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 completeurl($url, $scheme) {
  13         $urlarr = parse_url($url);
  14
  15         if (isset($urlarr["scheme"]))
  16                 return($url);
  17
  18         $schemearr = parse_url($scheme);
  19
  20         $complete = $schemearr["scheme"]."://".$schemearr["host"];
  21
  22         if ($schemearr["port"] != "")
  23                 $complete .= ":".$schemearr["port"];
  24
  25         $complete .= $urlarr["path"];
  26
  27         if ($urlarr["query"] != "")
  28                 $complete .= "?".$urlarr["query"];
  29
  30         if ($urlarr["fragment"] != "")
  31                 $complete .= "#".$urlarr["fragment"];
  32
  33         return($complete);
  34 }
  35
  36 function parseurl_getsiteinfo($url) {
  37         $siteinfo = array();
  38
  39         $ch = curl_init();
  40         curl_setopt($ch, CURLOPT_URL, $url);
  41         curl_setopt($ch, CURLOPT_HEADER, 1);
  42         curl_setopt($ch, CURLOPT_NOBODY, 0);
  43         curl_setopt($ch, CURLOPT_TIMEOUT, 3);
  44         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
  45         curl_setopt($ch,CURLOPT_USERAGENT,'Opera/9.64(Windows NT 5.1; U; de) Presto/2.1.1');
  46
  47         $header = curl_exec($ch);
  48         curl_close($ch);
  49
  50         // Fetch the first mentioned charset. Can be in body or header
  51         if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
  52                 $charset = trim(array_pop($matches));
  53         else
  54                 $charset = "utf-8";
  55
  56         $pos = strpos($header, "\r\n\r\n");
  57
  58         if ($pos)
  59                 $body = trim(substr($header, $pos));
  60         else
  61                 $body = $header;
  62
  63         $body = mb_convert_encoding($body, "UTF-8", $charset);
  64         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
  65
  66         $doc = new DOMDocument();
  67         @$doc->loadHTML($body);
  68
  69         deletenode($doc, 'style');
  70         deletenode($doc, 'script');
  71         deletenode($doc, 'option');
  72         deletenode($doc, 'h1');
  73         deletenode($doc, 'h2');
  74         deletenode($doc, 'h3');
  75         deletenode($doc, 'h4');
  76         deletenode($doc, 'h5');
  77         deletenode($doc, 'h6');
  78         deletenode($doc, 'ol');
  79         deletenode($doc, 'ul');
  80
  81         $xpath = new DomXPath($doc);
  82
  83         //$list = $xpath->query("head/title");
  84         $list = $xpath->query("//title");
  85         foreach ($list as $node)
  86                 $siteinfo["title"] =  html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
  87
  88         //$list = $xpath->query("head/meta[@name]");
  89         $list = $xpath->query("//meta[@name]");
  90         foreach ($list as $node) {
  91                 $attr = array();
  92                 if ($node->attributes->length)
  93                         foreach ($node->attributes as $attribute)
  94                                 $attr[$attribute->name] = $attribute->value;
  95
  96                 $attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
  97
  98                 switch (strtolower($attr["name"])) {
  99                         case "fulltitle":
 100                                 $siteinfo["title"] = $attr["content"];
 101                                 break;
 102                         case "description":
 103                                 $siteinfo["text"] = $attr["content"];
 104                                 break;
 105                         case "dc.title":
 106                                 $siteinfo["title"] = $attr["content"];
 107                                 break;
 108                         case "dc.description":
 109                                 $siteinfo["text"] = $attr["content"];
 110                                 break;
 111                 }
 112         }
 113
 114         //$list = $xpath->query("head/meta[@property]");
 115         $list = $xpath->query("//meta[@property]");
 116         foreach ($list as $node) {
 117                 $attr = array();
 118                 if ($node->attributes->length)
 119                         foreach ($node->attributes as $attribute)
 120                                 $attr[$attribute->name] = $attribute->value;
 121
 122                 $attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
 123
 124                 switch (strtolower($attr["property"])) {
 125                         case "og:image":
 126                                 $siteinfo["image"] = $attr["content"];
 127                                 break;
 128                         case "og:title":
 129                                 $siteinfo["title"] = $attr["content"];
 130                                 break;
 131                         case "og:description":
 132                                 $siteinfo["text"] = $attr["content"];
 133                                 break;
 134                 }
 135         }
 136
 137         if ($siteinfo["image"] == "") {
 138                 $list = $xpath->query("//img[@src]");
 139                 foreach ($list as $node) {
 140                         $attr = array();
 141                         if ($node->attributes->length)
 142                                 foreach ($node->attributes as $attribute)
 143                                         $attr[$attribute->name] = $attribute->value;
 144
 145                         $src = completeurl($attr["src"], $url);
 146                         $photodata = getimagesize($src);
 147
 148                         if (($photodata[0] > 150) and ($photodata[1] > 150)) {
 149                                 if ($photodata[0] > 300) {
 150                                         $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
 151                                         $photodata[0] = 300;
 152                                 }
 153                                 if ($photodata[1] > 300) {
 154                                         $photodata[0] = round($photodata[0] * (300 / $photodata[1]));
 155                                         $photodata[1] = 300;
 156                                 }
 157                                 $siteinfo["images"][] = array("src"=>$src,
 158                                                                 "width"=>$photodata[0],
 159                                                                 "height"=>$photodata[1]);
 160                         }
 161
 162                 }
 163         } else {
 164                 $src = completeurl($siteinfo["image"], $url);
 165                 $photodata = getimagesize($src);
 166
 167                 if (($photodata[0] > 10) and ($photodata[1] > 10))
 168                         $siteinfo["images"][] = array("src"=>$src,
 169                                                         "width"=>$photodata[0],
 170                                                         "height"=>$photodata[1]);
 171         }
 172
 173         if ($siteinfo["text"] == "") {
 174                 $text = "";
 175
 176                 $list = $xpath->query("//div[@class='article']");
 177                 foreach ($list as $node)
 178                         if (strlen($node->nodeValue) > 40)
 179                                 $text .= " ".trim($node->nodeValue);
 180
 181                 if ($text == "") {
 182                         $list = $xpath->query("//div[@class='content']");
 183                         foreach ($list as $node)
 184                                 if (strlen($node->nodeValue) > 40)
 185                                         $text .= " ".trim($node->nodeValue);
 186                 }
 187
 188                 // If none text was found then take the paragraph content
 189                 if ($text == "") {
 190                         $list = $xpath->query("//p");
 191                         foreach ($list as $node)
 192                                 if (strlen($node->nodeValue) > 40)
 193                                         $text .= " ".trim($node->nodeValue);
 194                 }
 195
 196                 if ($text != "") {
 197                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 198
 199                         while (strpos($text, "  "))
 200                                 $text = trim(str_replace("  ", " ", $text));
 201
 202                         $siteinfo["text"] = html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...';
 203                 }
 204         }
 205
 206         return($siteinfo);
 207 }
 208
 209 function arr_add_hashes(&$item,$k) {
 210         $item = '#' . $item;
 211 }
 212
 213 function parse_url_content(&$a) {
 214
 215         $text = null;
 216         $str_tags = '';
 217
 218         $textmode = false;
 219         if(local_user() && intval(get_pconfig(local_user(),'system','plaintext')))
 220                 $textmode = true;
 221
 222         //if($textmode)
 223         $br = (($textmode) ? "\n" : '<br />');
 224
 225         if(x($_GET,'binurl'))
 226                 $url = trim(hex2bin($_GET['binurl']));
 227         else
 228                 $url = trim($_GET['url']);
 229
 230         if($_GET['title'])
 231                 $title = strip_tags(trim($_GET['title']));
 232
 233         if($_GET['description'])
 234                 $text = strip_tags(trim($_GET['description']));
 235
 236         if($_GET['tags']) {
 237                 $arr_tags = str_getcsv($_GET['tags']);
 238                 if(count($arr_tags)) {
 239                         array_walk($arr_tags,'arr_add_hashes');
 240                         $str_tags = $br . implode(' ',$arr_tags) . $br;
 241                 }
 242         }
 243
 244         logger('parse_url: ' . $url);
 245
 246         if($textmode)
 247                 $template = $br . '[bookmark=%s]%s[/bookmark]%s' . $br;
 248         else
 249                 $template = "<br /><a class=\"bookmark\" href=\"%s\" >%s</a>%s<br />";
 250
 251         $arr = array('url' => $url, 'text' => '');
 252
 253         call_hooks('parse_link', $arr);
 254
 255         if(strlen($arr['text'])) {
 256                 echo $arr['text'];
 257                 killme();
 258         }
 259
 260
 261         if($url && $title && $text) {
 262
 263                 if($textmode)
 264                         $text = $br . '[quote]' . trim($text) . '[/quote]' . $br;
 265                 else
 266                         $text = '<br /><blockquote>' . trim($text) . '</blockquote><br />';
 267
 268                 $title = str_replace(array("\r","\n"),array('',''),$title);
 269
 270                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 271
 272                 logger('parse_url (unparsed): returns: ' . $result);
 273
 274                 echo $result;
 275                 killme();
 276         }
 277
 278         $siteinfo = parseurl_getsiteinfo($url);
 279
 280         if($siteinfo["title"] == "") {
 281                 echo print_r($siteinfo, true);
 282                 //echo sprintf($template,$url,$url,'') . $str_tags;
 283                 killme();
 284         } else {
 285                 $text = $siteinfo["text"];
 286                 $title = $siteinfo["title"];
 287         }
 288
 289         $image = "";
 290
 291         foreach ($siteinfo["images"] as $imagedata)
 292                 if($textmode)
 293                         $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]';
 294                 else
 295                         $image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" />';
 296
 297         if(strlen($text)) {
 298                 if($textmode)
 299                         $text = $br.'[quote]'.trim($text).'[/quote]'.$br ;
 300                 else
 301                         $text = '<br /><blockquote>'.trim($text).'</blockquote><br />';
 302         }
 303
 304         if($image) {
 305                 $text = $br.$br.$image.$text;
 306         }
 307         $title = str_replace(array("\r","\n"),array('',''),$title);
 308
 309         $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 310
 311         logger('parse_url: returns: ' . $result);
 312
 313         echo trim($result);
 314         killme();
 315 }