3 require_once('library/HTML5/Parser.php');
4 require_once('include/crypto.php');
6 if(! function_exists('scrape_dfrn')) {
7 function scrape_dfrn($url) {
13 logger('scrape_dfrn: url=' . $url);
20 $headers = $a->get_curl_headers();
21 logger('scrape_dfrn: headers=' . $headers, LOGGER_DEBUG);
24 $lines = explode("\n",$headers);
26 foreach($lines as $line) {
27 // don't try and run feeds through the html5 parser
28 if(stristr($line,'content-type:') && ((stristr($line,'application/atom+xml')) || (stristr($line,'application/rss+xml'))))
34 $dom = HTML5_Parser::parse($s);
35 } catch (DOMException $e) {
36 logger('scrape_dfrn: parse error: ' . $e);
42 $items = $dom->getElementsByTagName('link');
44 // get DFRN link elements
46 foreach($items as $item) {
47 $x = $item->getAttribute('rel');
48 if(($x === 'alternate') && ($item->getAttribute('type') === 'application/atom+xml'))
49 $ret['feed_atom'] = $item->getAttribute('href');
50 if(substr($x,0,5) == "dfrn-") {
51 $ret[$x] = $item->getAttribute('href');
54 $decoded = urldecode($item->getAttribute('href'));
55 if(preg_match('/acct:([^@]*)@/',$decoded,$matches))
56 $ret['nick'] = $matches[1];
60 // Pull out hCard profile elements
64 $items = $dom->getElementsByTagName('*');
65 foreach($items as $item) {
66 if(attribute_contains($item->getAttribute('class'), 'vcard')) {
67 $level2 = $item->getElementsByTagName('*');
68 foreach($level2 as $x) {
69 if(attribute_contains($x->getAttribute('class'),'fn')) {
70 $ret['fn'] = $x->textContent;
72 if((attribute_contains($x->getAttribute('class'),'photo'))
73 || (attribute_contains($x->getAttribute('class'),'avatar'))) {
74 $size = intval($x->getAttribute('width'));
75 // dfrn prefers 175, so if we find this, we set largest_size so it can't be topped.
76 if(($size > $largest_photo) || ($size == 175) || (! $largest_photo)) {
77 $ret['photo'] = $x->getAttribute('src');
78 $largest_photo = (($size == 175) ? 9999 : $size);
81 if(attribute_contains($x->getAttribute('class'),'key')) {
82 $ret['key'] = $x->textContent;
96 if(! function_exists('validate_dfrn')) {
97 function validate_dfrn($a) {
101 if(! x($a,'dfrn-request'))
103 if(! x($a,'dfrn-confirm'))
105 if(! x($a,'dfrn-notify'))
107 if(! x($a,'dfrn-poll'))
112 if(! function_exists('scrape_meta')) {
113 function scrape_meta($url) {
119 logger('scrape_meta: url=' . $url);
121 $s = fetch_url($url);
126 $headers = $a->get_curl_headers();
127 logger('scrape_meta: headers=' . $headers, LOGGER_DEBUG);
129 $lines = explode("\n",$headers);
131 foreach($lines as $line) {
132 // don't try and run feeds through the html5 parser
133 if(stristr($line,'content-type:') && ((stristr($line,'application/atom+xml')) || (stristr($line,'application/rss+xml'))))
139 $dom = HTML5_Parser::parse($s);
140 } catch (DOMException $e) {
141 logger('scrape_meta: parse error: ' . $e);
147 $items = $dom->getElementsByTagName('meta');
149 // get DFRN link elements
151 foreach($items as $item) {
152 $x = $item->getAttribute('name');
153 if(substr($x,0,5) == "dfrn-")
154 $ret[$x] = $item->getAttribute('content');
161 if(! function_exists('scrape_vcard')) {
162 function scrape_vcard($url) {
168 logger('scrape_vcard: url=' . $url);
170 $s = fetch_url($url);
175 $headers = $a->get_curl_headers();
176 $lines = explode("\n",$headers);
178 foreach($lines as $line) {
179 // don't try and run feeds through the html5 parser
180 if(stristr($line,'content-type:') && ((stristr($line,'application/atom+xml')) || (stristr($line,'application/rss+xml'))))
186 $dom = HTML5_Parser::parse($s);
187 } catch (DOMException $e) {
188 logger('scrape_vcard: parse error: ' . $e);
194 // Pull out hCard profile elements
198 $items = $dom->getElementsByTagName('*');
199 foreach($items as $item) {
200 if(attribute_contains($item->getAttribute('class'), 'vcard')) {
201 $level2 = $item->getElementsByTagName('*');
202 foreach($level2 as $x) {
203 if(attribute_contains($x->getAttribute('class'),'fn'))
204 $ret['fn'] = $x->textContent;
205 if((attribute_contains($x->getAttribute('class'),'photo'))
206 || (attribute_contains($x->getAttribute('class'),'avatar'))) {
207 $size = intval($x->getAttribute('width'));
208 if(($size > $largest_photo) || (! $largest_photo)) {
209 $ret['photo'] = $x->getAttribute('src');
210 $largest_photo = $size;
213 if((attribute_contains($x->getAttribute('class'),'nickname'))
214 || (attribute_contains($x->getAttribute('class'),'uid'))) {
215 $ret['nick'] = $x->textContent;
225 if(! function_exists('scrape_feed')) {
226 function scrape_feed($url) {
231 $s = fetch_url($url);
233 $headers = $a->get_curl_headers();
234 $code = $a->get_curl_code();
236 logger('scrape_feed: returns: ' . $code . ' headers=' . $headers, LOGGER_DEBUG);
239 logger('scrape_feed: no data returned for ' . $url);
244 $lines = explode("\n",$headers);
246 foreach($lines as $line) {
247 if(stristr($line,'content-type:')) {
248 if(stristr($line,'application/atom+xml') || stristr($s,'<feed')) {
249 $ret['feed_atom'] = $url;
252 if(stristr($line,'application/rss+xml') || stristr($s,'<rss')) {
253 $ret['feed_rss'] = $url;
258 // perhaps an RSS version 1 feed with a generic or incorrect content-type?
259 if(stristr($s,'</item>')) {
260 $ret['feed_rss'] = $url;
266 $dom = HTML5_Parser::parse($s);
267 } catch (DOMException $e) {
268 logger('scrape_feed: parse error: ' . $e);
272 logger('scrape_feed: failed to parse.');
277 $head = $dom->getElementsByTagName('base');
279 foreach($head as $head0) {
280 $basename = $head0->getAttribute('href');
285 $basename = implode('/', array_slice(explode('/',$url),0,3)) . '/';
287 $items = $dom->getElementsByTagName('link');
289 // get Atom/RSS link elements, take the first one of either.
292 foreach($items as $item) {
293 $x = $item->getAttribute('rel');
294 if(($x === 'alternate') && ($item->getAttribute('type') === 'application/atom+xml')) {
295 if(! x($ret,'feed_atom'))
296 $ret['feed_atom'] = $item->getAttribute('href');
298 if(($x === 'alternate') && ($item->getAttribute('type') === 'application/rss+xml')) {
299 if(! x($ret,'feed_rss'))
300 $ret['feed_rss'] = $item->getAttribute('href');
305 // Drupal and perhaps others only provide relative URL's. Turn them into absolute.
307 if(x($ret,'feed_atom') && (! strstr($ret['feed_atom'],'://')))
308 $ret['feed_atom'] = $basename . $ret['feed_atom'];
309 if(x($ret,'feed_rss') && (! strstr($ret['feed_rss'],'://')))
310 $ret['feed_rss'] = $basename . $ret['feed_rss'];
318 * Probe a network address to discover what kind of protocols we need to communicate with it.
320 * Warning: this function is a bit touchy and there are some subtle dependencies within the logic flow.
327 * PROBE_DIASPORA has a bias towards returning Diaspora information
328 * while PROBE_NORMAL has a bias towards dfrn/zot - in the case where
329 * an address (such as a Friendica address) supports more than one type
335 define ( 'PROBE_NORMAL', 0);
336 define ( 'PROBE_DIASPORA', 1);
338 function probe_url($url, $mode = PROBE_NORMAL) {
339 require_once('include/email.php');
352 $email_conversant = false;
354 $twitter = ((strpos($url,'twitter.com') !== false) ? true : false);
355 $lastfm = ((strpos($url,'last.fm/user') !== false) ? true : false);
357 $at_addr = ((strpos($url,'@') !== false) ? true : false);
359 if((! $twitter) && (! $lastfm)) {
361 if(strpos($url,'mailto:') !== false && $at_addr) {
362 $url = str_replace('mailto:','',$url);
371 logger('probe_url: found lrdd links: ' . print_r($links,true), LOGGER_DATA);
372 foreach($links as $link) {
373 if($link['@attributes']['rel'] === NAMESPACE_ZOT)
374 $zot = unamp($link['@attributes']['href']);
375 if($link['@attributes']['rel'] === NAMESPACE_DFRN)
376 $dfrn = unamp($link['@attributes']['href']);
377 if($link['@attributes']['rel'] === 'salmon')
378 $notify = unamp($link['@attributes']['href']);
379 if($link['@attributes']['rel'] === NAMESPACE_FEED)
380 $poll = unamp($link['@attributes']['href']);
381 if($link['@attributes']['rel'] === 'http://microformats.org/profile/hcard')
382 $hcard = unamp($link['@attributes']['href']);
383 if($link['@attributes']['rel'] === 'http://webfinger.net/rel/profile-page')
384 $profile = unamp($link['@attributes']['href']);
385 if($link['@attributes']['rel'] === 'http://portablecontacts.net/spec/1.0')
386 $poco = unamp($link['@attributes']['href']);
387 if($link['@attributes']['rel'] === 'http://joindiaspora.com/seed_location') {
388 $diaspora_base = unamp($link['@attributes']['href']);
391 if($link['@attributes']['rel'] === 'http://joindiaspora.com/guid') {
392 $diaspora_guid = unamp($link['@attributes']['href']);
395 if($link['@attributes']['rel'] === 'diaspora-public-key') {
396 $diaspora_key = base64_decode(unamp($link['@attributes']['href']));
397 if(strstr($diaspora_key,'RSA '))
398 $pubkey = rsatopem($diaspora_key);
400 $pubkey = $diaspora_key;
405 // Status.Net can have more than one profile URL. We need to match the profile URL
406 // to a contact on incoming messages to prevent spam, and we won't know which one
407 // to match. So in case of two, one of them is stored as an alias. Only store URL's
408 // and not webfinger user@host aliases. If they've got more than two non-email style
409 // aliases, let's hope we're lucky and get one that matches the feed author-uri because
410 // otherwise we're screwed.
412 foreach($links as $link) {
413 if($link['@attributes']['rel'] === 'alias') {
414 if(strpos($link['@attributes']['href'],'@') === false) {
415 if(isset($profile)) {
416 if($link['@attributes']['href'] !== $profile)
417 $alias = unamp($link['@attributes']['href']);
420 $profile = unamp($link['@attributes']['href']);
425 elseif($mode == PROBE_NORMAL) {
430 if((strpos($orig_url,'@')) && validate_email($orig_url)) {
431 $x = q("SELECT `prvkey` FROM `user` WHERE `uid` = %d LIMIT 1",
434 $r = q("SELECT * FROM `mailacct` WHERE `uid` = %d AND `server` != '' LIMIT 1",
437 if(count($x) && count($r)) {
438 $mailbox = construct_mailbox_name($r[0]);
440 openssl_private_decrypt(hex2bin($r[0]['pass']),$password,$x[0]['prvkey']);
441 $mbox = email_connect($mailbox,$r[0]['user'],$password);
443 logger('probe_url: email_connect failed.');
447 $msgs = email_poll($mbox,$orig_url);
448 logger('probe_url: searching ' . $orig_url . ', ' . count($msgs) . ' messages found.', LOGGER_DEBUG);
451 $network = NETWORK_MAIL;
452 $name = substr($url,0,strpos($url,'@'));
453 $phost = substr($url,strpos($url,'@')+1);
454 $profile = 'http://' . $phost;
455 // fix nick character range
456 $vcard = array('fn' => $name, 'nick' => $name, 'photo' => avatar_img($url));
457 $notify = 'smtp ' . random_string();
458 $poll = 'email ' . random_string();
460 $x = email_msg_meta($mbox,$msgs[0]);
461 if(stristr($x[0]->from,$orig_url))
462 $adr = imap_rfc822_parse_adrlist($x[0]->from,'');
463 elseif(stristr($x[0]->to,$orig_url))
464 $adr = imap_rfc822_parse_adrlist($x[0]->to,'');
466 foreach($adr as $feadr) {
467 if((strcasecmp($feadr->mailbox,$name) == 0)
468 &&(strcasecmp($feadr->host,$phost) == 0)
469 && (strlen($feadr->personal))) {
471 $personal = imap_mime_header_decode($feadr->personal);
473 foreach($personal as $perspart)
474 if ($perspart->charset != "default")
475 $vcard['fn'] .= iconv($perspart->charset, 'UTF-8//IGNORE', $perspart->text);
477 $vcard['fn'] .= $perspart->text;
479 $vcard['fn'] = notags($vcard['fn']);
490 if($mode == PROBE_NORMAL) {
492 $s = fetch_url($zot);
494 $j = json_decode($s);
496 $network = NETWORK_ZOT;
498 'fn' => $j->fullname,
499 'nick' => $j->nickname,
504 $pubkey = $j->pubkey;
511 $ret = scrape_dfrn(($hcard) ? $hcard : $dfrn);
512 if(is_array($ret) && x($ret,'dfrn-request')) {
513 $network = NETWORK_DFRN;
514 $request = $ret['dfrn-request'];
515 $confirm = $ret['dfrn-confirm'];
516 $notify = $ret['dfrn-notify'];
517 $poll = $ret['dfrn-poll'];
520 $vcard['fn'] = $ret['fn'];
521 $vcard['nick'] = $ret['nick'];
522 $vcard['photo'] = $ret['photo'];
527 if($diaspora && $diaspora_base && $diaspora_guid) {
528 if($mode == PROBE_DIASPORA || ! $notify) {
529 $notify = $diaspora_base . 'receive/users/' . $diaspora_guid;
530 $batch = $diaspora_base . 'receive/public' ;
533 $addr = str_replace('acct:', '', $url);
536 if($network !== NETWORK_ZOT && $network !== NETWORK_DFRN && $network !== NETWORK_MAIL) {
538 $network = NETWORK_DIASPORA;
540 $network = NETWORK_OSTATUS;
543 if($hcard && ! $vcard) {
544 $vcard = scrape_vcard($hcard);
546 // Google doesn't use absolute url in profile photos
548 if((x($vcard,'photo')) && substr($vcard['photo'],0,1) == '/') {
549 $h = @parse_url($hcard);
551 $vcard['photo'] = $h['scheme'] . '://' . $h['host'] . $vcard['photo'];
554 logger('probe_url: scrape_vcard: ' . print_r($vcard,true), LOGGER_DATA);
557 if($diaspora && $addr) {
558 // Diaspora returns the name as the nick. As the nick will never be updated,
559 // let's use the Diaspora nickname (the first part of the handle) as the nick instead
560 $addr_parts = explode('@', $addr);
561 $vcard['nick'] = $addr_parts[0];
565 logger('twitter: setup');
566 $tid = basename($url);
567 $tapi = 'https://api.twitter.com/1/statuses/user_timeline.rss';
569 $poll = $tapi . '?user_id=' . $tid;
571 $poll = $tapi . '?screen_name=' . $tid;
572 $profile = 'http://twitter.com/#!/' . $tid;
573 //$vcard['photo'] = 'https://api.twitter.com/1/users/profile_image/' . $tid;
574 $vcard['photo'] = 'https://api.twitter.com/1/users/profile_image?screen_name=' . $tid . '&size=bigger';
575 $vcard['nick'] = $tid;
581 $poll = str_replace(array('www.','last.fm/'),array('','ws.audioscrobbler.com/1.0/'),$url) . '/recenttracks.rss';
582 $vcard['nick'] = basename($url);
583 $vcard['fn'] = $vcard['nick'] . t(' on Last.fm');
584 $network = NETWORK_FEED;
589 $vcard['fn'] = $vcard['nick'];
593 if(stristr($url,'tumblr.com') && (! stristr($url,'/rss'))) {
594 $poll = $url . '/rss';
596 // Will leave it to others to figure out how to grab the avatar, which is on the $url page in the open graph meta links
599 if($twitter || ! $poll)
601 if((! isset($vcard)) || (! x($vcard,'fn')) || (! $profile))
603 if(($at_addr) && (! count($links)))
608 $feedret = scrape_feed(($poll) ? $poll : $url);
609 logger('probe_url: scrape_feed ' . (($poll)? $poll : $url) . ' returns: ' . print_r($feedret,true), LOGGER_DATA);
610 if(count($feedret) && ($feedret['feed_atom'] || $feedret['feed_rss'])) {
611 $poll = ((x($feedret,'feed_atom')) ? unamp($feedret['feed_atom']) : unamp($feedret['feed_rss']));
616 if(x($feedret,'photo') && (! x($vcard,'photo')))
617 $vcard['photo'] = $feedret['photo'];
618 require_once('library/simplepie/simplepie.inc');
619 $feed = new SimplePie();
620 $xml = fetch_url($poll);
622 logger('probe_url: fetch feed: ' . $poll . ' returns: ' . $xml, LOGGER_DATA);
625 logger('probe_url: scrape_feed: headers: ' . $a->get_curl_headers(), LOGGER_DATA);
627 // Don't try and parse an empty string
628 $feed->set_raw_data(($xml) ? $xml : '<?xml version="1.0" encoding="utf-8" ?><xml></xml>');
632 logger('probe_url: scrape_feed: Error parsing XML: ' . $feed->error());
635 if(! x($vcard,'photo'))
636 $vcard['photo'] = $feed->get_image_url();
637 $author = $feed->get_author();
640 $vcard['fn'] = unxmlify(trim($author->get_name()));
642 $vcard['fn'] = trim(unxmlify($author->get_email()));
643 if(strpos($vcard['fn'],'@') !== false)
644 $vcard['fn'] = substr($vcard['fn'],0,strpos($vcard['fn'],'@'));
645 $email = unxmlify($author->get_email());
646 if(! $profile && $author->get_link())
647 $profile = trim(unxmlify($author->get_link()));
648 if(! $vcard['photo']) {
649 $rawtags = $feed->get_feed_tags( SIMPLEPIE_NAMESPACE_ATOM_10, 'author');
651 $elems = $rawtags[0]['child'][SIMPLEPIE_NAMESPACE_ATOM_10];
652 if((x($elems,'link')) && ($elems['link'][0]['attribs']['']['rel'] === 'photo'))
653 $vcard['photo'] = $elems['link'][0]['attribs']['']['href'];
658 $item = $feed->get_item(0);
660 $author = $item->get_author();
662 $vcard['fn'] = trim(unxmlify($author->get_name()));
664 $vcard['fn'] = trim(unxmlify($author->get_email()));
665 if(strpos($vcard['fn'],'@') !== false)
666 $vcard['fn'] = substr($vcard['fn'],0,strpos($vcard['fn'],'@'));
667 $email = unxmlify($author->get_email());
668 if(! $profile && $author->get_link())
669 $profile = trim(unxmlify($author->get_link()));
671 if(! $vcard['photo']) {
672 $rawmedia = $item->get_item_tags('http://search.yahoo.com/mrss/','thumbnail');
673 if($rawmedia && $rawmedia[0]['attribs']['']['url'])
674 $vcard['photo'] = unxmlify($rawmedia[0]['attribs']['']['url']);
676 if(! $vcard['photo']) {
677 $rawtags = $item->get_item_tags( SIMPLEPIE_NAMESPACE_ATOM_10, 'author');
679 $elems = $rawtags[0]['child'][SIMPLEPIE_NAMESPACE_ATOM_10];
680 if((x($elems,'link')) && ($elems['link'][0]['attribs']['']['rel'] === 'photo'))
681 $vcard['photo'] = $elems['link'][0]['attribs']['']['href'];
687 if((! $vcard['photo']) && strlen($email))
688 $vcard['photo'] = avatar_img($email);
689 if($poll === $profile)
690 $lnk = $feed->get_permalink();
691 if(isset($lnk) && strlen($lnk))
694 if(! (x($vcard,'fn')))
695 $vcard['fn'] = notags($feed->get_title());
696 if(! (x($vcard,'fn')))
697 $vcard['fn'] = notags($feed->get_description());
699 if(strpos($vcard['fn'],'Twitter / ') !== false) {
700 $vcard['fn'] = substr($vcard['fn'],strpos($vcard['fn'],'/')+1);
701 $vcard['fn'] = trim($vcard['fn']);
703 if(! x($vcard,'nick')) {
704 $vcard['nick'] = strtolower(notags(unxmlify($vcard['fn'])));
705 if(strpos($vcard['nick'],' '))
706 $vcard['nick'] = trim(substr($vcard['nick'],0,strpos($vcard['nick'],' ')));
709 $network = NETWORK_FEED;
715 if(! x($vcard,'photo')) {
717 $vcard['photo'] = $a->get_baseurl() . '/images/person-175.jpg' ;
723 // No human could be associated with this link, use the URL as the contact name
725 if(($network === NETWORK_FEED) && ($poll) && (! x($vcard,'fn')))
728 $vcard['fn'] = notags($vcard['fn']);
729 $vcard['nick'] = str_replace(' ','',notags($vcard['nick']));
731 $result['name'] = $vcard['fn'];
732 $result['nick'] = $vcard['nick'];
733 $result['url'] = $profile;
734 $result['addr'] = $addr;
735 $result['batch'] = $batch;
736 $result['notify'] = $notify;
737 $result['poll'] = $poll;
738 $result['request'] = $request;
739 $result['confirm'] = $confirm;
740 $result['poco'] = $poco;
741 $result['photo'] = $vcard['photo'];
742 $result['priority'] = $priority;
743 $result['network'] = $network;
744 $result['alias'] = $alias;
745 $result['pubkey'] = $pubkey;
747 logger('probe_url: ' . print_r($result,true), LOGGER_DEBUG);