]> git.mxchange.org Git - friendica.git/blob - include/Scrape.php
ensure host xrd is correct, even if installed in subpath
[friendica.git] / include / Scrape.php
1 <?php
2
3 require_once('library/HTML5/Parser.php');
4
5 if(! function_exists('scrape_dfrn')) {
6 function scrape_dfrn($url) {
7
8         $a = get_app();
9
10         $ret = array();
11
12         logger('scrape_dfrn: url=' . $url);
13
14         $s = fetch_url($url);
15
16         if(! $s) 
17                 return $ret;
18
19         $headers = $a->get_curl_headers();
20         logger('scrape_dfrn: headers=' . $headers, LOGGER_DEBUG);
21
22
23         $lines = explode("\n",$headers);
24         if(count($lines)) {
25                 foreach($lines as $line) {                              
26                         // don't try and run feeds through the html5 parser
27                         if(stristr($line,'content-type:') && ((stristr($line,'application/atom+xml')) || (stristr($line,'application/rss+xml'))))
28                                 return ret;
29                 }
30         }
31
32
33         $dom = HTML5_Parser::parse($s);
34
35         if(! $dom)
36                 return $ret;
37
38         $items = $dom->getElementsByTagName('link');
39
40         // get DFRN link elements
41
42         foreach($items as $item) {
43                 $x = $item->getAttribute('rel');
44                 if(($x === 'alternate') && ($item->getAttribute('type') === 'application/atom+xml'))
45                         $ret['feed_atom'] = $item->getAttribute('href');
46                 if(substr($x,0,5) == "dfrn-")
47                         $ret[$x] = $item->getAttribute('href');
48                 if($x === 'lrdd') {
49                         $decoded = urldecode($item->getAttribute('href'));
50                         if(preg_match('/acct:([^@]*)@/',$decoded,$matches))
51                                 $ret['nick'] = $matches[1];
52                 }
53         }
54
55         // Pull out hCard profile elements
56
57         $items = $dom->getElementsByTagName('*');
58         foreach($items as $item) {
59                 if(attribute_contains($item->getAttribute('class'), 'vcard')) {
60                         $level2 = $item->getElementsByTagName('*');
61                         foreach($level2 as $x) {
62                                 if(attribute_contains($x->getAttribute('class'),'fn'))
63                                         $ret['fn'] = $x->textContent;
64                                 if(attribute_contains($x->getAttribute('class'),'photo'))
65                                         $ret['photo'] = $x->getAttribute('src');
66                                 if(attribute_contains($x->getAttribute('class'),'key'))
67                                         $ret['key'] = $x->textContent;
68                         }
69                 }
70         }
71
72         return $ret;
73 }}
74
75
76
77
78
79
80 if(! function_exists('validate_dfrn')) {
81 function validate_dfrn($a) {
82         $errors = 0;
83         if(! x($a,'key'))
84                 $errors ++;
85         if(! x($a,'dfrn-request'))
86                 $errors ++;
87         if(! x($a,'dfrn-confirm'))
88                 $errors ++;
89         if(! x($a,'dfrn-notify'))
90                 $errors ++;
91         if(! x($a,'dfrn-poll'))
92                 $errors ++;
93         return $errors;
94 }}
95
96 if(! function_exists('scrape_meta')) {
97 function scrape_meta($url) {
98
99         $a = get_app();
100
101         $ret = array();
102
103         logger('scrape_meta: url=' . $url);
104
105         $s = fetch_url($url);
106
107         if(! $s) 
108                 return $ret;
109
110         $headers = $a->get_curl_headers();
111         logger('scrape_meta: headers=' . $headers, LOGGER_DEBUG);
112
113         $lines = explode("\n",$headers);
114         if(count($lines)) {
115                 foreach($lines as $line) {                              
116                         // don't try and run feeds through the html5 parser
117                         if(stristr($line,'content-type:') && ((stristr($line,'application/atom+xml')) || (stristr($line,'application/rss+xml'))))
118                                 return ret;
119                 }
120         }
121
122
123
124         $dom = HTML5_Parser::parse($s);
125
126         if(! $dom)
127                 return $ret;
128
129         $items = $dom->getElementsByTagName('meta');
130
131         // get DFRN link elements
132
133         foreach($items as $item) {
134                 $x = $item->getAttribute('name');
135                 if(substr($x,0,5) == "dfrn-")
136                         $ret[$x] = $item->getAttribute('content');
137         }
138
139         return $ret;
140 }}
141
142
143 if(! function_exists('scrape_vcard')) {
144 function scrape_vcard($url) {
145
146         $a = get_app();
147
148         $ret = array();
149
150         logger('scrape_vcard: url=' . $url);
151
152         $s = fetch_url($url);
153
154         if(! $s) 
155                 return $ret;
156
157         $headers = $a->get_curl_headers();
158         $lines = explode("\n",$headers);
159         if(count($lines)) {
160                 foreach($lines as $line) {                              
161                         // don't try and run feeds through the html5 parser
162                         if(stristr($line,'content-type:') && ((stristr($line,'application/atom+xml')) || (stristr($line,'application/rss+xml'))))
163                                 return ret;
164                 }
165         }
166
167         $dom = HTML5_Parser::parse($s);
168
169         if(! $dom)
170                 return $ret;
171
172         // Pull out hCard profile elements
173
174         $items = $dom->getElementsByTagName('*');
175         foreach($items as $item) {
176                 if(attribute_contains($item->getAttribute('class'), 'vcard')) {
177                         $level2 = $item->getElementsByTagName('*');
178                         foreach($level2 as $x) {
179                                 if(attribute_contains($x->getAttribute('class'),'fn'))
180                                         $ret['fn'] = $x->textContent;
181                                 if((attribute_contains($x->getAttribute('class'),'photo'))
182                                         || (attribute_contains($x->getAttribute('class'),'avatar')))
183                                         $ret['photo'] = $x->getAttribute('src');
184                                 if((attribute_contains($x->getAttribute('class'),'nickname'))
185                                         || (attribute_contains($x->getAttribute('class'),'uid')))
186                                         $ret['nick'] = $x->textContent;
187                         }
188                 }
189         }
190
191         return $ret;
192 }}
193
194
195 if(! function_exists('scrape_feed')) {
196 function scrape_feed($url) {
197
198         $a = get_app();
199
200         $ret = array();
201         $s = fetch_url($url);
202
203         if(! $s) 
204                 return $ret;
205
206         $headers = $a->get_curl_headers();
207         logger('scrape_feed: headers=' . $headers, LOGGER_DEBUG);
208
209         $lines = explode("\n",$headers);
210         if(count($lines)) {
211                 foreach($lines as $line) {                              
212                         if(stristr($line,'content-type:')) {
213                                 if(stristr($line,'application/atom+xml') || stristr($s,'<feed')) {
214                                         $ret['feed_atom'] = $url;
215                                         return $ret;
216                                 }
217                                 if(stristr($line,'application/rss+xml') || stristr($s,'<rss')) {
218                                         $ret['feed_rss'] = $url;
219                                         return $ret;
220                                 }
221                         }
222                 }
223         }
224
225         $dom = HTML5_Parser::parse($s);
226
227         if(! $dom)
228                 return $ret;
229
230
231         $items = $dom->getElementsByTagName('img');
232
233         // get img elements (twitter)
234
235         if($items) {
236                 foreach($items as $item) {
237                         $x = $item->getAttribute('id');
238                         if($x === 'profile-image') {
239                                 $ret['photo'] = $item->getAttribute('src');
240                         }
241                 }
242         }
243
244         $items = $dom->getElementsByTagName('link');
245
246         // get Atom/RSS link elements, take the first one of either.
247
248         if($items) {
249                 foreach($items as $item) {
250                         $x = $item->getAttribute('rel');
251                         if(($x === 'alternate') && ($item->getAttribute('type') === 'application/atom+xml')) {
252                                 if(! x($ret,'feed_atom'))
253                                         $ret['feed_atom'] = $item->getAttribute('href');
254                         }
255                         if(($x === 'alternate') && ($item->getAttribute('type') === 'application/rss+xml')) {
256                                 if(! x($ret,'feed_rss'))
257                                         $ret['feed_rss'] = $item->getAttribute('href');
258                         }
259                 }       
260         }
261
262         return $ret;
263 }}