mod/parse_url.php

   1 <?php
   2 if(!function_exists('deletenode')) {
   3         function deletenode(&$doc, $node)
   4         {
   5                 $xpath = new DomXPath($doc);
   6                 $list = $xpath->query("//".$node);
   7                 foreach ($list as $child)
   8                         $child->parentNode->removeChild($child);
   9         }
  10 }
  11
  12 function parseurl_getsiteinfo($url) {
  13         $siteinfo = array();
  14
  15         $ch = curl_init();
  16         curl_setopt($ch, CURLOPT_URL, $url);
  17         curl_setopt($ch, CURLOPT_HEADER, 1);
  18         curl_setopt($ch, CURLOPT_NOBODY, 0);
  19         curl_setopt($ch, CURLOPT_TIMEOUT, 3);
  20         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
  21         curl_setopt($ch,CURLOPT_USERAGENT,'Opera/9.64(Windows NT 5.1; U; de) Presto/2.1.1');
  22
  23         $header = curl_exec($ch);
  24         curl_close($ch);
  25
  26         if (preg_match('/charset=(.*?)\n/', $header, $matches))
  27                 $charset = trim(array_pop($matches));
  28         else
  29                 $charset = "utf-8";
  30
  31         $pos = strpos($header, "\r\n\r\n");
  32
  33         if ($pos)
  34                 $body = trim(substr($header, $pos));
  35         else
  36                 $body = $header;
  37
  38         $body = mb_convert_encoding($body, "UTF-8", $charset);
  39         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
  40
  41         $doc = new DOMDocument();
  42         @$doc->loadHTML($body);
  43
  44         deletenode($doc, 'style');
  45         deletenode($doc, 'script');
  46         deletenode($doc, 'option');
  47         deletenode($doc, 'h1');
  48         deletenode($doc, 'h2');
  49         deletenode($doc, 'h3');
  50         deletenode($doc, 'h4');
  51         deletenode($doc, 'h5');
  52         deletenode($doc, 'h6');
  53         deletenode($doc, 'ol');
  54         deletenode($doc, 'ul');
  55
  56         $xpath = new DomXPath($doc);
  57
  58         $list = $xpath->query("head/title");
  59         foreach ($list as $node)
  60                 $siteinfo["title"] =  html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
  61
  62         $list = $xpath->query("head/meta[@name]");
  63         foreach ($list as $node) {
  64                 $attr = array();
  65                 if ($node->attributes->length)
  66                         foreach ($node->attributes as $attribute)
  67                                 $attr[$attribute->name] = $attribute->value;
  68
  69                 $attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
  70
  71                 switch (strtolower($attr["name"])) {
  72                         case "fulltitle":
  73                                 $siteinfo["title"] = $attr["content"];
  74                                 break;
  75                         case "description":
  76                                 $siteinfo["text"] = $attr["content"];
  77                                 break;
  78                         case "dc.title":
  79                                 $siteinfo["title"] = $attr["content"];
  80                                 break;
  81                         case "dc.description":
  82                                 $siteinfo["text"] = $attr["content"];
  83                                 break;
  84                 }
  85         }
  86
  87         $list = $xpath->query("head/meta[@property]");
  88         foreach ($list as $node) {
  89                 $attr = array();
  90                 if ($node->attributes->length)
  91                         foreach ($node->attributes as $attribute)
  92                                 $attr[$attribute->name] = $attribute->value;
  93
  94                 $attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
  95
  96                 switch (strtolower($attr["property"])) {
  97                         case "og:image":
  98                                 $siteinfo["image"] = $attr["content"];
  99                                 break;
 100                         case "og:title":
 101                                 $siteinfo["title"] = $attr["content"];
 102                                 break;
 103                         case "og:description":
 104                                 $siteinfo["text"] = $attr["content"];
 105                                 break;
 106                 }
 107         }
 108
 109         if ($siteinfo["image"] == "") {
 110                 require_once('include/Photo.php');
 111                 $list = $xpath->query("//img[@src]");
 112                 foreach ($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                         // guess mimetype from headers or filename
 119                         $type = guess_image_type($attr["src"],true);
 120
 121                         $i = fetch_url($attr["src"]);
 122                         $ph = new Photo($i, $type);
 123
 124                         if(($ph->getWidth() > 200) and ($ph->getHeight() > 200))
 125                                 $siteinfo["image"] = $attr["src"];
 126                 }
 127         }
 128
 129         if ($siteinfo["text"] == "") {
 130                 $text = "";
 131
 132                 $list = $xpath->query("//div[@class='article']");
 133                 foreach ($list as $node)
 134                         $text .= " ".trim($node->nodeValue);
 135
 136                 if ($text == "") {
 137                         $list = $xpath->query("//div[@class='content']");
 138                         foreach ($list as $node)
 139                                 $text .= " ".trim($node->nodeValue);
 140                 }
 141
 142                 // If none text was found then take the paragraph content
 143                 if ($text == "") {
 144                         $list = $xpath->query("//p");
 145                         foreach ($list as $node)
 146                                 $text .= " ".trim($node->nodeValue);
 147                 }
 148
 149                 if ($text != "") {
 150                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 151
 152                         while (strpos($text, "  "))
 153                                 $text = trim(str_replace("  ", " ", $text));
 154
 155                         $siteinfo["text"] = html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...';
 156                 }
 157         }
 158
 159         return($siteinfo);
 160 }
 161
 162 function arr_add_hashes(&$item,$k) {
 163         $item = '#' . $item;
 164 }
 165
 166 function parse_url_content(&$a) {
 167
 168         $text = null;
 169         $str_tags = '';
 170
 171         $textmode = false;
 172         if(local_user() && intval(get_pconfig(local_user(),'system','plaintext')))
 173                 $textmode = true;
 174
 175         if($textmode)
 176         $br = (($textmode) ? "\n" : '<br /?');
 177
 178         if(x($_GET,'binurl'))
 179                 $url = trim(hex2bin($_GET['binurl']));
 180         else
 181                 $url = trim($_GET['url']);
 182
 183         if($_GET['title'])
 184                 $title = strip_tags(trim($_GET['title']));
 185
 186         if($_GET['description'])
 187                 $text = strip_tags(trim($_GET['description']));
 188
 189         if($_GET['tags']) {
 190                 $arr_tags = str_getcsv($_GET['tags']);
 191                 if(count($arr_tags)) {
 192                         array_walk($arr_tags,'arr_add_hashes');
 193                         $str_tags = $br . implode(' ',$arr_tags) . $br;
 194                 }
 195         }
 196
 197         logger('parse_url: ' . $url);
 198
 199         if($textmode)
 200                 $template = $br . '[bookmark=%s]%s[/bookmark]%s' . $br;
 201         else
 202                 $template = "<br /><a class=\"bookmark\" href=\"%s\" >%s</a>%s<br />";
 203
 204         $arr = array('url' => $url, 'text' => '');
 205
 206         call_hooks('parse_link', $arr);
 207
 208         if(strlen($arr['text'])) {
 209                 echo $arr['text'];
 210                 killme();
 211         }
 212
 213
 214         if($url && $title && $text) {
 215
 216                 if($textmode)
 217                         $text = $br . $br . '[quote]' . trim($text) . '[/quote]' . $br;
 218                 else
 219                         $text = '<br /><br /><blockquote>' . trim($text) . '</blockquote><br />';
 220
 221                 $title = str_replace(array("\r","\n"),array('',''),$title);
 222
 223                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 224
 225                 logger('parse_url (unparsed): returns: ' . $result);
 226
 227                 echo $result;
 228                 killme();
 229         }
 230
 231         $siteinfo = parseurl_getsiteinfo($url);
 232
 233         if($siteinfo["title"] == "") {
 234                 echo sprintf($template,$url,$url,'') . $str_tags;
 235                 killme();
 236         } else {
 237                 $image = $siteinfo["image"];
 238                 $text = $siteinfo["text"];
 239                 $title = $siteinfo["title"];
 240         }
 241
 242         if ($image != "") {
 243                 $i = fetch_url($image);
 244                 if($i) {
 245                         require_once('include/Photo.php');
 246                         // guess mimetype from headers or filename
 247                         $type = guess_image_type($image,true);
 248
 249                         $ph = new Photo($i, $type);
 250                         if($ph->is_valid()) {
 251                                 if($ph->getWidth() > 300 || $ph->getHeight() > 300) {
 252                                         $ph->scaleImage(300);
 253                                         $new_width = $ph->getWidth();
 254                                         $new_height = $ph->getHeight();
 255                                         if($textmode)
 256                                                 $image = $br . $br . '[img=' . $new_width . 'x' . $new_height . ']' . $image . '[/img]';
 257                                         else
 258                                                 $image = '<br /><br /><img height="' . $new_height . '" width="' . $new_width . '" src="' .$image . '" alt="photo" />';
 259                                 } else {
 260                                         if($textmode)
 261                                                 $image = $br.$br.'[img]'.$image.'[/img]';
 262                                         else
 263                                                 $image = '<br /><br /><img src="'.$image.'" alt="photo" />';
 264                                 }
 265                         }
 266                 }
 267         }
 268
 269         if(strlen($text)) {
 270                 if($textmode)
 271                         $text = $br.$br.'[quote]'.trim($text).'[/quote]'.$br ;
 272                 else
 273                         $text = '<br /><br /><blockquote>'.trim($text).'</blockquote><br />';
 274         }
 275
 276         if($image) {
 277                 $text = $image.$br.$text;
 278         }
 279         $title = str_replace(array("\r","\n"),array('',''),$title);
 280
 281         $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 282
 283         logger('parse_url: returns: ' . $result);
 284
 285         echo $result;
 286         killme();
 287 }