relocated "follow" url scraping code so it can be used elsewhere (contact repair, etc.)

This commit is contained in:
Friendika 2011-04-15 00:59:00 -07:00
parent d21144d0a3
commit 13eaa802e4
2 changed files with 215 additions and 183 deletions

View File

@ -260,4 +260,194 @@ function scrape_feed($url) {
} }
return $ret; return $ret;
}} }}
function probe_url($url) {
$result = array();
if(! $url)
return $result;
$diaspora = false;
$email_conversant = false;
if($url) {
$links = lrdd($url);
if(count($links)) {
foreach($links as $link) {
if($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://joindiaspora.com/seed_location')
$diaspora = true;
}
// Status.Net can have more than one profile URL. We need to match the profile URL
// to a contact on incoming messages to prevent spam, and we won't know which one
// to match. So in case of two, one of them is stored as an alias. Only store URL's
// and not webfinger user@host aliases. If they've got more than two non-email style
// aliases, let's hope we're lucky and get one that matches the feed author-uri because
// otherwise we're screwed.
foreach($links as $link) {
if($link['@attributes']['rel'] === 'alias') {
if(strpos($link['@attributes']['href'],'@') === false) {
if(isset($profile)) {
if($link['@attributes']['href'] !== $profile)
$alias = unamp($link['@attributes']['href']);
}
else
$profile = unamp($link['@attributes']['href']);
}
}
}
}
else {
if((strpos($orig_url,'@')) && validate_email($orig_url)) {
$email_conversant = true;
}
}
}
if(strlen($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 = $ret['dfrn-poll'];
}
}
if($network !== NETWORK_DFRN) {
$network = NETWORK_OSTATUS;
$priority = 0;
if($hcard) {
$vcard = scrape_vcard($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(! $profile) {
if($diaspora)
$profile = $hcard;
else
$profile = $url;
}
if(! x($vcard,'fn'))
if(x($vcard,'nick'))
$vcard['fn'] = $vcard['nick'];
if((! isset($vcard)) && (! $poll)) {
$ret = scrape_feed($url);
logger('probe_url: scrape_feed returns: ' . print_r($ret,true), LOGGER_DATA);
if(count($ret) && ($ret['feed_atom'] || $ret['feed_rss'])) {
$poll = ((x($ret,'feed_atom')) ? unamp($ret['feed_atom']) : unamp($ret['feed_rss']));
$vcard = array();
if(x($ret,'photo'))
$vcard['photo'] = $ret['photo'];
require_once('simplepie/simplepie.inc');
$feed = new SimplePie();
$xml = fetch_url($poll);
$feed->set_raw_data($xml);
$feed->init();
if(! x($vcard,'photo'))
$vcard['photo'] = $feed->get_image_url();
$author = $feed->get_author();
if($author) {
$vcard['fn'] = unxmlify(trim($author->get_name()));
if(! $vcard['fn'])
$vcard['fn'] = trim(unxmlify($author->get_email()));
if(strpos($vcard['fn'],'@') !== false)
$vcard['fn'] = substr($vcard['fn'],0,strpos($vcard['fn'],'@'));
$vcard['nick'] = strtolower(notags(unxmlify($vcard['fn'])));
if(strpos($vcard['nick'],' '))
$vcard['nick'] = trim(substr($vcard['nick'],0,strpos($vcard['nick'],' ')));
$email = unxmlify($author->get_email());
}
else {
$item = $feed->get_item(0);
if($item) {
$author = $item->get_author();
if($author) {
$vcard['fn'] = trim(unxmlify($author->get_name()));
if(! $vcard['fn'])
$vcard['fn'] = trim(unxmlify($author->get_email()));
if(strpos($vcard['fn'],'@') !== false)
$vcard['fn'] = substr($vcard['fn'],0,strpos($vcard['fn'],'@'));
$vcard['nick'] = strtolower(unxmlify($vcard['fn']));
if(strpos($vcard['nick'],' '))
$vcard['nick'] = trim(substr($vcard['nick'],0,strpos($vcard['nick'],' ')));
$email = unxmlify($author->get_email());
}
if(! $vcard['photo']) {
$rawmedia = $item->get_item_tags('http://search.yahoo.com/mrss/','thumbnail');
if($rawmedia && $rawmedia[0]['attribs']['']['url'])
$vcard['photo'] = unxmlify($rawmedia[0]['attribs']['']['url']);
}
}
}
if((! $vcard['photo']) && strlen($email))
$vcard['photo'] = gravatar_img($email);
if($poll === $profile)
$lnk = $feed->get_permalink();
if(isset($lnk) && strlen($lnk))
$profile = $lnk;
if(! (x($vcard,'fn')))
$vcard['fn'] = notags($feed->get_title());
if(! (x($vcard,'fn')))
$vcard['fn'] = notags($feed->get_description());
$network = 'feed';
$priority = 2;
}
}
}
if(! x($vcard,'photo')) {
$a = get_app();
$vcard['photo'] = $a->get_baseurl() . '/images/default-profile.jpg' ;
}
$vcard['fn'] = notags($vcard['fn']);
$vcard['nick'] = notags($vcard['nick']);
$result['name'] = $vcard['fn'];
$result['nick'] = $vcard['nick'];
$result['url'] = $profile;
$result['notify'] = $notify;
$result['poll'] = $poll;
$result['request'] = $request;
$result['confirm'] = $confirm;
$result['photo'] = $vcard['photo'];
$result['priority'] = $priority;
$result['network'] = $network;
$result['alias'] = $alias;
logger('probe_url: ' . print_r($result,true), LOGGER_DEBUG);
return $result;
}

View File

@ -11,192 +11,34 @@ function follow_post(&$a) {
} }
$url = $orig_url = notags(trim($_POST['url'])); $url = $orig_url = notags(trim($_POST['url']));
$diaspora = false;
$email_conversant = false;
if($url) { $ret = probe_url($url);
$links = lrdd($url);
if(count($links)) {
foreach($links as $link) {
if($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://joindiaspora.com/seed_location')
$diaspora = true;
} if($ret['network'] === NETWORK_DFRN) {
if(strlen($a->path))
// Status.Net can have more than one profile URL. We need to match the profile URL $myaddr = bin2hex($a->get_baseurl() . '/profile/' . $a->user['nickname']);
// to a contact on incoming messages to prevent spam, and we won't know which one
// to match. So in case of two, one of them is stored as an alias. Only store URL's
// and not webfinger user@host aliases. If they've got more than two non-email style
// aliases, let's hope we're lucky and get one that matches the feed author-uri because
// otherwise we're screwed.
foreach($links as $link) {
if($link['@attributes']['rel'] === 'alias') {
if(strpos($link['@attributes']['href'],'@') === false) {
if(isset($profile)) {
if($link['@attributes']['href'] !== $profile)
$alias = unamp($link['@attributes']['href']);
}
else
$profile = unamp($link['@attributes']['href']);
}
}
}
}
else {
if((strpos($orig_url,'@')) && validate_email($orig_url)) {
$email_conversant = true;
}
}
}
// If we find a DFRN site, send our subscriber to the other person's
// dfrn_request page and all the other details will get sorted.
if(strlen($dfrn)) {
$ret = scrape_dfrn($dfrn);
if(is_array($ret) && x($ret,'dfrn-request')) {
if(strlen($a->path))
$myaddr = bin2hex($a->get_baseurl() . '/profile/' . $a->user['nickname']);
else
$myaddr = bin2hex($a->user['nickname'] . '@' . $a->get_hostname());
goaway($ret['dfrn-request'] . "&addr=$myaddr");
// NOTREACHED
}
}
$network = 'stat';
$priority = 0;
if($hcard) {
$vcard = scrape_vcard($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'] . $vcard['photo'];
}
logger('mod_follow: scrape_vcard: ' . print_r($vcard,true), LOGGER_DATA);
}
if(! $profile) {
if($diaspora)
$profile = $hcard;
else else
$profile = $url; $myaddr = bin2hex($a->user['nickname'] . '@' . $a->get_hostname());
goaway($ret['request'] . "&addr=$myaddr");
// NOTREACHED
} }
if(! x($vcard,'fn'))
if(x($vcard,'nick'))
$vcard['fn'] = $vcard['nick'];
if((! isset($vcard)) && (! $poll)) {
$ret = scrape_feed($url);
logger('mod_follow: scrape_feed returns: ' . print_r($ret,true), LOGGER_DATA);
if(count($ret) && ($ret['feed_atom'] || $ret['feed_rss'])) {
$poll = ((x($ret,'feed_atom')) ? unamp($ret['feed_atom']) : unamp($ret['feed_rss']));
$vcard = array();
if(x($ret,'photo'))
$vcard['photo'] = $ret['photo'];
require_once('simplepie/simplepie.inc');
$feed = new SimplePie();
$xml = fetch_url($poll);
$feed->set_raw_data($xml);
$feed->init();
if(! x($vcard,'photo'))
$vcard['photo'] = $feed->get_image_url();
$author = $feed->get_author();
if($author) {
$vcard['fn'] = unxmlify(trim($author->get_name()));
if(! $vcard['fn'])
$vcard['fn'] = trim(unxmlify($author->get_email()));
if(strpos($vcard['fn'],'@') !== false)
$vcard['fn'] = substr($vcard['fn'],0,strpos($vcard['fn'],'@'));
$vcard['nick'] = strtolower(notags(unxmlify($vcard['fn'])));
if(strpos($vcard['nick'],' '))
$vcard['nick'] = trim(substr($vcard['nick'],0,strpos($vcard['nick'],' ')));
$email = unxmlify($author->get_email());
}
else {
$item = $feed->get_item(0);
if($item) {
$author = $item->get_author();
if($author) {
$vcard['fn'] = trim(unxmlify($author->get_name()));
if(! $vcard['fn'])
$vcard['fn'] = trim(unxmlify($author->get_email()));
if(strpos($vcard['fn'],'@') !== false)
$vcard['fn'] = substr($vcard['fn'],0,strpos($vcard['fn'],'@'));
$vcard['nick'] = strtolower(unxmlify($vcard['fn']));
if(strpos($vcard['nick'],' '))
$vcard['nick'] = trim(substr($vcard['nick'],0,strpos($vcard['nick'],' ')));
$email = unxmlify($author->get_email());
}
if(! $vcard['photo']) {
$rawmedia = $item->get_item_tags('http://search.yahoo.com/mrss/','thumbnail');
if($rawmedia && $rawmedia[0]['attribs']['']['url'])
$vcard['photo'] = unxmlify($rawmedia[0]['attribs']['']['url']);
}
}
}
if((! $vcard['photo']) && strlen($email))
$vcard['photo'] = gravatar_img($email);
if($poll === $profile)
$lnk = $feed->get_permalink();
if(isset($lnk) && strlen($lnk))
$profile = $lnk;
if(! (x($vcard,'fn')))
$vcard['fn'] = notags($feed->get_title());
if(! (x($vcard,'fn')))
$vcard['fn'] = notags($feed->get_description());
$network = 'feed';
$priority = 2;
}
}
logger('follow: poll=' . $poll . ' notify=' . $notify . ' profile=' . $profile . ' vcard=' . print_r($vcard,true));
$vcard['fn'] = notags($vcard['fn']);
$vcard['nick'] = notags($vcard['nick']);
// do we have enough information? // do we have enough information?
if(! ((x($vcard['fn'])) && ($poll) && ($profile))) { if(! ((x($ret,'name')) && (x($ret,'poll')) && (x($ret,'url')))) {
notice( t('The profile address specified does not provide adequate information.') . EOL); notice( t('The profile address specified does not provide adequate information.') . EOL);
goaway($_SESSION['return_url']); goaway($_SESSION['return_url']);
} }
if(! $notify) { if(! $ret['notify']) {
notice( t('Limited profile. This person will be unable to receive direct/personal notifications from you.') . EOL); notice( t('Limited profile. This person will be unable to receive direct/personal notifications from you.') . EOL);
} }
if(! x($vcard,'photo')) $writeable = ((($ret['network'] === NETWORK_OSTATUS) && ($ret['notify'])) ? 1 : 0);
$vcard['photo'] = $a->get_baseurl() . '/images/default-profile.jpg' ;
$writeable = ((($network === 'stat') && ($notify)) ? 1 : 0);
// check if we already have a contact // check if we already have a contact
// the poll url is more reliable than the profile url, as we may have // the poll url is more reliable than the profile url, as we may have
@ -204,7 +46,7 @@ function follow_post(&$a) {
$r = q("SELECT * FROM `contact` WHERE `uid` = %d AND `poll` = '%s' LIMIT 1", $r = q("SELECT * FROM `contact` WHERE `uid` = %d AND `poll` = '%s' LIMIT 1",
intval(local_user()), intval(local_user()),
dbesc($poll) dbesc($ret['poll'])
); );
if(count($r)) { if(count($r)) {
@ -224,22 +66,22 @@ function follow_post(&$a) {
VALUES ( %d, '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', %d, %d, %d, 0, 0, 0 ) ", VALUES ( %d, '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', %d, %d, %d, 0, 0, 0 ) ",
intval(local_user()), intval(local_user()),
dbesc(datetime_convert()), dbesc(datetime_convert()),
dbesc($profile), dbesc($ret['url']),
dbesc($alias), dbesc($ret['alias']),
dbesc($notify), dbesc($ret['notify']),
dbesc($poll), dbesc($ret['poll']),
dbesc($vcard['fn']), dbesc($ret['name']),
dbesc($vcard['nick']), dbesc($ret['nick']),
dbesc($vcard['photo']), dbesc($ret['photo']),
dbesc($network), dbesc($ret['network']),
intval(REL_FAN), intval(REL_FAN),
intval($priority), intval($ret['priority']),
intval($writable) intval($writable)
); );
} }
$r = q("SELECT * FROM `contact` WHERE `url` = '%s' AND `uid` = %d LIMIT 1", $r = q("SELECT * FROM `contact` WHERE `url` = '%s' AND `uid` = %d LIMIT 1",
dbesc($profile), dbesc($ret['url']),
intval(local_user()) intval(local_user())
); );
@ -254,7 +96,7 @@ function follow_post(&$a) {
require_once("Photo.php"); require_once("Photo.php");
$photos = import_profile_photo($vcard['photo'],local_user(),$contact_id); $photos = import_profile_photo($ret['photo'],local_user(),$contact_id);
$r = q("UPDATE `contact` SET `photo` = '%s', $r = q("UPDATE `contact` SET `photo` = '%s',
`thumb` = '%s', `thumb` = '%s',