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