($doc, 'h4');
 209         deletenode($doc, 'h5');
 210         deletenode($doc, 'h6');
 211         deletenode($doc, 'ol');
 212         deletenode($doc, 'ul');
 213
 214         $xpath = new DomXPath($doc);
 215
 216         $list = $xpath->query("//meta[@content]");
 217         foreach ($list as $node) {
 218                 $attr = array();
 219                 if ($node->attributes->length)
 220                         foreach ($node->attributes as $attribute)
 221                                 $attr[$attribute->name] = $attribute->value;
 222
 223                 if (@$attr["http-equiv"] == 'refresh') {
 224                         $path = $attr["content"];
 225                         $pathinfo = explode(";", $path);
 226                         $content = "";
 227                         foreach ($pathinfo AS $value) {
 228                                 if (substr(strtolower($value), 0, 4) == "url=")
 229                                         $content = substr($value, 4);
 230                         }
 231                         if ($content != "") {
 232                                 $siteinfo = parseurl_getsiteinfo($content, $no_guessing, $do_oembed, ++$count);
 233                                 return($siteinfo);
 234                         }
 235                 }
 236         }
 237
 238         $list = $xpath->query("//title");
 239         if ($list->length > 0)
 240                 $siteinfo["title"] = $list->item(0)->nodeValue;
 241
 242         //$list = $xpath->query("head/meta[@name]");
 243         $list = $xpath->query("//meta[@name]");
 244         foreach ($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                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 251
 252                 if ($attr["content"] != "")
 253                         switch (strtolower($attr["name"])) {
 254                                 case "fulltitle":
 255                                         $siteinfo["title"] = $attr["content"];
 256                                         break;
 257                                 case "description":
 258                                         $siteinfo["text"] = $attr["content"];
 259                                         break;
 260                                 case "thumbnail":
 261                                         $siteinfo["image"] = $attr["content"];
 262                                         break;
 263                                 case "twitter:image":
 264                                         $siteinfo["image"] = $attr["content"];
 265                                         break;
 266                                 case "twitter:image:src":
 267                                         $siteinfo["image"] = $attr["content"];
 268                                         break;
 269                                 case "twitter:card":
 270                                         if (($siteinfo["type"] == "") OR ($attr["content"] == "photo"))
 271                                                 $siteinfo["type"] = $attr["content"];
 272                                         break;
 273                                 case "twitter:description":
 274                                         $siteinfo["text"] = $attr["content"];
 275                                         break;
 276                                 case "twitter:title":
 277                                         $siteinfo["title"] = $attr["content"];
 278                                         break;
 279                                 case "dc.title":
 280                                         $siteinfo["title"] = $attr["content"];
 281                                         break;
 282                                 case "dc.description":
 283                                         $siteinfo["text"] = $attr["content"];
 284                                         break;
 285                                 case "keywords":
 286                                         $keywords = explode(",", $attr["content"]);
 287                                         break;
 288                                 case "news_keywords":
 289                                         $keywords = explode(",", $attr["content"]);
 290                                         break;
 291                         }
 292                 if ($siteinfo["type"] == "summary")
 293                         $siteinfo["type"] = "link";
 294         }
 295
 296         if (isset($keywords)) {
 297                 $siteinfo["keywords"] = array();
 298                 foreach ($keywords as $keyword)
 299                         if (!in_array(trim($keyword), $siteinfo["keywords"]))
 300                                 $siteinfo["keywords"][] = trim($keyword);
 301         }
 302
 303         //$list = $xpath->query("head/meta[@property]");
 304         $list = $xpath->query("//meta[@property]");
 305         foreach ($list as $node) {
 306                 $attr = array();
 307                 if ($node->attributes->length)
 308                         foreach ($node->attributes as $attribute)
 309                                 $attr[$attribute->name] = $attribute->value;
 310
 311                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
 312
 313                 if ($attr["content"] != "")
 314                         switch (strtolower($attr["property"])) {
 315                                 case "og:image":
 316                                         $siteinfo["image"] = $attr["content"];
 317                                         break;
 318                                 case "og:title":
 319                                         $siteinfo["title"] = $attr["content"];
 320                                         break;
 321                                 case "og:description":
 322                                         $siteinfo["text"] = $attr["content"];
 323                                         break;
 324                         }
 325         }
 326
 327         if ((@$siteinfo["image"] == "") AND !$no_guessing) {
 328             $list = $xpath->query("//img[@src]");
 329             foreach ($list as $node) {
 330                 $attr = array();
 331                 if ($node->attributes->length)
 332                     foreach ($node->attributes as $attribute)
 333                         $attr[$attribute->name] = $attribute->value;
 334
 335                         $src = completeurl($attr["src"], $url);
 336                         $photodata = get_photo_info($src);
 337
 338                         if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
 339                                 if ($photodata[0] > 300) {
 340                                         $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
 341                                         $photodata[0] = 300;
 342                                 }
 343                                 if ($photodata[1] > 300) {
 344                                         $photodata[0] = round($photodata[0] * (300 / $photodata[1]));
 345                                         $photodata[1] = 300;
 346                                 }
 347                                 $siteinfo["images"][] = array("src"=>$src,
 348                                                                 "width"=>$photodata[0],
 349                                                                 "height"=>$photodata[1]);
 350                         }
 351
 352                 }
 353     } elseif ($siteinfo["image"] != "") {
 354                 $src = completeurl($siteinfo["image"], $url);
 355
 356                 unset($siteinfo["image"]);
 357
 358                 $photodata = get_photo_info($src);
 359
 360                 if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
 361                         $siteinfo["images"][] = array("src"=>$src,
 362                                                         "width"=>$photodata[0],
 363                                                         "height"=>$photodata[1]);
 364         }
 365
 366         if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$no_guessing) {
 367                 $text = "";
 368
 369                 $list = $xpath->query("//div[@class='article']");
 370                 foreach ($list as $node)
 371                         if (strlen($node->nodeValue) > 40)
 372                                 $text .= " ".trim($node->nodeValue);
 373
 374                 if ($text == "") {
 375                         $list = $xpath->query("//div[@class='content']");
 376                         foreach ($list as $node)
 377                                 if (strlen($node->nodeValue) > 40)
 378                                         $text .= " ".trim($node->nodeValue);
 379                 }
 380
 381                 // If none text was found then take the paragraph content
 382                 if ($text == "") {
 383                         $list = $xpath->query("//p");
 384                         foreach ($list as $node)
 385                                 if (strlen($node->nodeValue) > 40)
 386                                         $text .= " ".trim($node->nodeValue);
 387                 }
 388
 389                 if ($text != "") {
 390                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 391
 392                         while (strpos($text, "  "))
 393                                 $text = trim(str_replace("  ", " ", $text));
 394
 395                         $siteinfo["text"] = trim(html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...');
 396                 }
 397         }
 398
 399         logger("parseurl_getsiteinfo: Siteinfo for ".$url." ".print_r($siteinfo, true), LOGGER_DEBUG);
 400
 401         call_hooks('getsiteinfo', $siteinfo);
 402
 403         return($siteinfo);
 404 }
 405
 406 function arr_add_hashes(&$item,$k) {
 407         $item = '#' . $item;
 408 }
 409
 410 function parse_url_content(&$a) {
 411
 412         require_once("include/items.php");
 413
 414         $text = null;
 415         $str_tags = '';
 416
 417         $textmode = false;
 418
 419         if(local_user() && (! feature_enabled(local_user(),'richtext')))
 420                 $textmode = true;
 421
 422         //if($textmode)
 423         $br = (($textmode) ? "\n" : '<br />');
 424
 425         if(x($_GET,'binurl'))
 426                 $url = trim(hex2bin($_GET['binurl']));
 427         else
 428                 $url = trim($_GET['url']);
 429
 430         if($_GET['title'])
 431                 $title = strip_tags(trim($_GET['title']));
 432
 433         if($_GET['description'])
 434                 $text = strip_tags(trim($_GET['description']));
 435
 436         if($_GET['tags']) {
 437                 $arr_tags = str_getcsv($_GET['tags']);
 438                 if(count($arr_tags)) {
 439                         array_walk($arr_tags,'arr_add_hashes');
 440                         $str_tags = $br . implode(' ',$arr_tags) . $br;
 441                 }
 442         }
 443
 444         // add url scheme if missing
 445         $arrurl = parse_url($url);
 446         if (!x($arrurl, 'scheme')) {
 447                 if (x($arrurl, 'host'))
 448                         $url = "http:".$url;
 449                 else
 450                         $url = "http://".$url;
 451         }
 452
 453         logger('parse_url: ' . $url);
 454
 455         if($textmode)
 456                 $template = '[bookmark=%s]%s[/bookmark]%s';
 457         else
 458                 $template = "<a class=\"bookmark\" href=\"%s\" >%s</a>%s";
 459
 460         $arr = array('url' => $url, 'text' => '');
 461
 462         call_hooks('parse_link', $arr);
 463
 464         if(strlen($arr['text'])) {
 465                 echo $arr['text'];
 466                 killme();
 467         }
 468
 469
 470         if($url && $title && $text) {
 471
 472                 $title = str_replace(array("\r","\n"),array('',''),$title);
 473
 474                 if($textmode)
 475                         $text = '[quote]' . trim($text) . '[/quote]' . $br;
 476                 else {
 477                         $text = '<blockquote>' . htmlspecialchars(trim($text)) . '</blockquote><br />';
 478                         $title = htmlspecialchars($title);
 479                 }
 480
 481                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 482
 483                 logger('parse_url (unparsed): returns: ' . $result);
 484
 485                 echo $result;
 486                 killme();
 487         }
 488
 489         $siteinfo = parseurl_getsiteinfo($url);
 490
 491         unset($siteinfo["keywords"]);
 492
 493         $info = add_page_info_data($siteinfo);
 494
 495         if (!$textmode)
 496                 // Replace ' with ’ - not perfect - but the richtext editor has problems otherwise
 497                 $info = str_replace(array("&#039;"), array("&#8217;"), $info);
 498
 499         echo $info;
 500
 501         killme();
 502 }
 503 ?>