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) {
  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,'Opera/9.64(Windows NT 5.1; U; de) Presto/2.1.1');
  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         // Fetch the first mentioned charset. Can be in body or header
  79         if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
  80                 $charset = trim(array_pop($matches));
  81         else
  82                 $charset = "utf-8";
  83
  84         $pos = strpos($header, "\r\n\r\n");
  85
  86         if ($pos)
  87                 $body = trim(substr($header, $pos));
  88         else
  89                 $body = $header;
  90
  91         $body = mb_convert_encoding($body, "UTF-8", $charset);
  92         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
  93
  94         $doc = new DOMDocument();
  95         @$doc->loadHTML($body);
  96
  97         deletenode($doc, 'style');
  98         deletenode($doc, 'script');
  99         deletenode($doc, 'option');
 100         deletenode($doc, 'h1');
 101         deletenode($doc, 'h2');
 102         deletenode($doc, 'h3');
 103         deletenode($doc, 'h4');
 104         deletenode($doc, 'h5');
 105         deletenode($doc, 'h6');
 106         deletenode($doc, 'ol');
 107         deletenode($doc, 'ul');
 108
 109         $xpath = new DomXPath($doc);
 110
 111         $list = $xpath->query("//meta[@content]");
 112         foreach ($list as $node) {
 113                 $attr = array();
 114                 if ($node->attributes->length)
 115                         foreach ($node->attributes as $attribute)
 116                                 $attr[$attribute->name] = $attribute->value;
 117
 118                 if (@$attr["http-equiv"] == 'refresh') {
 119                         $path = $attr["content"];
 120                         $pathinfo = explode(";", $path);
 121                         $content = "";
 122                         foreach ($pathinfo AS $value) {
 123                                 if (substr(strtolower($value), 0, 4) == "url=")
 124                                         $content = substr($value, 4);
 125                         }
 126                         if ($content != "") {
 127                                 $siteinfo = parseurl_getsiteinfo($content);
 128                                 return($siteinfo);
 129                         }
 130                 }
 131         }
 132
 133         //$list = $xpath->query("head/title");
 134         $list = $xpath->query("//title");
 135         foreach ($list as $node)
 136                 $siteinfo["title"] =  html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
 137
 138         //$list = $xpath->query("head/meta[@name]");
 139         $list = $xpath->query("//meta[@name]");
 140         foreach ($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                 $attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
 147
 148                 switch (strtolower($attr["name"])) {
 149                         case "fulltitle":
 150                                 $siteinfo["title"] = $attr["content"];
 151                                 break;
 152                         case "description":
 153                                 $siteinfo["text"] = $attr["content"];
 154                                 break;
 155                         case "dc.title":
 156                                 $siteinfo["title"] = $attr["content"];
 157                                 break;
 158                         case "dc.description":
 159                                 $siteinfo["text"] = $attr["content"];
 160                                 break;
 161                 }
 162         }
 163
 164         //$list = $xpath->query("head/meta[@property]");
 165         $list = $xpath->query("//meta[@property]");
 166         foreach ($list as $node) {
 167                 $attr = array();
 168                 if ($node->attributes->length)
 169                         foreach ($node->attributes as $attribute)
 170                                 $attr[$attribute->name] = $attribute->value;
 171
 172                 $attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
 173
 174                 switch (strtolower($attr["property"])) {
 175                         case "og:image":
 176                                 $siteinfo["image"] = $attr["content"];
 177                                 break;
 178                         case "og:title":
 179                                 $siteinfo["title"] = $attr["content"];
 180                                 break;
 181                         case "og:description":
 182                                 $siteinfo["text"] = $attr["content"];
 183                                 break;
 184                 }
 185         }
 186
 187         if (@$siteinfo["image"] == "") {
 188             $list = $xpath->query("//img[@src]");
 189             foreach ($list as $node) {
 190                 $attr = array();
 191                 if ($node->attributes->length)
 192                     foreach ($node->attributes as $attribute)
 193                         $attr[$attribute->name] = $attribute->value;
 194
 195                         $src = completeurl($attr["src"], $url);
 196                         $photodata = @getimagesize($src);
 197
 198                         if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
 199                                 if ($photodata[0] > 300) {
 200                                         $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
 201                                         $photodata[0] = 300;
 202                                 }
 203                                 if ($photodata[1] > 300) {
 204                                         $photodata[0] = round($photodata[0] * (300 / $photodata[1]));
 205                                         $photodata[1] = 300;
 206                                 }
 207                                 $siteinfo["images"][] = array("src"=>$src,
 208                                                                 "width"=>$photodata[0],
 209                                                                 "height"=>$photodata[1]);
 210                         }
 211
 212                 }
 213     } else {
 214                 $src = completeurl($siteinfo["image"], $url);
 215
 216                 unset($siteinfo["image"]);
 217
 218                 $photodata = @getimagesize($src);
 219
 220                 if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
 221                         $siteinfo["images"][] = array("src"=>$src,
 222                                                         "width"=>$photodata[0],
 223                                                         "height"=>$photodata[1]);
 224         }
 225
 226         if (@$siteinfo["text"] == "") {
 227                 $text = "";
 228
 229                 $list = $xpath->query("//div[@class='article']");
 230                 foreach ($list as $node)
 231                         if (strlen($node->nodeValue) > 40)
 232                                 $text .= " ".trim($node->nodeValue);
 233
 234                 if ($text == "") {
 235                         $list = $xpath->query("//div[@class='content']");
 236                         foreach ($list as $node)
 237                                 if (strlen($node->nodeValue) > 40)
 238                                         $text .= " ".trim($node->nodeValue);
 239                 }
 240
 241                 // If none text was found then take the paragraph content
 242                 if ($text == "") {
 243                         $list = $xpath->query("//p");
 244                         foreach ($list as $node)
 245                                 if (strlen($node->nodeValue) > 40)
 246                                         $text .= " ".trim($node->nodeValue);
 247                 }
 248
 249                 if ($text != "") {
 250                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 251
 252                         while (strpos($text, "  "))
 253                                 $text = trim(str_replace("  ", " ", $text));
 254
 255                         $siteinfo["text"] = html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...';
 256                 }
 257         }
 258
 259         return($siteinfo);
 260 }
 261
 262 function arr_add_hashes(&$item,$k) {
 263         $item = '#' . $item;
 264 }
 265
 266 function parse_url_content(&$a) {
 267
 268         $text = null;
 269         $str_tags = '';
 270
 271         $textmode = false;
 272
 273         if(local_user() && (! feature_enabled(local_user(),'richtext')))
 274                 $textmode = true;
 275
 276         //if($textmode)
 277         $br = (($textmode) ? "\n" : '<br />');
 278
 279         if(x($_GET,'binurl'))
 280                 $url = trim(hex2bin($_GET['binurl']));
 281         else
 282                 $url = trim($_GET['url']);
 283
 284         if($_GET['title'])
 285                 $title = strip_tags(trim($_GET['title']));
 286
 287         if($_GET['description'])
 288                 $text = strip_tags(trim($_GET['description']));
 289
 290         if($_GET['tags']) {
 291                 $arr_tags = str_getcsv($_GET['tags']);
 292                 if(count($arr_tags)) {
 293                         array_walk($arr_tags,'arr_add_hashes');
 294                         $str_tags = $br . implode(' ',$arr_tags) . $br;
 295                 }
 296         }
 297
 298         logger('parse_url: ' . $url);
 299
 300         if($textmode)
 301                 $template = $br . '[bookmark=%s]%s[/bookmark]%s' . $br;
 302         else
 303                 $template = "<br /><a class=\"bookmark\" href=\"%s\" >%s</a>%s<br />";
 304
 305         $arr = array('url' => $url, 'text' => '');
 306
 307         call_hooks('parse_link', $arr);
 308
 309         if(strlen($arr['text'])) {
 310                 echo $arr['text'];
 311                 killme();
 312         }
 313
 314
 315         if($url && $title && $text) {
 316
 317                 if($textmode)
 318                         $text = $br . '[quote]' . trim($text) . '[/quote]' . $br;
 319                 else
 320                         $text = '<br /><blockquote>' . trim($text) . '</blockquote><br />';
 321
 322                 $title = str_replace(array("\r","\n"),array('',''),$title);
 323
 324                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 325
 326                 logger('parse_url (unparsed): returns: ' . $result);
 327
 328                 echo $result;
 329                 killme();
 330         }
 331
 332         $siteinfo = parseurl_getsiteinfo($url);
 333
 334         if($siteinfo["title"] == "") {
 335                 echo sprintf($template,$url,$url,'') . $str_tags;
 336                 killme();
 337         } else {
 338                 $text = $siteinfo["text"];
 339                 $title = $siteinfo["title"];
 340         }
 341
 342         $image = "";
 343
 344         if(sizeof($siteinfo["images"]) > 0){
 345                 /* Execute below code only if image is present in siteinfo */
 346
 347                 $total_images = 0;
 348                 $max_images = get_config('system','max_bookmark_images');
 349                 if($max_images === false)
 350                         $max_images = 2;
 351                 else
 352                         $max_images = intval($max_images);
 353
 354                 foreach ($siteinfo["images"] as $imagedata) {
 355                         if($textmode)
 356                                 $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]' . "\n";
 357                         else
 358                                 $image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" /><br />';
 359                         $total_images ++;
 360                         if($max_images && $max_images >= $total_images)
 361                                 break;
 362         }
 363         }
 364
 365         if(strlen($text)) {
 366                 if($textmode)
 367                         $text = $br.'[quote]'.trim($text).'[/quote]'.$br ;
 368                 else
 369                         $text = '<br /><blockquote>'.trim($text).'</blockquote><br />';
 370         }
 371
 372         if($image) {
 373                 $text = $br.$br.$image.$text;
 374         }
 375         $title = str_replace(array("\r","\n"),array('',''),$title);
 376
 377         $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 378
 379         logger('parse_url: returns: ' . $result);
 380
 381         echo trim($result);
 382         killme();
 383 }