keith/friendica
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Merge pull request #2669 from annando/1607-new-probe

Browse source 
New probing mechanism
This commit is contained in:
rabuzarus 2016-07-11 21:42:48 +02:00 committed by GitHub
commit 4e4580f326
12 changed files with 1221 additions and 1280 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
doc/database/db_fcontact.md
View file

@ -4,6 +4,7 @@ Table fcontact
| Field | Description | Type | Null | Key | Default | Extra |
| -------- | ------------- | ---------------- | ---- | --- | ------------------- | --------------- |
| id | sequential ID | int(10) unsigned | NO | PRI | NULL | auto_increment |
| guid | unique id | varchar(64) | NO | | | |
| url | | varchar(255) | NO | | | |
| name | | varchar(255) | NO | | | |
| photo | | varchar(255) | NO | | | |

1121
include/Probe.php Normal file
View file

File diff suppressed because it is too large Load diff

942
include/Scrape.php
View file

@ -1,341 +1,5 @@
<?php
require_once('library/HTML5/Parser.php');
require_once('include/crypto.php');
require_once('include/feed.php');
if(! function_exists('scrape_dfrn')) {
function scrape_dfrn($url, $dont_probe = false) {
$a = get_app();
$ret = array();
logger('scrape_dfrn: url=' . $url);
// Try to fetch the data from noscrape. This is faster than parsing the HTML
$noscrape = str_replace("/hcard/", "/noscrape/", $url);
$noscrapejson = fetch_url($noscrape);
$noscrapedata = array();
if ($noscrapejson) {
$noscrapedata = json_decode($noscrapejson, true);
if (is_array($noscrapedata)) {
if ($noscrapedata["nick"] != "")
return($noscrapedata);
else
unset($noscrapedata["nick"]);
} else
$noscrapedata = array();
}
$s = fetch_url($url);
if (!$s)
return $ret;
if (!$dont_probe) {
$probe = probe_url($url);
if (isset($probe["addr"]))
$ret["addr"] = $probe["addr"];
}
$headers = $a->get_curl_headers();
logger('scrape_dfrn: headers=' . $headers, LOGGER_DEBUG);
$lines = explode("\n",$headers);
if(count($lines)) {
foreach($lines as $line) {
// don't try and run feeds through the html5 parser
if(stristr($line,'content-type:') && ((stristr($line,'application/atom+xml')) || (stristr($line,'application/rss+xml'))))
return ret;
}
}
try {
$dom = HTML5_Parser::parse($s);
} catch (DOMException $e) {
logger('scrape_dfrn: parse error: ' . $e);
}
if(! $dom)
return $ret;
$items = $dom->getElementsByTagName('link');
// get DFRN link elements
foreach($items as $item) {
$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-") {
$ret[$x] = $item->getAttribute('href');
}
if($x === 'lrdd') {
$decoded = urldecode($item->getAttribute('href'));
if(preg_match('/acct:([^@]*)@/',$decoded,$matches))
$ret['nick'] = $matches[1];
}
}
// 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'),'uid'))
$ret['guid'] = $x->textContent;
if(attribute_contains($x->getAttribute('class'),'nickname'))
$ret['nickname'] = $x->textContent;
if(attribute_contains($x->getAttribute('class'),'fn'))
$ret['fn'] = $x->textContent;
if(attribute_contains($x->getAttribute('class'),'searchable'))
$ret['searchable'] = $x->textContent;
if(attribute_contains($x->getAttribute('class'),'key'))
$ret['key'] = $x->textContent;
if(attribute_contains($x->getAttribute('class'),'url'))
$ret['url'] = $x->textContent;
if((attribute_contains($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 : $size);
}
}
}
}
}
return array_merge($ret, $noscrapedata);
}}
if(! function_exists('validate_dfrn')) {
function validate_dfrn($a) {
$errors = 0;
if(! x($a,'key'))
$errors ++;
if(! x($a,'dfrn-request'))
$errors ++;
if(! x($a,'dfrn-confirm'))
$errors ++;
if(! x($a,'dfrn-notify'))
$errors ++;
if(! x($a,'dfrn-poll'))
$errors ++;
return $errors;
}}
if(! function_exists('scrape_meta')) {
function scrape_meta($url) {
$a = get_app();
$ret = array();
logger('scrape_meta: url=' . $url);
$s = fetch_url($url);
if(! $s)
return $ret;
$headers = $a->get_curl_headers();
logger('scrape_meta: headers=' . $headers, LOGGER_DEBUG);
$lines = explode("\n",$headers);
if(count($lines)) {
foreach($lines as $line) {
// don't try and run feeds through the html5 parser
if(stristr($line,'content-type:') && ((stristr($line,'application/atom+xml')) || (stristr($line,'application/rss+xml'))))
return ret;
}
}
try {
$dom = HTML5_Parser::parse($s);
} catch (DOMException $e) {
logger('scrape_meta: parse error: ' . $e);
}
if(! $dom)
return $ret;
$items = $dom->getElementsByTagName('meta');
// get DFRN link elements
foreach($items as $item) {
$x = $item->getAttribute('name');
if(substr($x,0,5) == "dfrn-")
$ret[$x] = $item->getAttribute('content');
}
return $ret;
}}
if(! function_exists('scrape_vcard')) {
function scrape_vcard($url) {
$a = get_app();
$ret = array();
logger('scrape_vcard: url=' . $url);
$s = fetch_url($url);
if(! $s)
return $ret;
$headers = $a->get_curl_headers();
$lines = explode("\n",$headers);
if(count($lines)) {
foreach($lines as $line) {
// don't try and run feeds through the html5 parser
if(stristr($line,'content-type:') && ((stristr($line,'application/atom+xml')) || (stristr($line,'application/rss+xml'))))
return ret;
}
}
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')) {
$level2 = $item->getElementsByTagName('*');
foreach($level2 as $x) {
if(attribute_contains($x->getAttribute('class'),'fn'))
$ret['fn'] = $x->textContent;
if((attribute_contains($x->getAttribute('class'),'photo'))
|| (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'))) {
$ret['nick'] = $x->textContent;
}
}
}
}
return $ret;
}}
if(! function_exists('scrape_feed')) {
function scrape_feed($url) {
$a = get_app();
$ret = array();
$cookiejar = tempnam(get_temppath(), 'cookiejar-scrape-feed-');
$s = fetch_url($url, false, $redirects, 0, Null, $cookiejar);
unlink($cookiejar);
$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;
}
$lines = explode("\n",$headers);
if(count($lines)) {
foreach($lines as $line) {
if(stristr($line,'content-type:')) {
if(stristr($line,'application/atom+xml') || stristr($s,'<feed')) {
$ret['feed_atom'] = $url;
return $ret;
}
if(stristr($line,'application/rss+xml') || stristr($s,'<rss')) {
$ret['feed_rss'] = $url;
return $ret;
}
}
}
// perhaps an RSS version 1 feed with a generic or incorrect content-type?
if(stristr($s,'</item>')) {
$ret['feed_rss'] = $url;
return $ret;
}
}
$basename = implode('/', array_slice(explode('/',$url),0,3)) . '/';
$doc = new DOMDocument();
@$doc->loadHTML($s);
$xpath = new DomXPath($doc);
$base = $xpath->query("//base");
foreach ($base as $node) {
$attr = array();
if ($node->attributes->length)
foreach ($node->attributes as $attribute)
$attr[$attribute->name] = $attribute->value;
if ($attr["href"] != "")
$basename = $attr["href"] ;
}
$list = $xpath->query("//link");
foreach ($list as $node) {
$attr = array();
if ($node->attributes->length)
foreach ($node->attributes as $attribute)
$attr[$attribute->name] = $attribute->value;
if (($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"] = $attr["href"];
}
// Drupal and perhaps others only provide relative URLs. Turn them into absolute.
if(x($ret,'feed_atom') && (! strstr($ret['feed_atom'],'://')))
$ret['feed_atom'] = $basename . $ret['feed_atom'];
if(x($ret,'feed_rss') && (! strstr($ret['feed_rss'],'://')))
$ret['feed_rss'] = $basename . $ret['feed_rss'];
return $ret;
}}
require_once('include/Probe.php');
/**
*
@ -355,605 +19,17 @@ function scrape_feed($url) {
*
*/
define ( 'PROBE_NORMAL', 0);
define ( 'PROBE_DIASPORA', 1);
define('PROBE_NORMAL', 0);
define('PROBE_DIASPORA', 1);
function probe_url($url, $mode = PROBE_NORMAL, $level = 1) {
require_once('include/email.php');
$result = array();
if ($mode == PROBE_DIASPORA)
$network = NETWORK_DIASPORA;
else
$network = "";
if (!$url)
return $result;
$data = Probe::uri($url, $network);
$result = Cache::get("probe_url:".$mode.":".$url);
if (!is_null($result)) {
$result = unserialize($result);
return $result;
}
$original_url = $url;
$network = null;
$diaspora = false;
$diaspora_base = '';
$diaspora_guid = '';
$diaspora_key = '';
$has_lrdd = false;
$email_conversant = false;
$connectornetworks = false;
$appnet = false;
if (strpos($url,'twitter.com')) {
$connectornetworks = true;
$network = NETWORK_TWITTER;
}
$lastfm = ((strpos($url,'last.fm/user') !== false) ? true : false);
$at_addr = ((strpos($url,'@') !== false) ? true : false);
if((!$appnet) && (!$lastfm) && !$connectornetworks) {
if(strpos($url,'mailto:') !== false && $at_addr) {
$url = str_replace('mailto:','',$url);
$links = array();
}
else
$links = lrdd($url);
if ((count($links) == 0) AND strstr($url, "/index.php")) {
$url = str_replace("/index.php", "", $url);
$links = lrdd($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)
$zot = unamp($link['@attributes']['href']);
if($link['@attributes']['rel'] === NAMESPACE_DFRN)
$dfrn = unamp($link['@attributes']['href']);
if($link['@attributes']['rel'] === 'salmon')
$notify = unamp($link['@attributes']['href']);
if($link['@attributes']['rel'] === NAMESPACE_FEED)
$poll = unamp($link['@attributes']['href']);
if($link['@attributes']['rel'] === 'http://microformats.org/profile/hcard')
$hcard = unamp($link['@attributes']['href']);
if($link['@attributes']['rel'] === 'http://webfinger.net/rel/profile-page')
$profile = unamp($link['@attributes']['href']);
if($link['@attributes']['rel'] === 'http://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;
$diaspora = true;
}
if(($link['@attributes']['rel'] === 'http://ostatus.org/schema/1.0/subscribe') AND ($mode == PROBE_NORMAL)) {
$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
// otherwise we're screwed.
$backup_alias = "";
foreach($links as $link) {
if($link['@attributes']['rel'] === 'alias') {
if(strpos($link['@attributes']['href'],'@') === false) {
if(isset($profile)) {
$alias_url = $link['@attributes']['href'];
if(($alias_url !== $profile) AND ($backup_alias == "") AND
($alias_url !== str_replace("/index.php", "", $profile)))
$backup_alias = $alias_url;
if(($alias_url !== $profile) AND !strstr($alias_url, "index.php") AND
($alias_url !== str_replace("/index.php", "", $profile)))
$alias = $alias_url;
}
else
$profile = unamp($link['@attributes']['href']);
}
}
}
if ($alias == "")
$alias = $backup_alias;
// If the profile is different from the url then the url is abviously an alias
if (($alias == "") AND ($profile != "") AND !$at_addr AND (normalise_link($profile) != normalise_link($url)))
$alias = $url;
}
elseif($mode == PROBE_NORMAL) {
// Check email
$orig_url = $url;
if((strpos($orig_url,'@')) && validate_email($orig_url)) {
$x = q("SELECT `prvkey` FROM `user` WHERE `uid` = %d LIMIT 1",
intval(local_user())
);
$r = q("SELECT * FROM `mailacct` WHERE `uid` = %d AND `server` != '' LIMIT 1",
intval(local_user())
);
if(count($x) && count($r)) {
$mailbox = construct_mailbox_name($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,'@'));
$phost = substr($url,strpos($url,'@')+1);
$profile = 'http://' . $phost;
// fix nick character range
$vcard = array('fn' => $name, 'nick' => $name, 'photo' => avatar_img($url));
$notify = 'smtp ' . random_string();
$poll = 'email ' . random_string();
$priority = 0;
$x = email_msg_meta($mbox,$msgs[0]);
if(stristr($x[0]->from,$orig_url))
$adr = imap_rfc822_parse_adrlist($x[0]->from,'');
elseif(stristr($x[0]->to,$orig_url))
$adr = imap_rfc822_parse_adrlist($x[0]->to,'');
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($mbox);
}
}
}
}
if($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)) {
$ret = scrape_dfrn(($hcard) ? $hcard : $dfrn, true);
if(is_array($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'] = $ret['photo'];
}
}
}
// Scrape the public key from the hcard.
// Diaspora will remove it from the webfinger somewhere in the future.
if (($hcard != "") AND ($pubkey == "")) {
$ret = scrape_dfrn(($hcard) ? $hcard : $dfrn, true);
if (isset($ret["guid"]))
$diaspora_guid = $ret["guid"];
if (isset($ret["url"]))
$diaspora_base = $ret["url"];
if (isset($ret["key"])) {
$hcard_key = $ret["key"];
if(strstr($hcard_key,'RSA '))
$pubkey = rsatopem($hcard_key);
else
$pubkey = $hcard_key;
}
}
if(($network == "") && $diaspora_base && $diaspora_guid) {
$diaspora_notify = $diaspora_base.'receive/users/'.$diaspora_guid;
$diaspora = true;
if($mode == PROBE_DIASPORA || !$notify || ($notify == $diaspora_notify)) {
$notify = $diaspora_notify;
$batch = $diaspora_base . 'receive/public' ;
}
if(strpos($url,'@'))
$addr = str_replace('acct:', '', $url);
}
if($network !== NETWORK_ZOT && $network !== NETWORK_DFRN && $network !== NETWORK_MAIL) {
if($diaspora)
$network = NETWORK_DIASPORA;
elseif($has_lrdd AND ($notify))
$network = NETWORK_OSTATUS;
if(strpos($url,'@'))
$addr = str_replace('acct:', '', $url);
$priority = 0;
if($hcard && ! $vcard) {
$vcard = scrape_vcard($hcard);
// Google doesn't use absolute url in profile photos
if((x($vcard,'photo')) && substr($vcard['photo'],0,1) == '/') {
$h = @parse_url($hcard);
if($h)
$vcard['photo'] = $h['scheme'] . '://' . $h['host'] . $vcard['photo'];
}
logger('probe_url: scrape_vcard: ' . print_r($vcard,true), LOGGER_DATA);
}
if($diaspora && $addr) {
// Diaspora returns the name as the nick. As the nick will never be updated,
// let's use the Diaspora nickname (the first part of the handle) as the nick instead
$addr_parts = explode('@', $addr);
$vcard['nick'] = $addr_parts[0];
}
if($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'];
$check_feed = false;
if(stristr($url,'tumblr.com') && (! stristr($url,'/rss'))) {
$poll = $url . '/rss';
$check_feed = true;
// Will leave it to others to figure out how to grab the avatar, which is on the $url page in the open graph meta links
}
if($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']));
if(! x($vcard))
$vcard = array();
}
if(x($feedret,'photo') && (! x($vcard,'photo')))
$vcard['photo'] = $feedret['photo'];
$cookiejar = tempnam(get_temppath(), 'cookiejar-scrape-feed-');
$xml = fetch_url($poll, false, $redirects, 0, Null, $cookiejar);
unlink($cookiejar);
logger('probe_url: fetch feed: ' . $poll . ' returns: ' . $xml, LOGGER_DATA);
if ($xml == "") {
logger("scrape_feed: XML is empty for feed ".$poll);
$network = NETWORK_PHANTOM;
} else {
$data = feed_import($xml,$dummy1,$dummy2, $dummy3, true);
if (!is_array($data)) {
logger("scrape_feed: This doesn't seem to be a feed: ".$poll);
$network = NETWORK_PHANTOM;
} else {
if (($vcard["photo"] == "") AND ($data["header"]["author-avatar"] != ""))
$vcard["photo"] = $data["header"]["author-avatar"];
if (($vcard["fn"] == "") AND ($data["header"]["author-name"] != ""))
$vcard["fn"] = $data["header"]["author-name"];
if (($vcard["nick"] == "") AND ($data["header"]["author-nick"] != ""))
$vcard["nick"] = $data["header"]["author-nick"];
if ($network == NETWORK_OSTATUS) {
if ($data["header"]["author-id"] != "")
$alias = $data["header"]["author-id"];
if ($data["header"]["author-link"] != "")
$profile = $data["header"]["author-link"];
} elseif(!$profile AND ($data["header"]["author-link"] != "") AND !in_array($network, array("", NETWORK_FEED)))
$profile = $data["header"]["author-link"];
}
}
// Workaround for misconfigured Friendica servers
if (($network == "") AND (strstr($url, "/profile/"))) {
$noscrape = str_replace("/profile/", "/noscrape/", $url);
$noscrapejson = fetch_url($noscrape);
if ($noscrapejson) {
$network = NETWORK_DFRN;
$poco = str_replace("/profile/", "/poco/", $url);
$noscrapedata = json_decode($noscrapejson, true);
if (isset($noscrapedata["addr"]))
$addr = $noscrapedata["addr"];
if (isset($noscrapedata["fn"]))
$vcard["fn"] = $noscrapedata["fn"];
if (isset($noscrapedata["key"]))
$pubkey = $noscrapedata["key"];
if (isset($noscrapedata["photo"]))
$vcard["photo"] = $noscrapedata["photo"];
if (isset($noscrapedata["dfrn-request"]))
$request = $noscrapedata["dfrn-request"];
if (isset($noscrapedata["dfrn-confirm"]))
$confirm = $noscrapedata["dfrn-confirm"];
if (isset($noscrapedata["dfrn-notify"]))
$notify = $noscrapedata["dfrn-notify"];
if (isset($noscrapedata["dfrn-poll"]))
$poll = $noscrapedata["dfrn-poll"];
}
}
if(! $network)
$network = NETWORK_FEED;
if(! x($vcard,'nick')) {
$vcard['nick'] = strtolower(notags(unxmlify($vcard['fn'])));
if(strpos($vcard['nick'],' '))
$vcard['nick'] = trim(substr($vcard['nick'],0,strpos($vcard['nick'],' ')));
}
if(! $priority)
$priority = 2;
}
}
if(! x($vcard,'photo')) {
$a = get_app();
$vcard['photo'] = App::get_baseurl() . '/images/person-175.jpg' ;
}
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'] = $url;
if ($diaspora_base != "")
$baseurl = $diaspora_base;
if (($baseurl == "") AND ($notify != "") AND ($poll != "")) {
$baseurl = matching_url(normalise_link($notify), normalise_link($poll));
$baseurl2 = matching_url($baseurl, normalise_link($profile));
if ($baseurl2 != "")
$baseurl = $baseurl2;
}
if (($baseurl == "") AND ($notify != ""))
$baseurl = matching_url(normalise_link($profile), normalise_link($notify));
if (($baseurl == "") AND ($poll != ""))
$baseurl = matching_url(normalise_link($profile), normalise_link($poll));
if (substr($baseurl, -10) == "/index.php")
$baseurl = str_replace("/index.php", "", $baseurl);
if ($network == "")
$network = NETWORK_PHANTOM;
$baseurl = rtrim($baseurl, "/");
if(strpos($url,'@') AND ($addr == "") AND ($network == NETWORK_DFRN))
$addr = str_replace('acct:', '', $url);
$vcard['fn'] = notags($vcard['fn']);
$vcard['nick'] = str_replace(' ','',notags($vcard['nick']));
$result['name'] = $vcard['fn'];
$result['nick'] = $vcard['nick'];
$result['guid'] = $diaspora_guid;
$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['pubkey'] = $pubkey;
$result['baseurl'] = $baseurl;
logger('probe_url: ' . print_r($result,true), LOGGER_DEBUG);
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 (!in_array($result2['network'], array("", NETWORK_PHANTOM, NETWORK_FEED)))
$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 (!in_array($result2['network'], array("", NETWORK_PHANTOM, NETWORK_FEED)))
$result = $result2;
}
// Quickfix for Hubzilla systems with enabled OStatus plugin
if (($result['network'] == NETWORK_DIASPORA) AND ($result["batch"] == "")) {
$result2 = probe_url($url, PROBE_DIASPORA, ++$level);
if ($result2['network'] == NETWORK_DIASPORA) {
$addr = $result["addr"];
$result = $result2;
if (($result["addr"] == "") AND ($addr != ""))
$result["addr"] = $addr;
}
}
}
// Only store into the cache if the value seems to be valid
if ($result['network'] != NETWORK_PHANTOM) {
Cache::set("probe_url:".$mode.":".$original_url,serialize($result), CACHE_DAY);
/// @todo temporary fix - we need a real contact update function that updates only changing fields
/// The biggest problem is the avatar picture that could have a reduced image size.
/// It should only be updated if the existing picture isn't existing anymore.
if (($result['network'] != NETWORK_FEED) AND ($mode == PROBE_NORMAL) AND
$result["name"] AND $result["nick"] AND $result["url"] AND $result["addr"] AND $result["poll"])
q("UPDATE `contact` SET `name` = '%s', `nick` = '%s', `url` = '%s', `addr` = '%s',
`notify` = '%s', `poll` = '%s', `alias` = '%s', `success_update` = '%s'
WHERE `nurl` = '%s' AND NOT `self` AND `uid` = 0",
dbesc($result["name"]),
dbesc($result["nick"]),
dbesc($result["url"]),
dbesc($result["addr"]),
dbesc($result["notify"]),
dbesc($result["poll"]),
dbesc($result["alias"]),
dbesc(datetime_convert()),
dbesc(normalise_link($result['url']))
);
}
return $result;
}
/**
* @brief Find the matching part between two url
*
* @param string $url1
* @param string $url2
* @return string The matching part
*/
function matching_url($url1, $url2) {
if (($url1 == "") OR ($url2 == ""))
return "";
$url1 = normalise_link($url1);
$url2 = normalise_link($url2);
$parts1 = parse_url($url1);
$parts2 = parse_url($url2);
if (!isset($parts1["host"]) OR !isset($parts2["host"]))
return "";
if ($parts1["scheme"] != $parts2["scheme"])
return "";
if ($parts1["host"] != $parts2["host"])
return "";
if ($parts1["port"] != $parts2["port"])
return "";
$match = $parts1["scheme"]."://".$parts1["host"];
if ($parts1["port"])
$match .= ":".$parts1["port"];
$pathparts1 = explode("/", $parts1["path"]);
$pathparts2 = explode("/", $parts2["path"]);
$i = 0;
$path = "";
do {
$path1 = $pathparts1[$i];
$path2 = $pathparts2[$i];
if ($path1 == $path2)
$path .= $path1."/";
} while (($path1 == $path2) AND ($i++ <= count($pathparts1)));
$match .= $path;
return normalise_link($match);
return $data;
}

25
include/feed.php
View file

@ -59,8 +59,6 @@ function feed_import($xml,$importer,&$contact, &$hub, $simulate = false) {
if ($attributes->name == "href")
$author["author-link"] = $attributes->textContent;
$author["author-id"] = $xpath->evaluate('/atom:feed/atom:author/atom:uri/text()')->item(0)->nodeValue;
if ($author["author-link"] == "")
$author["author-link"] = $author["author-id"];
@ -89,9 +87,22 @@ function feed_import($xml,$importer,&$contact, &$hub, $simulate = false) {
if ($value != "")
$author["author-name"] = $value;
$value = $xpath->evaluate('atom:author/poco:preferredUsername/text()')->item(0)->nodeValue;
if ($value != "")
$author["author-nick"] = $value;
if ($simulate) {
$author["author-id"] = $xpath->evaluate('/atom:feed/atom:author/atom:uri/text()')->item(0)->nodeValue;
$value = $xpath->evaluate('atom:author/poco:preferredUsername/text()')->item(0)->nodeValue;
if ($value != "")
$author["author-nick"] = $value;
$value = $xpath->evaluate('atom:author/poco:address/poco:formatted/text()', $context)->item(0)->nodeValue;
if ($value != "")
$author["author-location"] = $value;
$value = $xpath->evaluate('atom:author/poco:note/text()')->item(0)->nodeValue;
if ($value != "")
$author["author-about"] = $value;
}
$author["edited"] = $author["created"] = $xpath->query('/atom:feed/atom:updated/text()')->item(0)->nodeValue;
@ -131,10 +142,6 @@ function feed_import($xml,$importer,&$contact, &$hub, $simulate = false) {
$author["owner-link"] = $contact["url"];
$author["owner-name"] = $contact["name"];
$author["owner-avatar"] = $contact["thumb"];
// This is no field in the item table. So we have to unset it.
unset($author["author-nick"]);
unset($author["author-id"]);
}
$header = array();

376
include/network.php
View file

@ -5,7 +5,7 @@
*/
require_once("include/xml.php");
require_once('include/Probe.php');
/**
* @brief Curl wrapper
@ -371,324 +371,6 @@ function http_status_exit($val, $description = array()) {
}
// Given an email style address, perform webfinger lookup and
// return the resulting DFRN profile URL, or if no DFRN profile URL
// is located, returns an OStatus subscription template (prefixed
// with the string 'stat:' to identify it as on OStatus template).
// If this isn't an email style address just return $webbie.
// Return an empty string if email-style addresses but webfinger fails,
// or if the resultant personal XRD doesn't contain a supported
// subscription/friend-request attribute.
// amended 7/9/2011 to return an hcard which could save potentially loading
// a lengthy content page to scrape dfrn attributes
function webfinger_dfrn($webbie,&$hcard) {
if(! strstr($webbie,'@')) {
return $webbie;
}
$profile_link = '';
$links = webfinger($webbie);
logger('webfinger_dfrn: ' . $webbie . ':' . print_r($links,true), LOGGER_DATA);
if(count($links)) {
foreach($links as $link) {
if($link['@attributes']['rel'] === NAMESPACE_DFRN)
$profile_link = $link['@attributes']['href'];
if($link['@attributes']['rel'] === NAMESPACE_OSTATUSSUB)
$profile_link = 'stat:' . $link['@attributes']['template'];
if($link['@attributes']['rel'] === 'http://microformats.org/profile/hcard')
$hcard = $link['@attributes']['href'];
}
}
return $profile_link;
}
/**
* @brief Perform webfinger lookup on an email style address
*
* @param string $webbi An email style address
* @param boolean $debug
*
* @return array of link attributes from the personal XRD file
* empty array on error/failure
*/
function webfinger($webbie, $debug = false) {
$host = '';
if(strstr($webbie,'@')) {
$host = substr($webbie,strpos($webbie,'@') + 1);
}
if(strlen($host)) {
$tpl = fetch_lrdd_template($host);
logger('webfinger: lrdd template: ' . $tpl);
if(strlen($tpl)) {
$pxrd = str_replace('{uri}', urlencode('acct:' . $webbie), $tpl);
logger('webfinger: pxrd: ' . $pxrd);
$links = fetch_xrd_links($pxrd);
if(! count($links)) {
// try with double slashes
$pxrd = str_replace('{uri}', urlencode('acct://' . $webbie), $tpl);
logger('webfinger: pxrd: ' . $pxrd);
$links = fetch_xrd_links($pxrd);
}
return $links;
}
}
return array();
}
function lrdd($uri, $debug = false) {
$a = get_app();
// default priority is host priority, host-meta first
$priority = 'host';
// All we have is an email address. Resource-priority is irrelevant
// because our URI isn't directly resolvable.
if(strstr($uri,'@')) {
return(webfinger($uri));
}
// get the host meta file
$host = @parse_url($uri);
if($host) {
$url = ((x($host,'scheme')) ? $host['scheme'] : 'http') . '://';
$url .= $host['host'] . '/.well-known/host-meta' ;
}
else
return array();
logger('lrdd: constructed url: ' . $url);
$xml = fetch_url($url);
$headers = $a->get_curl_headers();
if (! $xml)
return array();
logger('lrdd: host_meta: ' . $xml, LOGGER_DATA);
if(! stristr($xml,'<xrd'))
return array();
$h = parse_xml_string($xml);
if(! $h)
return array();
$arr = xml::element_to_array($h);
if(isset($arr['xrd']['property'])) {
$property = $arr['crd']['property'];
if(! isset($property[0]))
$properties = array($property);
else
$properties = $property;
foreach($properties as $prop)
if((string) $prop['@attributes'] === 'http://lrdd.net/priority/resource')
$priority = 'resource';
}
// save the links in case we need them
$links = array();
if(isset($arr['xrd']['link'])) {
$link = $arr['xrd']['link'];
if(! isset($link[0]))
$links = array($link);
else
$links = $link;
}
// do we have a template or href?
if(count($links)) {
foreach($links as $link) {
if($link['@attributes']['rel'] && attribute_contains($link['@attributes']['rel'],'lrdd')) {
if(x($link['@attributes'],'template'))
$tpl = $link['@attributes']['template'];
elseif(x($link['@attributes'],'href'))
$href = $link['@attributes']['href'];
}
}
}
if((! isset($tpl)) || (! strpos($tpl,'{uri}')))
$tpl = '';
if($priority === 'host') {
if(strlen($tpl))
$pxrd = str_replace('{uri}', urlencode($uri), $tpl);
elseif(isset($href))
$pxrd = $href;
if(isset($pxrd)) {
logger('lrdd: (host priority) pxrd: ' . $pxrd);
$links = fetch_xrd_links($pxrd);
return $links;
}
$lines = explode("\n",$headers);
if(count($lines)) {
foreach($lines as $line) {
if((stristr($line,'link:')) && preg_match('/<([^>].*)>.*rel\=[\'\"]lrdd[\'\"]/',$line,$matches)) {
return(fetch_xrd_links($matches[1]));
break;
}
}
}
}
// priority 'resource'
$html = fetch_url($uri);
$headers = $a->get_curl_headers();
logger('lrdd: headers=' . $headers, LOGGER_DEBUG);
// don't try and parse raw xml as html
if(! strstr($html,'<?xml')) {
require_once('library/HTML5/Parser.php');
try {
$dom = HTML5_Parser::parse($html);
} catch (DOMException $e) {
logger('lrdd: parse error: ' . $e);
}
if(isset($dom) && $dom) {
$items = $dom->getElementsByTagName('link');
foreach($items as $item) {
$x = $item->getAttribute('rel');
if($x == "lrdd") {
$pagelink = $item->getAttribute('href');
break;
}
}
}
}
if(isset($pagelink))
return(fetch_xrd_links($pagelink));
// next look in HTTP headers
$lines = explode("\n",$headers);
if(count($lines)) {
foreach($lines as $line) {
/// @TODO Alter the following regex to support multiple relations (space separated)
if((stristr($line,'link:')) && preg_match('/<([^>].*)>.*rel\=[\'\"]lrdd[\'\"]/',$line,$matches)) {
$pagelink = $matches[1];
break;
}
// don't try and run feeds through the html5 parser
if(stristr($line,'content-type:') && ((stristr($line,'application/atom+xml')) || (stristr($line,'application/rss+xml'))))
return array();
if(stristr($html,'<rss') || stristr($html,'<feed'))
return array();
}
}
if(isset($pagelink))
return(fetch_xrd_links($pagelink));
// If we haven't found any links, return the host xrd links (which we have already fetched)
if(isset($links))
return $links;
return array();
}
// Given a host name, locate the LRDD template from that
// host. Returns the LRDD template or an empty string on
// error/failure.
function fetch_lrdd_template($host) {
$tpl = '';
$url1 = 'https://' . $host . '/.well-known/host-meta' ;
$url2 = 'http://' . $host . '/.well-known/host-meta' ;
$links = fetch_xrd_links($url1);
logger('fetch_lrdd_template from: ' . $url1);
logger('template (https): ' . print_r($links,true));
if(! count($links)) {
logger('fetch_lrdd_template from: ' . $url2);
$links = fetch_xrd_links($url2);
logger('template (http): ' . print_r($links,true));
}
if(count($links)) {
foreach($links as $link)
if($link['@attributes']['rel'] && $link['@attributes']['rel'] === 'lrdd' && (!$link['@attributes']['type'] || $link['@attributes']['type'] === 'application/xrd+xml'))
$tpl = $link['@attributes']['template'];
}
if(! strpos($tpl,'{uri}'))
$tpl = '';
return $tpl;
}
/**
* @brief Given a URL, retrieve the page as an XRD document.
*
* @param string $url An url
* @return array of links
* return empty array on error/failure
*/
function fetch_xrd_links($url) {
$xrd_timeout = intval(get_config('system','xrd_timeout'));
$redirects = 0;
$xml = fetch_url($url,false,$redirects,(($xrd_timeout) ? $xrd_timeout : 20), "application/xrd+xml");
logger('fetch_xrd_links: ' . $xml, LOGGER_DATA);
if ((! $xml) || (! stristr($xml,'<xrd')))
return array();
// fix diaspora's bad xml
$xml = str_replace(array('href=&quot;','&quot;/>'),array('href="','"/>'),$xml);
$h = parse_xml_string($xml);
if(! $h)
return array();
$arr = xml::element_to_array($h);
$links = array();
if(isset($arr['xrd']['link'])) {
$link = $arr['xrd']['link'];
if(! isset($link[0]))
$links = array($link);
else
$links = $link;
}
if(isset($arr['xrd']['alias'])) {
$alias = $arr['xrd']['alias'];
if(! isset($alias[0]))
$aliases = array($alias);
else
$aliases = $alias;
if(is_array($aliases) && count($aliases)) {
foreach($aliases as $alias) {
$links[]['@attributes'] = array('rel' => 'alias' , 'href' => $alias);
}
}
}
logger('fetch_xrd_links: ' . print_r($links,true), LOGGER_DATA);
return $links;
}
/**
* @brief Check URL to se if ts's real
*
@ -1123,7 +805,7 @@ function short_link($url) {
* This function encodes an array to json format
* and adds an application/json HTTP header to the output.
* After finishing the process is getting killed.
*
*
* @param array $x The input content
*/
function json_return_and_die($x) {
@ -1131,3 +813,57 @@ function json_return_and_die($x) {
echo json_encode($x);
killme();
}
/**
* @brief Find the matching part between two url
*
* @param string $url1
* @param string $url2
* @return string The matching part
*/
function matching_url($url1, $url2) {
if (($url1 == "") OR ($url2 == ""))
return "";
$url1 = normalise_link($url1);
$url2 = normalise_link($url2);
$parts1 = parse_url($url1);
$parts2 = parse_url($url2);
if (!isset($parts1["host"]) OR !isset($parts2["host"]))
return "";
if ($parts1["scheme"] != $parts2["scheme"])
return "";
if ($parts1["host"] != $parts2["host"])
return "";
if ($parts1["port"] != $parts2["port"])
return "";
$match = $parts1["scheme"]."://".$parts1["host"];
if ($parts1["port"])
$match .= ":".$parts1["port"];
$pathparts1 = explode("/", $parts1["path"]);
$pathparts2 = explode("/", $parts2["path"]);
$i = 0;
$path = "";
do {
$path1 = $pathparts1[$i];
$path2 = $pathparts2[$i];
if ($path1 == $path2)
$path .= $path1."/";
} while (($path1 == $path2) AND ($i++ <= count($pathparts1)));
$match .= $path;
return normalise_link($match);
}

5
include/salmon.php
View file

@ -1,15 +1,14 @@
<?php
require_once('include/crypto.php');
require_once('include/Probe.php');
function get_salmon_key($uri,$keyhash) {
$ret = array();
logger('Fetching salmon key for '.$uri);
$arr = lrdd($uri);
$arr = Probe::lrdd($uri);
if(is_array($arr)) {
foreach($arr as $a) {

1
include/socgraph.php
View file

@ -9,6 +9,7 @@
require_once('include/datetime.php');
require_once("include/Scrape.php");
require_once("include/network.php");
require_once("include/html2bbcode.php");
require_once("include/Contact.php");
require_once("include/Photo.php");

3
mod/dfrn_confirm.php
View file

@ -16,6 +16,7 @@
require_once('include/enotify.php');
require_once('include/group.php');
require_once('include/Probe.php');
function dfrn_confirm_post(&$a,$handsfree = null) {
@ -356,7 +357,7 @@ function dfrn_confirm_post(&$a,$handsfree = null) {
$poll = (($contact['poll']) ? $contact['poll'] : '');
if((! $contact['notify']) || (! $contact['poll'])) {
$arr = lrdd($contact['url']);
$arr = Probe::lrdd($contact['url']);
if(count($arr)) {
foreach($arr as $link) {
if($link['@attributes']['rel'] === 'salmon')

11
mod/dfrn_request.php
View file

@ -11,6 +11,7 @@
require_once('include/enotify.php');
require_once('include/Scrape.php');
require_once('include/Probe.php');
require_once('include/group.php');
if(! function_exists('dfrn_request_init')) {
@ -116,7 +117,7 @@ function dfrn_request_post(&$a) {
* Scrape the other site's profile page to pick up the dfrn links, key, fn, and photo
*/
$parms = scrape_dfrn($dfrn_url);
$parms = Probe::profile($dfrn_url);
if(! count($parms)) {
notice( t('Profile location is not valid or does not contain profile information.') . EOL );
@ -127,7 +128,7 @@ function dfrn_request_post(&$a) {
notice( t('Warning: profile location has no identifiable owner name.') . EOL );
if(! x($parms,'photo'))
notice( t('Warning: profile location has no profile photo.') . EOL );
$invalid = validate_dfrn($parms);
$invalid = Probe::valid_dfrn($parms);
if($invalid) {
notice( sprintf( tt("%d required parameter was not found at the given location",
"%d required parameters were not found at the given location",
@ -447,7 +448,7 @@ function dfrn_request_post(&$a) {
$network = $data["network"];
// Canonicalise email-style profile locator
$url = webfinger_dfrn($url,$hcard);
$url = Probe::webfinger_dfrn($url,$hcard);
if (substr($url,0,5) === 'stat:') {
@ -511,7 +512,7 @@ function dfrn_request_post(&$a) {
require_once('include/Scrape.php');
$parms = scrape_dfrn(($hcard) ? $hcard : $url);
$parms = Probe::profile(($hcard) ? $hcard : $url);
if(! count($parms)) {
notice( t('Profile location is not valid or does not contain profile information.') . EOL );
@ -522,7 +523,7 @@ function dfrn_request_post(&$a) {
notice( t('Warning: profile location has no identifiable owner name.') . EOL );
if(! x($parms,'photo'))
notice( t('Warning: profile location has no profile photo.') . EOL );
$invalid = validate_dfrn($parms);
$invalid = Probe::valid_dfrn($parms);
if($invalid) {
notice( sprintf( tt("%d required parameter was not found at the given location",
"%d required parameters were not found at the given location",

3
mod/photos.php
View file

@ -8,6 +8,7 @@ require_once('include/security.php');
require_once('include/redir.php');
require_once('include/tags.php');
require_once('include/threads.php');
require_once('include/Probe.php');
function photos_init(&$a) {
@ -552,7 +553,7 @@ function photos_post(&$a) {
$name = substr($tag,1);
if((strpos($name,'@')) || (strpos($name,'http://'))) {
$newname = $name;
$links = @lrdd($name);
$links = @Probe::lrdd($name);
if(count($links)) {
foreach($links as $link) {
if($link['@attributes']['rel'] === 'http://webfinger.net/rel/profile-page')

3
mod/profiles.php
View file

@ -1,5 +1,6 @@
<?php
require_once("include/Contact.php");
require_once('include/Probe.php');
function profiles_init(&$a) {
@ -251,7 +252,7 @@ function profiles_post(&$a) {
$lookup = str_replace('_',' ', $lookup);
if(strpos($lookup,'@') || (strpos($lookup,'http://'))) {
$newname = $lookup;
$links = @lrdd($lookup);
$links = @Probe::lrdd($lookup);
if(count($links)) {
foreach($links as $link) {
if($link['@attributes']['rel'] === 'http://webfinger.net/rel/profile-page') {

10
mod/webfinger.php
View file

@ -1,6 +1,5 @@
<?php
require_once("include/Probe.php");
function webfinger_content(&$a) {
@ -8,16 +7,13 @@ function webfinger_content(&$a) {
$o .= '<form action="webfinger" method="get">';
$o .= 'Lookup address: <input type="text" style="width: 250px;" name="addr" value="' . $_GET['addr'] .'" />';
$o .= '<input type="submit" name="submit" value="Submit" /></form>';
$o .= '<input type="submit" name="submit" value="Submit" /></form>';
$o .= '<br /><br />';
if(x($_GET,'addr')) {
$addr = trim($_GET['addr']);
if(strpos($addr,'@' !== false))
$res = webfinger($addr);
else
$res = lrdd($addr);
$res = Probe::lrdd($addr);
$o .= '<pre>';
$o .= str_replace("\n",'<br />',print_r($res,true));
$o .= '</pre>';