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
  56         $ch = curl_init();
  57         curl_setopt($ch, CURLOPT_URL, $url);
  58         curl_setopt($ch, CURLOPT_HEADER, 1);
  59         curl_setopt($ch, CURLOPT_NOBODY, 0);
  60         curl_setopt($ch, CURLOPT_TIMEOUT, 3);
  61         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
  62         curl_setopt($ch,CURLOPT_USERAGENT,'Opera/9.64(Windows NT 5.1; U; de) Presto/2.1.1');
  63
  64         $header = curl_exec($ch);
  65         curl_close($ch);
  66
  67         // Fetch the first mentioned charset. Can be in body or header
  68         if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
  69                 $charset = trim(array_pop($matches));
  70         else
  71                 $charset = "utf-8";
  72
  73         $pos = strpos($header, "\r\n\r\n");
  74
  75         if ($pos)
  76                 $body = trim(substr($header, $pos));
  77         else
  78                 $body = $header;
  79
  80         $body = mb_convert_encoding($body, "UTF-8", $charset);
  81         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
  82
  83         $doc = new DOMDocument();
  84         @$doc->loadHTML($body);
  85
  86         deletenode($doc, 'style');
  87         deletenode($doc, 'script');
  88         deletenode($doc, 'option');
  89         deletenode($doc, 'h1');
  90         deletenode($doc, 'h2');
  91         deletenode($doc, 'h3');
  92         deletenode($doc, 'h4');
  93         deletenode($doc, 'h5');
  94         deletenode($doc, 'h6');
  95         deletenode($doc, 'ol');
  96         deletenode($doc, 'ul');
  97
  98         $xpath = new DomXPath($doc);
  99
 100         $list = $xpath->query("//meta[@content]");
 101         foreach ($list as $node) {
 102                 $attr = array();
 103                 if ($node->attributes->length)
 104                         foreach ($node->attributes as $attribute)
 105                                 $attr[$attribute->name] = $attribute->value;
 106
 107                 if (@$attr["http-equiv"] == 'refresh') {
 108                         $path = $attr["content"];
 109                         $pathinfo = explode(";", $path);
 110                         $content = "";
 111                         foreach ($pathinfo AS $value) {
 112                                 if (substr(strtolower($value), 0, 4) == "url=")
 113                                         $content = substr($value, 4);
 114                         }
 115                         if ($content != "") {
 116                                 $siteinfo = parseurl_getsiteinfo($content);
 117                                 return($siteinfo);
 118                         }
 119                 }
 120         }
 121
 122         //$list = $xpath->query("head/title");
 123         $list = $xpath->query("//title");
 124         foreach ($list as $node)
 125                 $siteinfo["title"] =  html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
 126
 127         //$list = $xpath->query("head/meta[@name]");
 128         $list = $xpath->query("//meta[@name]");
 129         foreach ($list as $node) {
 130                 $attr = array();
 131                 if ($node->attributes->length)
 132                         foreach ($node->attributes as $attribute)
 133                                 $attr[$attribute->name] = $attribute->value;
 134
 135                 $attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
 136
 137                 switch (strtolower($attr["name"])) {
 138                         case "fulltitle":
 139                                 $siteinfo["title"] = $attr["content"];
 140                                 break;
 141                         case "description":
 142                                 $siteinfo["text"] = $attr["content"];
 143                                 break;
 144                         case "dc.title":
 145                                 $siteinfo["title"] = $attr["content"];
 146                                 break;
 147                         case "dc.description":
 148                                 $siteinfo["text"] = $attr["content"];
 149                                 break;
 150                 }
 151         }
 152
 153         //$list = $xpath->query("head/meta[@property]");
 154         $list = $xpath->query("//meta[@property]");
 155         foreach ($list as $node) {
 156                 $attr = array();
 157                 if ($node->attributes->length)
 158                         foreach ($node->attributes as $attribute)
 159                                 $attr[$attribute->name] = $attribute->value;
 160
 161                 $attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
 162
 163                 switch (strtolower($attr["property"])) {
 164                         case "og:image":
 165                                 $siteinfo["image"] = $attr["content"];
 166                                 break;
 167                         case "og:title":
 168                                 $siteinfo["title"] = $attr["content"];
 169                                 break;
 170                         case "og:description":
 171                                 $siteinfo["text"] = $attr["content"];
 172                                 break;
 173                 }
 174         }
 175
 176         if ($siteinfo["image"] == "") {
 177             $list = $xpath->query("//img[@src]");
 178             foreach ($list as $node) {
 179                 $attr = array();
 180                 if ($node->attributes->length)
 181                     foreach ($node->attributes as $attribute)
 182                         $attr[$attribute->name] = $attribute->value;
 183
 184                         $src = completeurl($attr["src"], $url);
 185                         $photodata = @getimagesize($src);
 186
 187                         if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
 188                                 if ($photodata[0] > 300) {
 189                                         $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
 190                                         $photodata[0] = 300;
 191                                 }
 192                                 if ($photodata[1] > 300) {
 193                                         $photodata[0] = round($photodata[0] * (300 / $photodata[1]));
 194                                         $photodata[1] = 300;
 195                                 }
 196                                 $siteinfo["images"][] = array("src"=>$src,
 197                                                                 "width"=>$photodata[0],
 198                                                                 "height"=>$photodata[1]);
 199                         }
 200
 201                 }
 202     } else {
 203                 $src = completeurl($siteinfo["image"], $url);
 204
 205                 unset($siteinfo["image"]);
 206
 207                 $photodata = @getimagesize($src);
 208
 209                 if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
 210                         $siteinfo["images"][] = array("src"=>$src,
 211                                                         "width"=>$photodata[0],
 212                                                         "height"=>$photodata[1]);
 213         }
 214
 215         if ($siteinfo["text"] == "") {
 216                 $text = "";
 217
 218                 $list = $xpath->query("//div[@class='article']");
 219                 foreach ($list as $node)
 220                         if (strlen($node->nodeValue) > 40)
 221                                 $text .= " ".trim($node->nodeValue);
 222
 223                 if ($text == "") {
 224                         $list = $xpath->query("//div[@class='content']");
 225                         foreach ($list as $node)
 226                                 if (strlen($node->nodeValue) > 40)
 227                                         $text .= " ".trim($node->nodeValue);
 228                 }
 229
 230                 // If none text was found then take the paragraph content
 231                 if ($text == "") {
 232                         $list = $xpath->query("//p");
 233                         foreach ($list as $node)
 234                                 if (strlen($node->nodeValue) > 40)
 235                                         $text .= " ".trim($node->nodeValue);
 236                 }
 237
 238                 if ($text != "") {
 239                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 240
 241                         while (strpos($text, "  "))
 242                                 $text = trim(str_replace("  ", " ", $text));
 243
 244                         $siteinfo["text"] = html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...';
 245                 }
 246         }
 247
 248         return($siteinfo);
 249 }
 250
 251 function arr_add_hashes(&$item,$k) {
 252         $item = '#' . $item;
 253 }
 254
 255 function parse_url_content(&$a) {
 256
 257         $text = null;
 258         $str_tags = '';
 259
 260         $textmode = false;
 261
 262         if(local_user() && (! feature_enabled(local_user(),'richtext')))
 263                 $textmode = true;
 264
 265         //if($textmode)
 266         $br = (($textmode) ? "\n" : '<br />');
 267
 268         if(x($_GET,'binurl'))
 269                 $url = trim(hex2bin($_GET['binurl']));
 270         else
 271                 $url = trim($_GET['url']);
 272
 273         if($_GET['title'])
 274                 $title = strip_tags(trim($_GET['title']));
 275
 276         if($_GET['description'])
 277                 $text = strip_tags(trim($_GET['description']));
 278
 279         if($_GET['tags']) {
 280                 $arr_tags = str_getcsv($_GET['tags']);
 281                 if(count($arr_tags)) {
 282                         array_walk($arr_tags,'arr_add_hashes');
 283                         $str_tags = $br . implode(' ',$arr_tags) . $br;
 284                 }
 285         }
 286
 287         logger('parse_url: ' . $url);
 288
 289         if($textmode)
 290                 $template = $br . '[bookmark=%s]%s[/bookmark]%s' . $br;
 291         else
 292                 $template = "<br /><a class=\"bookmark\" href=\"%s\" >%s</a>%s<br />";
 293
 294         $arr = array('url' => $url, 'text' => '');
 295
 296         call_hooks('parse_link', $arr);
 297
 298         if(strlen($arr['text'])) {
 299                 echo $arr['text'];
 300                 killme();
 301         }
 302
 303
 304         if($url && $title && $text) {
 305
 306                 if($textmode)
 307                         $text = $br . '[quote]' . trim($text) . '[/quote]' . $br;
 308                 else
 309                         $text = '<br /><blockquote>' . trim($text) . '</blockquote><br />';
 310
 311                 $title = str_replace(array("\r","\n"),array('',''),$title);
 312
 313                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 314
 315                 logger('parse_url (unparsed): returns: ' . $result);
 316
 317                 echo $result;
 318                 killme();
 319         }
 320
 321         $siteinfo = parseurl_getsiteinfo($url);
 322
 323         if($siteinfo["title"] == "") {
 324                 echo sprintf($template,$url,$url,'') . $str_tags;
 325                 killme();
 326         } else {
 327                 $text = $siteinfo["text"];
 328                 $title = $siteinfo["title"];
 329         }
 330
 331         $image = "";
 332
 333         if(sizeof($siteinfo["images"]) > 0){
 334                 /* Execute below code only if image is present in siteinfo */
 335
 336                 $total_images = 0;
 337                 $max_images = get_config('system','max_bookmark_images');
 338                 if($max_images === false)
 339                         $max_images = 2;
 340                 else
 341                         $max_images = intval($max_images);
 342
 343                 foreach ($siteinfo["images"] as $imagedata) {
 344                         if($textmode)
 345                                 $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]' . "\n";
 346                         else
 347                                 $image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" /><br />';
 348                         $total_images ++;
 349                         if($max_images && $max_images >= $total_images)
 350                                 break;
 351         }
 352         }
 353
 354         if(strlen($text)) {
 355                 if($textmode)
 356                         $text = $br.'[quote]'.trim($text).'[/quote]'.$br ;
 357                 else
 358                         $text = '<br /><blockquote>'.trim($text).'</blockquote><br />';
 359         }
 360
 361         if($image) {
 362                 $text = $br.$br.$image.$text;
 363         }
 364         $title = str_replace(array("\r","\n"),array('',''),$title);
 365
 366         $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 367
 368         logger('parse_url: returns: ' . $result);
 369
 370         echo trim($result);
 371         killme();
 372 }