$ret['feed_atom'] = $url;
return $ret;
}
- if(stristr($line,'application/rss+xml') || stristr($s,'<rss')) {
+ if(stristr($line,'application/rss+xml') || stristr($s,'<rss')) {
$ret['feed_rss'] = $url;
return $ret;
}
}
}
- try {
- $dom = HTML5_Parser::parse($s);
- } catch (DOMException $e) {
- logger('scrape_feed: parse error: ' . $e);
- }
+ $basename = implode('/', array_slice(explode('/',$url),0,3)) . '/';
- if(! $dom) {
- logger('scrape_feed: failed to parse.');
- return $ret;
- }
+ $doc = new DOMDocument();
+ @$doc->loadHTML($s);
+ $xpath = new DomXPath($doc);
+ $base = $xpath->query("//base");
+ foreach ($base as $node) {
+ $attr = array();
- $head = $dom->getElementsByTagName('base');
- if($head) {
- foreach($head as $head0) {
- $basename = $head0->getAttribute('href');
- break;
- }
+ if ($node->attributes->length)
+ foreach ($node->attributes as $attribute)
+ $attr[$attribute->name] = $attribute->value;
+
+ if ($attr["href"] != "")
+ $basename = $attr["href"] ;
}
- if(! $basename)
- $basename = implode('/', array_slice(explode('/',$url),0,3)) . '/';
- $items = $dom->getElementsByTagName('link');
+ $list = $xpath->query("//link");
+ foreach ($list as $node) {
+ $attr = array();
- // get Atom/RSS link elements, take the first one of either.
+ if ($node->attributes->length)
+ foreach ($node->attributes as $attribute)
+ $attr[$attribute->name] = $attribute->value;
- if($items) {
- foreach($items as $item) {
- $x = $item->getAttribute('rel');
- if(($x === 'alternate') && ($item->getAttribute('type') === 'application/atom+xml')) {
- if(! x($ret,'feed_atom'))
- $ret['feed_atom'] = $item->getAttribute('href');
- }
- if(($x === 'alternate') && ($item->getAttribute('type') === 'application/rss+xml')) {
- if(! x($ret,'feed_rss'))
- $ret['feed_rss'] = $item->getAttribute('href');
- }
- }
+ if (($attr["rel"] == "alternate") AND ($attr["type"] == "application/atom+xml"))
+ $ret["feed_atom"] = $attr["href"];
+
+ if (($attr["rel"] == "alternate") AND ($attr["type"] == "application/rss+xml"))
+ $ret["feed_rss"] = $attr["href"];
}
- // Drupal and perhaps others only provide relative URL's. Turn them into absolute.
+ // Drupal and perhaps others only provide relative URLs. Turn them into absolute.
if(x($ret,'feed_atom') && (! strstr($ret['feed_atom'],'://')))
$ret['feed_atom'] = $basename . $ret['feed_atom'];
define ( 'PROBE_NORMAL', 0);
define ( 'PROBE_DIASPORA', 1);
-function probe_url($url, $mode = PROBE_NORMAL) {
+function probe_url($url, $mode = PROBE_NORMAL, $level = 1) {
require_once('include/email.php');
$result = array();
if(! $url)
return $result;
+ $result = Cache::get("probe_url:".$mode.":".$url);
+ if (!is_null($result)) {
+ $result = unserialize($result);
+ return $result;
+ }
+
$network = null;
$diaspora = false;
$diaspora_base = '';
$diaspora_key = '';
$has_lrdd = false;
$email_conversant = false;
+ $connectornetworks = false;
+ $appnet = false;
+
+ if (strpos($url,'twitter.com')) {
+ $connectornetworks = true;
+ $network = NETWORK_TWITTER;
+ }
// Twitter is deactivated since twitter closed its old API
//$twitter = ((strpos($url,'twitter.com') !== false) ? true : false);
$at_addr = ((strpos($url,'@') !== false) ? true : false);
- if((! $twitter) && (! $lastfm)) {
+ if((!$appnet) && (!$lastfm) && !$connectornetworks) {
if(strpos($url,'mailto:') !== false && $at_addr) {
$url = str_replace('mailto:','',$url);
$pubkey = $diaspora_key;
$diaspora = true;
}
+ if($link['@attributes']['rel'] === 'http://ostatus.org/schema/1.0/subscribe') {
+ $diaspora = false;
+ }
}
// Status.Net can have more than one profile URL. We need to match the profile URL
// to a contact on incoming messages to prevent spam, and we won't know which one
// to match. So in case of two, one of them is stored as an alias. Only store URL's
// and not webfinger user@host aliases. If they've got more than two non-email style
- // aliases, let's hope we're lucky and get one that matches the feed author-uri because
+ // aliases, let's hope we're lucky and get one that matches the feed author-uri because
// otherwise we're screwed.
foreach($links as $link) {
}
}
}
+
+ // If the profile is different from the url then the url is abviously an alias
+ if (($alias == "") AND ($profile != "") AND !$at_addr AND (normalise_link($profile) != normalise_link($url)))
+ $alias = $url;
}
elseif($mode == PROBE_NORMAL) {
}
if($mode == PROBE_NORMAL) {
+
if(strlen($zot)) {
$s = fetch_url($zot);
if($s) {
if($j) {
$network = NETWORK_ZOT;
$vcard = array(
- 'fn' => $j->fullname,
- 'nick' => $j->nickname,
+ 'fn' => $j->fullname,
+ 'nick' => $j->nickname,
'photo' => $j->photo
);
$profile = $j->url;
}
}
+
if(strlen($dfrn)) {
$ret = scrape_dfrn(($hcard) ? $hcard : $dfrn);
if(is_array($ret) && x($ret,'dfrn-request')) {
if($network !== NETWORK_ZOT && $network !== NETWORK_DFRN && $network !== NETWORK_MAIL) {
if($diaspora)
$network = NETWORK_DIASPORA;
- elseif($has_lrdd)
+ elseif($has_lrdd AND ($notify))
$network = NETWORK_OSTATUS;
+
+ if(strpos($url,'@'))
+ $addr = str_replace('acct:', '', $url);
+
$priority = 0;
if($hcard && ! $vcard) {
// Will leave it to others to figure out how to grab the avatar, which is on the $url page in the open graph meta links
}
- if($twitter || ! $poll)
+ if($appnet || ! $poll)
$check_feed = true;
if((! isset($vcard)) || (! x($vcard,'fn')) || (! $profile))
$check_feed = true;
if(($at_addr) && (! count($links)))
$check_feed = false;
+ if ($connectornetworks)
+ $check_feed = false;
+
if($check_feed) {
$feedret = scrape_feed(($poll) ? $poll : $url);
+
logger('probe_url: scrape_feed ' . (($poll)? $poll : $url) . ' returns: ' . print_r($feedret,true), LOGGER_DATA);
if(count($feedret) && ($feedret['feed_atom'] || $feedret['feed_rss'])) {
$poll = ((x($feedret,'feed_atom')) ? unamp($feedret['feed_atom']) : unamp($feedret['feed_rss']));
logger('probe_url: scrape_feed: headers: ' . $a->get_curl_headers(), LOGGER_DATA);
// Don't try and parse an empty string
- $feed->set_raw_data(($xml) ? $xml : '<?xml version="1.0" encoding="utf-8" ?><xml></xml>');
+ $feed->set_raw_data(($xml) ? $xml : '<?xml version="1.0" encoding="utf-8" ?><xml></xml>');
$feed->init();
- if($feed->error())
+ if($feed->error()) {
logger('probe_url: scrape_feed: Error parsing XML: ' . $feed->error());
-
+ $network = NETWORK_PHANTOM;
+ }
if(! x($vcard,'photo'))
$vcard['photo'] = $feed->get_image_url();
$vcard['fn'] = trim(unxmlify($author->get_email()));
if(strpos($vcard['fn'],'@') !== false)
$vcard['fn'] = substr($vcard['fn'],0,strpos($vcard['fn'],'@'));
+
$email = unxmlify($author->get_email());
if(! $profile && $author->get_link())
$profile = trim(unxmlify($author->get_link()));
$vcard['photo'] = $elems['link'][0]['attribs']['']['href'];
}
}
+ // Fetch fullname via poco:displayName
+ $pocotags = $feed->get_feed_tags(SIMPLEPIE_NAMESPACE_ATOM_10, 'author');
+ if ($pocotags) {
+ $elems = $pocotags[0]['child']['http://portablecontacts.net/spec/1.0'];
+ if (isset($elems["displayName"]))
+ $vcard['fn'] = $elems["displayName"][0]["data"];
+ if (isset($elems["preferredUsername"]))
+ $vcard['nick'] = $elems["preferredUsername"][0]["data"];
+ }
}
else {
$item = $feed->get_item(0);
if(($network === NETWORK_FEED) && ($poll) && (! x($vcard,'fn')))
$vcard['fn'] = $url;
+ if (($notify != "") AND ($poll != "")) {
+ $baseurl = matching(normalise_link($notify), normalise_link($poll));
+
+ $baseurl2 = matching($baseurl, normalise_link($profile));
+ if ($baseurl2 != "")
+ $baseurl = $baseurl2;
+ }
+
+ if (($baseurl == "") AND ($notify != ""))
+ $baseurl = matching(normalise_link($profile), normalise_link($notify));
+
+ if (($baseurl == "") AND ($poll != ""))
+ $baseurl = matching(normalise_link($profile), normalise_link($poll));
+
+ $baseurl = rtrim($baseurl, "/");
+
$vcard['fn'] = notags($vcard['fn']);
$vcard['nick'] = str_replace(' ','',notags($vcard['nick']));
$result['network'] = $network;
$result['alias'] = $alias;
$result['pubkey'] = $pubkey;
+ $result['baseurl'] = $baseurl;
logger('probe_url: ' . print_r($result,true), LOGGER_DEBUG);
- // Trying if it maybe a diaspora account
- if ($result['network'] == NETWORK_FEED) {
- require_once('include/bbcode.php');
- $address = GetProfileUsername($url, "", true);
- $result2 = probe_url($address, $mode);
- if ($result2['network'] != "")
- $result = $result2;
+ if ($level == 1) {
+ // Trying if it maybe a diaspora account
+ if (($result['network'] == NETWORK_FEED) OR ($result['addr'] == "")) {
+ require_once('include/bbcode.php');
+ $address = GetProfileUsername($url, "", true);
+ $result2 = probe_url($address, $mode, ++$level);
+ if ($result2['network'] != "")
+ $result = $result2;
+ }
+
+ // Maybe it's some non standard GNU Social installation (Single user, subfolder or no uri rewrite)
+ if (($result['network'] == NETWORK_FEED) AND ($result['baseurl'] != "") AND ($result['nick'] != "")) {
+ $addr = $result['nick'].'@'.str_replace("http://", "", $result['baseurl']);
+ $result2 = probe_url($addr, $mode, ++$level);
+ if (($result2['network'] != "") AND ($result2['network'] != NETWORK_FEED))
+ $result = $result2;
+ }
}
+ // Only store into the cache if the value seems to be valid
+ if ($result['network'] != NETWORK_FEED)
+ Cache::set("probe_url:".$mode.":".$url,serialize($result), CACHE_DAY);
+
return $result;
}
+
+function matching($part1, $part2) {
+ $len = min(strlen($part1), strlen($part2));
+
+ $match = "";
+ $matching = true;
+ $i = 0;
+ while (($i <= $len) AND $matching) {
+ if (substr($part1, $i, 1) == substr($part2, $i, 1))
+ $match .= substr($part1, $i, 1);
+ else
+ $matching = false;
+
+ $i++;
+ }
+ return($match);
+}