<?php
require_once('library/HTML5/Parser.php');
+require_once('include/crypto.php');
if(! function_exists('scrape_dfrn')) {
function scrape_dfrn($url) {
}
}
-
- $dom = HTML5_Parser::parse($s);
+ try {
+ $dom = HTML5_Parser::parse($s);
+ } catch (DOMException $e) {
+ logger('scrape_dfrn: parse error: ' . $e);
+ }
if(! $dom)
return $ret;
$x = $item->getAttribute('rel');
if(($x === 'alternate') && ($item->getAttribute('type') === 'application/atom+xml'))
$ret['feed_atom'] = $item->getAttribute('href');
- if(substr($x,0,5) == "dfrn-")
+ if(substr($x,0,5) == "dfrn-") {
$ret[$x] = $item->getAttribute('href');
+ }
if($x === 'lrdd') {
$decoded = urldecode($item->getAttribute('href'));
if(preg_match('/acct:([^@]*)@/',$decoded,$matches))
// Pull out hCard profile elements
+ $largest_photo = 0;
+
$items = $dom->getElementsByTagName('*');
foreach($items as $item) {
if(attribute_contains($item->getAttribute('class'), 'vcard')) {
$level2 = $item->getElementsByTagName('*');
foreach($level2 as $x) {
- if(attribute_contains($x->getAttribute('class'),'fn'))
+ if(attribute_contains($x->getAttribute('class'),'fn')) {
$ret['fn'] = $x->textContent;
- if(attribute_contains($x->getAttribute('class'),'photo'))
- $ret['photo'] = $x->getAttribute('src');
- if(attribute_contains($x->getAttribute('class'),'key'))
+ }
+ if((attribute_contains($x->getAttribute('class'),'photo'))
+ || (attribute_contains($x->getAttribute('class'),'avatar'))) {
+ $size = intval($x->getAttribute('width'));
+ // dfrn prefers 175, so if we find this, we set largest_size so it can't be topped.
+ if(($size > $largest_photo) || ($size == 175) || (! $largest_photo)) {
+ $ret['photo'] = $x->getAttribute('src');
+ $largest_photo = (($size == 175) ? 9999 : $size);
+ }
+ }
+ if(attribute_contains($x->getAttribute('class'),'key')) {
$ret['key'] = $x->textContent;
+ }
}
}
}
}
}
-
-
- $dom = HTML5_Parser::parse($s);
+ try {
+ $dom = HTML5_Parser::parse($s);
+ } catch (DOMException $e) {
+ logger('scrape_meta: parse error: ' . $e);
+ }
if(! $dom)
return $ret;
}
}
- $dom = HTML5_Parser::parse($s);
+ try {
+ $dom = HTML5_Parser::parse($s);
+ } catch (DOMException $e) {
+ logger('scrape_vcard: parse error: ' . $e);
+ }
if(! $dom)
return $ret;
// Pull out hCard profile elements
+ $largest_photo = 0;
+
$items = $dom->getElementsByTagName('*');
foreach($items as $item) {
if(attribute_contains($item->getAttribute('class'), 'vcard')) {
if(attribute_contains($x->getAttribute('class'),'fn'))
$ret['fn'] = $x->textContent;
if((attribute_contains($x->getAttribute('class'),'photo'))
- || (attribute_contains($x->getAttribute('class'),'avatar')))
- $ret['photo'] = $x->getAttribute('src');
+ || (attribute_contains($x->getAttribute('class'),'avatar'))) {
+ $size = intval($x->getAttribute('width'));
+ if(($size > $largest_photo) || (! $largest_photo)) {
+ $ret['photo'] = $x->getAttribute('src');
+ $largest_photo = $size;
+ }
+ }
if((attribute_contains($x->getAttribute('class'),'nickname'))
- || (attribute_contains($x->getAttribute('class'),'uid')))
+ || (attribute_contains($x->getAttribute('class'),'uid'))) {
$ret['nick'] = $x->textContent;
+ }
}
}
}
$ret = array();
$s = fetch_url($url);
- if(! $s)
+ $headers = $a->get_curl_headers();
+ $code = $a->get_curl_code();
+
+ logger('scrape_feed: returns: ' . $code . ' headers=' . $headers, LOGGER_DEBUG);
+
+ if(! $s) {
+ logger('scrape_feed: no data returned for ' . $url);
return $ret;
+ }
- $headers = $a->get_curl_headers();
- logger('scrape_feed: headers=' . $headers, LOGGER_DEBUG);
$lines = explode("\n",$headers);
if(count($lines)) {
}
}
}
+ // perhaps an RSS version 1 feed with a generic or incorrect content-type?
+ if(stristr($s,'</item>')) {
+ $ret['feed_rss'] = $url;
+ return $ret;
+ }
}
- $dom = HTML5_Parser::parse($s);
+ try {
+ $dom = HTML5_Parser::parse($s);
+ } catch (DOMException $e) {
+ logger('scrape_feed: parse error: ' . $e);
+ }
- if(! $dom)
+ if(! $dom) {
+ logger('scrape_feed: failed to parse.');
return $ret;
-
-
- $items = $dom->getElementsByTagName('img');
-
- // get img elements (twitter)
-
- if($items) {
- foreach($items as $item) {
- $x = $item->getAttribute('id');
- if($x === 'profile-image') {
- $ret['photo'] = $item->getAttribute('src');
- }
- }
}
}
}
if(! $basename)
- $basename = substr($url,0,strrpos($url,'/')) . '/';
+ $basename = implode('/', array_slice(explode('/',$url),0,3)) . '/';
$items = $dom->getElementsByTagName('link');
}}
-function probe_url($url) {
+/**
+ *
+ * Probe a network address to discover what kind of protocols we need to communicate with it.
+ *
+ * Warning: this function is a bit touchy and there are some subtle dependencies within the logic flow.
+ * Edit with care.
+ *
+ */
+
+/**
+ *
+ * PROBE_DIASPORA has a bias towards returning Diaspora information
+ * while PROBE_NORMAL has a bias towards dfrn/zot - in the case where
+ * an address (such as a Friendica address) supports more than one type
+ * of network.
+ *
+ */
+
+
+define ( 'PROBE_NORMAL', 0);
+define ( 'PROBE_DIASPORA', 1);
+
+function probe_url($url, $mode = PROBE_NORMAL) {
require_once('include/email.php');
$result = array();
if(! $url)
return $result;
- $diaspora = false;
+ $network = null;
+ $diaspora = false;
+ $diaspora_base = '';
+ $diaspora_guid = '';
+ $diaspora_key = '';
+ $has_lrdd = false;
$email_conversant = false;
$twitter = ((strpos($url,'twitter.com') !== false) ? true : false);
+ $lastfm = ((strpos($url,'last.fm/user') !== false) ? true : false);
+
+ $at_addr = ((strpos($url,'@') !== false) ? true : false);
- if(! $twitter) {
- $links = lrdd($url);
+ if((! $twitter) && (! $lastfm)) {
+
+ if(strpos($url,'mailto:') !== false && $at_addr) {
+ $url = str_replace('mailto:','',$url);
+ $links = array();
+ }
+ else
+ $links = lrdd($url);
if(count($links)) {
+ $has_lrdd = true;
+
logger('probe_url: found lrdd links: ' . print_r($links,true), LOGGER_DATA);
foreach($links as $link) {
if($link['@attributes']['rel'] === NAMESPACE_ZOT)
$hcard = unamp($link['@attributes']['href']);
if($link['@attributes']['rel'] === 'http://webfinger.net/rel/profile-page')
$profile = unamp($link['@attributes']['href']);
- if($link['@attributes']['rel'] === 'http://joindiaspora.com/seed_location')
+ if($link['@attributes']['rel'] === 'http://portablecontacts.net/spec/1.0')
+ $poco = unamp($link['@attributes']['href']);
+ if($link['@attributes']['rel'] === 'http://joindiaspora.com/seed_location') {
+ $diaspora_base = unamp($link['@attributes']['href']);
+ $diaspora = true;
+ }
+ if($link['@attributes']['rel'] === 'http://joindiaspora.com/guid') {
+ $diaspora_guid = unamp($link['@attributes']['href']);
+ $diaspora = true;
+ }
+ if($link['@attributes']['rel'] === 'diaspora-public-key') {
+ $diaspora_key = base64_decode(unamp($link['@attributes']['href']));
+ if(strstr($diaspora_key,'RSA '))
+ $pubkey = rsatopem($diaspora_key);
+ else
+ $pubkey = $diaspora_key;
$diaspora = true;
+ }
}
// Status.Net can have more than one profile URL. We need to match the profile URL
}
}
}
- else {
+ elseif($mode == PROBE_NORMAL) {
// Check email
intval(local_user())
);
if(count($x) && count($r)) {
- $mailbox = construct_mailbox_name($r[0]);
+ $mailbox = construct_mailbox_name($r[0]);
$password = '';
openssl_private_decrypt(hex2bin($r[0]['pass']),$password,$x[0]['prvkey']);
$mbox = email_connect($mailbox,$r[0]['user'],$password);
+ if(! $mbox)
+ logger('probe_url: email_connect failed.');
unset($password);
}
if($mbox) {
$msgs = email_poll($mbox,$orig_url);
+ logger('probe_url: searching ' . $orig_url . ', ' . count($msgs) . ' messages found.', LOGGER_DEBUG);
if(count($msgs)) {
$addr = $orig_url;
$network = NETWORK_MAIL;
$name = substr($url,0,strpos($url,'@'));
- $profile = 'http://' . substr($url,strpos($url,'@')+1);
+ $phost = substr($url,strpos($url,'@')+1);
+ $profile = 'http://' . $phost;
// fix nick character range
- $vcard = array('fn' => $name, 'nick' => $name, 'photo' => gravatar_img($url));
+ $vcard = array('fn' => $name, 'nick' => $name, 'photo' => avatar_img($url));
$notify = 'smtp ' . random_string();
$poll = 'email ' . random_string();
$priority = 0;
$adr = imap_rfc822_parse_adrlist($x->from,'');
elseif(stristr($x->to,$orig_url))
$adr = imap_rfc822_parse_adrlist($x->to,'');
- if(isset($adr) && strlen($adr[0]->personal))
- $vcard['fn'] = notags($adr[0]->personal);
+ if(isset($adr)) {
+ foreach($adr as $feadr) {
+ if((strcasecmp($feadr->mailbox,$name) == 0)
+ &&(strcasecmp($feadr->host,$phost) == 0)
+ && (strlen($feadr->personal))) {
+
+ $personal = imap_mime_header_decode($feadr->personal);
+ $vcard['fn'] = "";
+ foreach($personal as $perspart)
+ if ($perspart->charset != "default")
+ $vcard['fn'] .= iconv($perspart->charset, 'UTF-8//IGNORE', $perspart->text);
+ else
+ $vcard['fn'] .= $perspart->text;
+
+ $vcard['fn'] = notags($vcard['fn']);
+ }
+ }
+ }
}
imap_close($mbox);
}
}
}
- if(strlen($zot)) {
- $s = fetch_url($zot);
- if($s) {
- $j = json_decode($s);
- if($j) {
- $network = NETWORK_ZOT;
- $vcard = array(
- 'fn' => $j->name,
- 'nick' => $j->username,
- 'photo' => $j->photo
- );
- $profile = $j->url;
- $notify = $j->post;
- $key = $j->pubkey;
- $poll = 'N/A';
+ if($mode == PROBE_NORMAL) {
+ if(strlen($zot)) {
+ $s = fetch_url($zot);
+ if($s) {
+ $j = json_decode($s);
+ if($j) {
+ $network = NETWORK_ZOT;
+ $vcard = array(
+ 'fn' => $j->fullname,
+ 'nick' => $j->nickname,
+ 'photo' => $j->photo
+ );
+ $profile = $j->url;
+ $notify = $j->post;
+ $pubkey = $j->pubkey;
+ $poll = 'N/A';
+ }
}
}
- }
- if(strlen($dfrn)) {
- $ret = scrape_dfrn($dfrn);
- if(is_array($ret) && x($ret,'dfrn-request')) {
- $network = NETWORK_DFRN;
- $request = $ret['dfrn-request'];
- $confirm = $ret['dfrn-confirm'];
- $notify = $ret['dfrn-notify'];
- $poll = $ret['dfrn-poll'];
+ if(strlen($dfrn)) {
+ $ret = scrape_dfrn(($hcard) ? $hcard : $dfrn);
+ if(is_array($ret) && x($ret,'dfrn-request')) {
+ $network = NETWORK_DFRN;
+ $request = $ret['dfrn-request'];
+ $confirm = $ret['dfrn-confirm'];
+ $notify = $ret['dfrn-notify'];
+ $poll = $ret['dfrn-poll'];
+
+ $vcard = array();
+ $vcard['fn'] = $ret['fn'];
+ $vcard['nick'] = $ret['nick'];
+ $vcard['photo'] = $ret['photo'];
+ }
}
}
+ if($diaspora && $diaspora_base && $diaspora_guid) {
+ if($mode == PROBE_DIASPORA || ! $notify) {
+ $notify = $diaspora_base . 'receive/users/' . $diaspora_guid;
+ $batch = $diaspora_base . 'receive/public' ;
+ }
+ if(strpos($url,'@'))
+ $addr = str_replace('acct:', '', $url);
+ }
+
if($network !== NETWORK_ZOT && $network !== NETWORK_DFRN && $network !== NETWORK_MAIL) {
- $network = NETWORK_OSTATUS;
+ if($diaspora)
+ $network = NETWORK_DIASPORA;
+ elseif($has_lrdd)
+ $network = NETWORK_OSTATUS;
$priority = 0;
- if($hcard) {
+ if($hcard && ! $vcard) {
$vcard = scrape_vcard($hcard);
// Google doesn't use absolute url in profile photos
logger('probe_url: scrape_vcard: ' . print_r($vcard,true), LOGGER_DATA);
}
- if(! $profile) {
- if($diaspora)
- $profile = $hcard;
- else
- $profile = $url;
- }
-
if($twitter) {
logger('twitter: setup');
$tid = basename($url);
else
$poll = $tapi . '?screen_name=' . $tid;
$profile = 'http://twitter.com/#!/' . $tid;
+ //$vcard['photo'] = 'https://api.twitter.com/1/users/profile_image/' . $tid;
+ $vcard['photo'] = 'https://api.twitter.com/1/users/profile_image?screen_name=' . $tid . '&size=bigger';
+ $vcard['nick'] = $tid;
+ $vcard['fn'] = $tid;
+ }
+
+ if($lastfm) {
+ $profile = $url;
+ $poll = str_replace(array('www.','last.fm/'),array('','ws.audioscrobbler.com/1.0/'),$url) . '/recenttracks.rss';
+ $vcard['nick'] = basename($url);
+ $vcard['fn'] = $vcard['nick'] . t(' on Last.fm');
+ $network = NETWORK_FEED;
}
if(! x($vcard,'fn'))
if(x($vcard,'nick'))
$vcard['fn'] = $vcard['nick'];
-
- if(((! isset($vcard)) && (! $poll)) || ($twitter)) {
+ $check_feed = false;
+
+ if($twitter || ! $poll)
+ $check_feed = true;
+ if((! isset($vcard)) || (! x($vcard,'fn')) || (! $profile))
+ $check_feed = true;
+ if(($at_addr) && (! count($links)))
+ $check_feed = false;
+
+ if($check_feed) {
- $feedret = scrape_feed($url);
- logger('probe_url: scrape_feed returns: ' . print_r($feedret,true), LOGGER_DATA);
+ $feedret = scrape_feed(($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']));
- $vcard = array();
+ if(! x($vcard))
+ $vcard = array();
}
- if(x($feedret,'photo'))
+ if(x($feedret,'photo') && (! x($vcard,'photo')))
$vcard['photo'] = $feedret['photo'];
require_once('library/simplepie/simplepie.inc');
$feed = new SimplePie();
logger('probe_url: fetch feed: ' . $poll . ' returns: ' . $xml, LOGGER_DATA);
$a = get_app();
- logger('probe_url: scrape_feed: headers: ' . $a->get_curl_headers(), $LOGGER_DATA);
+ logger('probe_url: scrape_feed: headers: ' . $a->get_curl_headers(), LOGGER_DATA);
$feed->set_raw_data($xml);
if($feed->error())
logger('probe_url: scrape_feed: Error parsing XML: ' . $feed->error());
+
if(! x($vcard,'photo'))
$vcard['photo'] = $feed->get_image_url();
$author = $feed->get_author();
+
if($author) {
$vcard['fn'] = unxmlify(trim($author->get_name()));
if(! $vcard['fn'])
if(strpos($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()));
if(! $vcard['photo']) {
$rawtags = $feed->get_feed_tags( SIMPLEPIE_NAMESPACE_ATOM_10, 'author');
if($rawtags) {
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()));
}
if(! $vcard['photo']) {
$rawmedia = $item->get_item_tags('http://search.yahoo.com/mrss/','thumbnail');
}
}
}
+
if((! $vcard['photo']) && strlen($email))
- $vcard['photo'] = gravatar_img($email);
+ $vcard['photo'] = avatar_img($email);
if($poll === $profile)
$lnk = $feed->get_permalink();
if(isset($lnk) && strlen($lnk))
if(strpos($vcard['nick'],' '))
$vcard['nick'] = trim(substr($vcard['nick'],0,strpos($vcard['nick'],' ')));
}
- $network = 'feed';
- $priority = 2;
+ if(! $network)
+ $network = NETWORK_FEED;
+ if(! $priority)
+ $priority = 2;
}
}
if(! x($vcard,'photo')) {
$a = get_app();
- $vcard['photo'] = $a->get_baseurl() . '/images/default-profile.jpg' ;
+ $vcard['photo'] = $a->get_baseurl() . '/images/person-175.jpg' ;
}
- $vcard['fn'] = notags($vcard['fn']);
- $vcard['nick'] = notags($vcard['nick']);
+ if(! $profile)
+ $profile = $url;
+ // No human could be associated with this link, use the URL as the contact name
+
+ if(($network === NETWORK_FEED) && ($poll) && (! x($vcard,'fn')))
+ $vcard['fn'] = $url;
+
+ $vcard['fn'] = notags($vcard['fn']);
+ $vcard['nick'] = str_replace(' ','',notags($vcard['nick']));
+
$result['name'] = $vcard['fn'];
$result['nick'] = $vcard['nick'];
$result['url'] = $profile;
$result['addr'] = $addr;
+ $result['batch'] = $batch;
$result['notify'] = $notify;
$result['poll'] = $poll;
$result['request'] = $request;
$result['confirm'] = $confirm;
+ $result['poco'] = $poco;
$result['photo'] = $vcard['photo'];
$result['priority'] = $priority;
$result['network'] = $network;
$result['alias'] = $alias;
- $result['key'] = $key;
+ $result['pubkey'] = $pubkey;
logger('probe_url: ' . print_r($result,true), LOGGER_DEBUG);