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         $complete .= $urlarr["path"];
  40
  41         if ($urlarr["query"] != "")
  42                 $complete .= "?".$urlarr["query"];
  43
  44         if ($urlarr["fragment"] != "")
  45                 $complete .= "#".$urlarr["fragment"];
  46
  47         return($complete);
  48 }
  49
  50 function parseurl_getsiteinfo($url) {
  51         $siteinfo = array();
  52
  53         $ch = curl_init();
  54         curl_setopt($ch, CURLOPT_URL, $url);
  55         curl_setopt($ch, CURLOPT_HEADER, 1);
  56         curl_setopt($ch, CURLOPT_NOBODY, 0);
  57         curl_setopt($ch, CURLOPT_TIMEOUT, 3);
  58         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
  59         curl_setopt($ch,CURLOPT_USERAGENT,'Opera/9.64(Windows NT 5.1; U; de) Presto/2.1.1');
  60
  61         $header = curl_exec($ch);
  62         curl_close($ch);
  63
  64         // Fetch the first mentioned charset. Can be in body or header
  65         if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
  66                 $charset = trim(array_pop($matches));
  67         else
  68                 $charset = "utf-8";
  69
  70         $pos = strpos($header, "\r\n\r\n");
  71
  72         if ($pos)
  73                 $body = trim(substr($header, $pos));
  74         else
  75                 $body = $header;
  76
  77         $body = mb_convert_encoding($body, "UTF-8", $charset);
  78         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
  79
  80         $doc = new DOMDocument();
  81         @$doc->loadHTML($body);
  82
  83         deletenode($doc, 'style');
  84         deletenode($doc, 'script');
  85         deletenode($doc, 'option');
  86         deletenode($doc, 'h1');
  87         deletenode($doc, 'h2');
  88         deletenode($doc, 'h3');
  89         deletenode($doc, 'h4');
  90         deletenode($doc, 'h5');
  91         deletenode($doc, 'h6');
  92         deletenode($doc, 'ol');
  93         deletenode($doc, 'ul');
  94
  95         $xpath = new DomXPath($doc);
  96
  97         //$list = $xpath->query("head/title");
  98         $list = $xpath->query("//title");
  99         foreach ($list as $node)
 100                 $siteinfo["title"] =  html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
 101
 102         //$list = $xpath->query("head/meta[@name]");
 103         $list = $xpath->query("//meta[@name]");
 104         foreach ($list as $node) {
 105                 $attr = array();
 106                 if ($node->attributes->length)
 107                         foreach ($node->attributes as $attribute)
 108                                 $attr[$attribute->name] = $attribute->value;
 109
 110                 $attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
 111
 112                 switch (strtolower($attr["name"])) {
 113                         case "fulltitle":
 114                                 $siteinfo["title"] = $attr["content"];
 115                                 break;
 116                         case "description":
 117                                 $siteinfo["text"] = $attr["content"];
 118                                 break;
 119                         case "dc.title":
 120                                 $siteinfo["title"] = $attr["content"];
 121                                 break;
 122                         case "dc.description":
 123                                 $siteinfo["text"] = $attr["content"];
 124                                 break;
 125                 }
 126         }
 127
 128         //$list = $xpath->query("head/meta[@property]");
 129         $list = $xpath->query("//meta[@property]");
 130         foreach ($list as $node) {
 131                 $attr = array();
 132                 if ($node->attributes->length)
 133                         foreach ($node->attributes as $attribute)
 134                                 $attr[$attribute->name] = $attribute->value;
 135
 136                 $attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
 137
 138                 switch (strtolower($attr["property"])) {
 139                         case "og:image":
 140                                 $siteinfo["image"] = $attr["content"];
 141                                 break;
 142                         case "og:title":
 143                                 $siteinfo["title"] = $attr["content"];
 144                                 break;
 145                         case "og:description":
 146                                 $siteinfo["text"] = $attr["content"];
 147                                 break;
 148                 }
 149         }
 150
 151         if ($siteinfo["image"] == "") {
 152                 $list = $xpath->query("//img[@src]");
 153                 foreach ($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                         $src = completeurl($attr["src"], $url);
 160                         $photodata = getimagesize($src);
 161
 162                         if (($photodata[0] > 150) and ($photodata[1] > 150)) {
 163                                 if ($photodata[0] > 300) {
 164                                         $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
 165                                         $photodata[0] = 300;
 166                                 }
 167                                 if ($photodata[1] > 300) {
 168                                         $photodata[0] = round($photodata[0] * (300 / $photodata[1]));
 169                                         $photodata[1] = 300;
 170                                 }
 171                                 $siteinfo["images"][] = array("src"=>$src,
 172                                                                 "width"=>$photodata[0],
 173                                                                 "height"=>$photodata[1]);
 174                         }
 175
 176                 }
 177         } else {
 178                 $src = completeurl($siteinfo["image"], $url);
 179                 $photodata = getimagesize($src);
 180
 181                 if (($photodata[0] > 10) and ($photodata[1] > 10))
 182                         $siteinfo["images"][] = array("src"=>$src,
 183                                                         "width"=>$photodata[0],
 184                                                         "height"=>$photodata[1]);
 185         }
 186
 187         if ($siteinfo["text"] == "") {
 188                 $text = "";
 189
 190                 $list = $xpath->query("//div[@class='article']");
 191                 foreach ($list as $node)
 192                         if (strlen($node->nodeValue) > 40)
 193                                 $text .= " ".trim($node->nodeValue);
 194
 195                 if ($text == "") {
 196                         $list = $xpath->query("//div[@class='content']");
 197                         foreach ($list as $node)
 198                                 if (strlen($node->nodeValue) > 40)
 199                                         $text .= " ".trim($node->nodeValue);
 200                 }
 201
 202                 // If none text was found then take the paragraph content
 203                 if ($text == "") {
 204                         $list = $xpath->query("//p");
 205                         foreach ($list as $node)
 206                                 if (strlen($node->nodeValue) > 40)
 207                                         $text .= " ".trim($node->nodeValue);
 208                 }
 209
 210                 if ($text != "") {
 211                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 212
 213                         while (strpos($text, "  "))
 214                                 $text = trim(str_replace("  ", " ", $text));
 215
 216                         $siteinfo["text"] = html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...';
 217                 }
 218         }
 219
 220         return($siteinfo);
 221 }
 222
 223 function arr_add_hashes(&$item,$k) {
 224         $item = '#' . $item;
 225 }
 226
 227 function parse_url_content(&$a) {
 228
 229         $text = null;
 230         $str_tags = '';
 231
 232         $textmode = false;
 233         if(local_user() && intval(get_pconfig(local_user(),'system','plaintext')))
 234                 $textmode = true;
 235
 236         //if($textmode)
 237         $br = (($textmode) ? "\n" : '<br />');
 238
 239         if(x($_GET,'binurl'))
 240                 $url = trim(hex2bin($_GET['binurl']));
 241         else
 242                 $url = trim($_GET['url']);
 243
 244         if($_GET['title'])
 245                 $title = strip_tags(trim($_GET['title']));
 246
 247         if($_GET['description'])
 248                 $text = strip_tags(trim($_GET['description']));
 249
 250         if($_GET['tags']) {
 251                 $arr_tags = str_getcsv($_GET['tags']);
 252                 if(count($arr_tags)) {
 253                         array_walk($arr_tags,'arr_add_hashes');
 254                         $str_tags = $br . implode(' ',$arr_tags) . $br;
 255                 }
 256         }
 257
 258         logger('parse_url: ' . $url);
 259
 260         if($textmode)
 261                 $template = $br . '[bookmark=%s]%s[/bookmark]%s' . $br;
 262         else
 263                 $template = "<br /><a class=\"bookmark\" href=\"%s\" >%s</a>%s<br />";
 264
 265         $arr = array('url' => $url, 'text' => '');
 266
 267         call_hooks('parse_link', $arr);
 268
 269         if(strlen($arr['text'])) {
 270                 echo $arr['text'];
 271                 killme();
 272         }
 273
 274
 275         if($url && $title && $text) {
 276
 277                 if($textmode)
 278                         $text = $br . '[quote]' . trim($text) . '[/quote]' . $br;
 279                 else
 280                         $text = '<br /><blockquote>' . trim($text) . '</blockquote><br />';
 281
 282                 $title = str_replace(array("\r","\n"),array('',''),$title);
 283
 284                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 285
 286                 logger('parse_url (unparsed): returns: ' . $result);
 287
 288                 echo $result;
 289                 killme();
 290         }
 291
 292         $siteinfo = parseurl_getsiteinfo($url);
 293
 294         if($siteinfo["title"] == "") {
 295                 echo print_r($siteinfo, true);
 296                 //echo sprintf($template,$url,$url,'') . $str_tags;
 297                 killme();
 298         } else {
 299                 $text = $siteinfo["text"];
 300                 $title = $siteinfo["title"];
 301         }
 302
 303         $image = "";
 304
 305         if($siteinfo["image"] != ""){
 306             /*
 307               Execute below code only if image is present in siteinfo
 308              */
 309             foreach ($siteinfo["images"] as $imagedata)
 310                 if($textmode)
 311                     $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]';
 312                 else
 313                     $image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" />';
 314         }
 315
 316         if(strlen($text)) {
 317                 if($textmode)
 318                         $text = $br.'[quote]'.trim($text).'[/quote]'.$br ;
 319                 else
 320                         $text = '<br /><blockquote>'.trim($text).'</blockquote><br />';
 321         }
 322
 323         if($image) {
 324                 $text = $br.$br.$image.$text;
 325         }
 326         $title = str_replace(array("\r","\n"),array('',''),$title);
 327
 328         $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 329
 330         logger('parse_url: returns: ' . $result);
 331
 332         echo trim($result);
 333         killme();
 334 }