]> git.mxchange.org Git - friendica.git/blob - mod/parse_url.php
Merge pull request #927 from annando/master
[friendica.git] / 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, $no_guessing = false) {
54         $siteinfo = array();
55
56         $siteinfo["type"] = "link";
57
58         $ch = curl_init();
59         curl_setopt($ch, CURLOPT_URL, $url);
60         curl_setopt($ch, CURLOPT_HEADER, 1);
61         curl_setopt($ch, CURLOPT_NOBODY, 0);
62         curl_setopt($ch, CURLOPT_TIMEOUT, 3);
63         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
64         //curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
65         curl_setopt($ch,CURLOPT_USERAGENT,' Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:24.0) Gecko/20100101 Firefox/24.0');
66
67         $header = curl_exec($ch);
68         $curl_info = @curl_getinfo($ch);
69         $http_code = $curl_info['http_code'];
70         curl_close($ch);
71
72         if ((($curl_info['http_code'] == "301") OR ($curl_info['http_code'] == "302"))
73                 AND (($curl_info['redirect_url'] != "") OR ($curl_info['location'] != ""))) {
74                 if ($curl_info['redirect_url'] != "")
75                         $siteinfo = parseurl_getsiteinfo($curl_info['redirect_url']);
76                 else
77                         $siteinfo = parseurl_getsiteinfo($curl_info['location']);
78                 return($siteinfo);
79         }
80
81         require_once("include/oembed.php");
82
83         $oembed_data = oembed_fetch_url($url);
84
85         if ($oembed_data->type != "error")
86                 $siteinfo["type"] = $oembed_data->type;
87
88         if ($oembed_data->type == "link") {
89                 if (isset($oembed_data->title))
90                         $siteinfo["title"] = $oembed_data->title;
91                 if (isset($oembed_data->description))
92                         $siteinfo["text"] = trim($oembed_data->description);
93                 if (isset($oembed_data->thumbnail_url))
94                         $siteinfo["image"] = $oembed_data->thumbnail_url;
95         }
96
97         // Fetch the first mentioned charset. Can be in body or header
98         if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
99                 $charset = trim(array_pop($matches));
100         else
101                 $charset = "utf-8";
102
103         $pos = strpos($header, "\r\n\r\n");
104
105         if ($pos)
106                 $body = trim(substr($header, $pos));
107         else
108                 $body = $header;
109
110         $body = mb_convert_encoding($body, "UTF-8", $charset);
111         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
112
113         $doc = new DOMDocument();
114         @$doc->loadHTML($body);
115
116         deletenode($doc, 'style');
117         deletenode($doc, 'script');
118         deletenode($doc, 'option');
119         deletenode($doc, 'h1');
120         deletenode($doc, 'h2');
121         deletenode($doc, 'h3');
122         deletenode($doc, 'h4');
123         deletenode($doc, 'h5');
124         deletenode($doc, 'h6');
125         deletenode($doc, 'ol');
126         deletenode($doc, 'ul');
127
128         $xpath = new DomXPath($doc);
129
130         $list = $xpath->query("//meta[@content]");
131         foreach ($list as $node) {
132                 $attr = array();
133                 if ($node->attributes->length)
134                         foreach ($node->attributes as $attribute)
135                                 $attr[$attribute->name] = $attribute->value;
136
137                 if (@$attr["http-equiv"] == 'refresh') {
138                         $path = $attr["content"];
139                         $pathinfo = explode(";", $path);
140                         $content = "";
141                         foreach ($pathinfo AS $value) {
142                                 if (substr(strtolower($value), 0, 4) == "url=")
143                                         $content = substr($value, 4);
144                         }
145                         if ($content != "") {
146                                 $siteinfo = parseurl_getsiteinfo($content);
147                                 return($siteinfo);
148                         }
149                 }
150         }
151
152         //$list = $xpath->query("head/title");
153         $list = $xpath->query("//title");
154         foreach ($list as $node)
155                 $siteinfo["title"] =  html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
156
157         //$list = $xpath->query("head/meta[@name]");
158         $list = $xpath->query("//meta[@name]");
159         foreach ($list as $node) {
160                 $attr = array();
161                 if ($node->attributes->length)
162                         foreach ($node->attributes as $attribute)
163                                 $attr[$attribute->name] = $attribute->value;
164
165                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
166
167                 switch (strtolower($attr["name"])) {
168                         case "fulltitle":
169                                 $siteinfo["title"] = $attr["content"];
170                                 break;
171                         case "description":
172                                 $siteinfo["text"] = $attr["content"];
173                                 break;
174                         case "twitter:image":
175                                 $siteinfo["image"] = $attr["content"];
176                                 break;
177                         case "twitter:card":
178                                 if ($siteinfo["type"] == "")
179                                         $siteinfo["type"] = $attr["content"];
180                                 break;
181                         case "twitter:description":
182                                 $siteinfo["text"] = $attr["content"];
183                                 break;
184                         case "twitter:title":
185                                 $siteinfo["title"] = $attr["content"];
186                                 break;
187                         case "dc.title":
188                                 $siteinfo["title"] = $attr["content"];
189                                 break;
190                         case "dc.description":
191                                 $siteinfo["text"] = $attr["content"];
192                                 break;
193                 }
194                 if ($siteinfo["type"] == "summary")
195                         $siteinfo["type"] = "link";
196         }
197
198         //$list = $xpath->query("head/meta[@property]");
199         $list = $xpath->query("//meta[@property]");
200         foreach ($list as $node) {
201                 $attr = array();
202                 if ($node->attributes->length)
203                         foreach ($node->attributes as $attribute)
204                                 $attr[$attribute->name] = $attribute->value;
205
206                 $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
207
208                 switch (strtolower($attr["property"])) {
209                         case "og:image":
210                                 $siteinfo["image"] = $attr["content"];
211                                 break;
212                         case "og:title":
213                                 $siteinfo["title"] = $attr["content"];
214                                 break;
215                         case "og:description":
216                                 $siteinfo["text"] = $attr["content"];
217                                 break;
218                 }
219         }
220
221         if ((@$siteinfo["image"] == "") AND !$no_guessing) {
222             $list = $xpath->query("//img[@src]");
223             foreach ($list as $node) {
224                 $attr = array();
225                 if ($node->attributes->length)
226                     foreach ($node->attributes as $attribute)
227                         $attr[$attribute->name] = $attribute->value;
228
229                         $src = completeurl($attr["src"], $url);
230                         $photodata = @getimagesize($src);
231
232                         if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
233                                 if ($photodata[0] > 300) {
234                                         $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
235                                         $photodata[0] = 300;
236                                 }
237                                 if ($photodata[1] > 300) {
238                                         $photodata[0] = round($photodata[0] * (300 / $photodata[1]));
239                                         $photodata[1] = 300;
240                                 }
241                                 $siteinfo["images"][] = array("src"=>$src,
242                                                                 "width"=>$photodata[0],
243                                                                 "height"=>$photodata[1]);
244                         }
245
246                 }
247     } else {
248                 $src = completeurl($siteinfo["image"], $url);
249
250                 unset($siteinfo["image"]);
251
252                 $photodata = @getimagesize($src);
253
254                 if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
255                         $siteinfo["images"][] = array("src"=>$src,
256                                                         "width"=>$photodata[0],
257                                                         "height"=>$photodata[1]);
258         }
259
260         if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$no_guessing) {
261                 $text = "";
262
263                 $list = $xpath->query("//div[@class='article']");
264                 foreach ($list as $node)
265                         if (strlen($node->nodeValue) > 40)
266                                 $text .= " ".trim($node->nodeValue);
267
268                 if ($text == "") {
269                         $list = $xpath->query("//div[@class='content']");
270                         foreach ($list as $node)
271                                 if (strlen($node->nodeValue) > 40)
272                                         $text .= " ".trim($node->nodeValue);
273                 }
274
275                 // If none text was found then take the paragraph content
276                 if ($text == "") {
277                         $list = $xpath->query("//p");
278                         foreach ($list as $node)
279                                 if (strlen($node->nodeValue) > 40)
280                                         $text .= " ".trim($node->nodeValue);
281                 }
282
283                 if ($text != "") {
284                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
285
286                         while (strpos($text, "  "))
287                                 $text = trim(str_replace("  ", " ", $text));
288
289                         $siteinfo["text"] = trim(html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...');
290                 }
291         }
292
293         return($siteinfo);
294 }
295
296 function arr_add_hashes(&$item,$k) {
297         $item = '#' . $item;
298 }
299
300 function parse_url_content(&$a) {
301
302         $text = null;
303         $str_tags = '';
304
305         $textmode = false;
306
307         if(local_user() && (! feature_enabled(local_user(),'richtext')))
308                 $textmode = true;
309
310         //if($textmode)
311         $br = (($textmode) ? "\n" : '<br />');
312
313         if(x($_GET,'binurl'))
314                 $url = trim(hex2bin($_GET['binurl']));
315         else
316                 $url = trim($_GET['url']);
317
318         if($_GET['title'])
319                 $title = strip_tags(trim($_GET['title']));
320
321         if($_GET['description'])
322                 $text = strip_tags(trim($_GET['description']));
323
324         if($_GET['tags']) {
325                 $arr_tags = str_getcsv($_GET['tags']);
326                 if(count($arr_tags)) {
327                         array_walk($arr_tags,'arr_add_hashes');
328                         $str_tags = $br . implode(' ',$arr_tags) . $br;
329                 }
330         }
331
332         logger('parse_url: ' . $url);
333
334         if($textmode)
335                 $template = '[bookmark=%s]%s[/bookmark]%s';
336         else
337                 $template = "<a class=\"bookmark\" href=\"%s\" >%s</a>%s";
338
339         $arr = array('url' => $url, 'text' => '');
340
341         call_hooks('parse_link', $arr);
342
343         if(strlen($arr['text'])) {
344                 echo $arr['text'];
345                 killme();
346         }
347
348
349         if($url && $title && $text) {
350
351                 $title = str_replace(array("\r","\n"),array('',''),$title);
352
353                 if($textmode)
354                         $text = '[quote]' . trim($text) . '[/quote]' . $br;
355                 else {
356                         $text = '<blockquote>' . htmlspecialchars(trim($text)) . '</blockquote><br />';
357                         $title = htmlspecialchars($title);
358                 }
359
360                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
361
362                 logger('parse_url (unparsed): returns: ' . $result);
363
364                 echo $result;
365                 killme();
366         }
367
368         $siteinfo = parseurl_getsiteinfo($url);
369
370         $sitedata = "";
371
372         if($siteinfo["title"] == "") {
373                 $sitedata .= sprintf($template,$url,$url,'') . $str_tags;
374                 killme();
375         } else {
376                 $text = $siteinfo["text"];
377                 $title = $siteinfo["title"];
378         }
379
380         $image = "";
381
382         if(sizeof($siteinfo["images"]) > 0){
383                 /* Execute below code only if image is present in siteinfo */
384
385                 $total_images = 0;
386                 $max_images = get_config('system','max_bookmark_images');
387                 if($max_images === false)
388                         $max_images = 2;
389                 else
390                         $max_images = intval($max_images);
391
392                 foreach ($siteinfo["images"] as $imagedata) {
393                         if($textmode)
394                                 $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]' . "\n";
395                         else
396                                 $image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" /><br />';
397                         $total_images ++;
398                         if($max_images && $max_images >= $total_images)
399                                 break;
400                 }
401         }
402
403         if(strlen($text)) {
404                 if($textmode)
405                         $text = '[quote]'.trim($text).'[/quote]';
406                 else
407                         $text = '<blockquote>'.htmlspecialchars(trim($text)).'</blockquote>';
408         }
409
410         if($image)
411                 $text = $br.$br.$image.$text;
412         else
413                 $text = $br.$text;
414
415         $title = str_replace(array("\r","\n"),array('',''),$title);
416
417         $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
418
419         logger('parse_url: returns: ' . $result);
420
421         $sitedata .=  trim($result);
422
423         if (($siteinfo["type"] == "video") AND ($url != ""))
424                 echo "[video]".$url."[/video]";
425         elseif (($siteinfo["type"] != "photo"))
426                 echo "[class=type-link]".$sitedata."[/class]";
427         else
428                 echo "[class=type-photo]".$title.$br.$image."[/class]";
429
430         killme();
431 }