function probe_url($url, $mode = PROBE_NORMAL)
{
require_once 'include/email.php';
$result = array();
if (!$url) {
return $result;
}
$network = null;
$diaspora = false;
$diaspora_base = '';
$diaspora_guid = '';
$diaspora_key = '';
$has_lrdd = false;
$email_conversant = false;
$twitter = strpos($url, 'twitter.com') !== false ? true : false;
$at_addr = strpos($url, '@') !== false ? true : false;
if (!$twitter) {
if (strpos($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) {
$zot = unamp($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']));
$pubkey = rsatopem($diaspora_key);
$diaspora = true;
}
}
// Status.Net can have more than one profile URL. We need to match the profile URL
// to a contact on incoming messages to prevent spam, and we won't know which one
// to match. So in case of two, one of them is stored as an alias. Only store URL's
// and not webfinger user@host aliases. If they've got more than two non-email style
// aliases, let's hope we're lucky and get one that matches the feed author-uri because
// otherwise we're screwed.
foreach ($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']);
}
}
}
}
} 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->from, $orig_url)) {
$adr = imap_rfc822_parse_adrlist($x->from, '');
} elseif (stristr($x->to, $orig_url)) {
$adr = imap_rfc822_parse_adrlist($x->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 : $dfrn);
if (is_array($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 && $network !== NETWORK_MAIL) {
if ($diaspora) {
$network = NETWORK_DIASPORA;
} elseif ($has_lrdd) {
$network = NETWORK_OSTATUS;
}
$priority = 0;
if ($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 ($twitter) {
logger('twitter: setup');
$tid = basename($url);
$tapi = 'https://api.twitter.com/1/statuses/user_timeline.rss';
if (intval($tid)) {
$poll = $tapi . '?user_id=' . $tid;
} else {
$poll = $tapi . '?screen_name=' . $tid;
}
$profile = 'http://twitter.com/#!/' . $tid;
$vcard['photo'] = 'https://api.twitter.com/1/users/profile_image/' . $tid;
$vcard['nick'] = $tid;
$vcard['fn'] = $tid . '@twitter';
}
if (!x($vcard, 'fn')) {
if (x($vcard, 'nick')) {
$vcard['fn'] = $vcard['nick'];
}
}
$check_feed = false;
if ($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) {
$feedret = scrape_feed($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'];
}
require_once 'library/simplepie/simplepie.inc';
$feed = new SimplePie();
$xml = fetch_url($poll);
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);
$feed->set_raw_data($xml);
$feed->init();
if ($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']) {
$vcard['fn'] = trim(unxmlify($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()));
}
if (!$vcard['photo']) {
$rawtags = $feed->get_feed_tags(SIMPLEPIE_NAMESPACE_ATOM_10, 'author');
if ($rawtags) {
$elems = $rawtags[0]['child'][SIMPLEPIE_NAMESPACE_ATOM_10];
if (x($elems, 'link') && $elems['link'][0]['attribs']['']['rel'] === 'photo') {
$vcard['photo'] = $elems['link'][0]['attribs']['']['href'];
}
}
}
} 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'], '@'));
}
$email = unxmlify($author->get_email());
if (!$profile && $author->get_link()) {
$profile = trim(unxmlify($author->get_link()));
}
}
if (!$vcard['photo']) {
$rawmedia = $item->get_item_tags('http://search.yahoo.com/mrss/', 'thumbnail');
if ($rawmedia && $rawmedia[0]['attribs']['']['url']) {
$vcard['photo'] = unxmlify($rawmedia[0]['attribs']['']['url']);
}
}
if (!$vcard['photo']) {
$rawtags = $item->get_item_tags(SIMPLEPIE_NAMESPACE_ATOM_10, 'author');
if ($rawtags) {
$elems = $rawtags[0]['child'][SIMPLEPIE_NAMESPACE_ATOM_10];
if (x($elems, 'link') && $elems['link'][0]['attribs']['']['rel'] === 'photo') {
$vcard['photo'] = $elems['link'][0]['attribs']['']['href'];
}
}
}
}
}
if (!$vcard['photo'] && strlen($email)) {
$vcard['photo'] = avatar_img($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());
}
if (strpos($vcard['fn'], 'Twitter / ') !== false) {
$vcard['fn'] = substr($vcard['fn'], strpos($vcard['fn'], '/') + 1);
$vcard['fn'] = trim($vcard['fn']);
}
if (!x($vcard, 'nick')) {
$vcard['nick'] = strtolower(notags(unxmlify($vcard['fn'])));
if (strpos($vcard['nick'], ' ')) {
$vcard['nick'] = trim(substr($vcard['nick'], 0, strpos($vcard['nick'], ' ')));
}
}
if (!$network) {
$network = NETWORK_FEED;
}
if (!$priority) {
$priority = 2;
}
}
}
if (!x($vcard, 'photo')) {
$a = get_app();
$vcard['photo'] = $a->get_baseurl() . '/images/person-175.jpg';
}
if (!$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;
}
$vcard['fn'] = notags($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['pubkey'] = $pubkey;
logger('probe_url: ' . print_r($result, true), LOGGER_DEBUG);
return $result;
}
function onepoll_run(&$argv, &$argc)
{
global $a, $db;
if (is_null($a)) {
$a = new App();
}
if (is_null($db)) {
@(include ".htconfig.php");
require_once "include/dba.php";
$db = new dba($db_host, $db_user, $db_pass, $db_data);
unset($db_host, $db_user, $db_pass, $db_data);
}
require_once 'include/session.php';
require_once 'include/datetime.php';
require_once 'library/simplepie/simplepie.inc';
require_once 'include/items.php';
require_once 'include/Contact.php';
require_once 'include/email.php';
require_once 'include/socgraph.php';
require_once 'include/pidfile.php';
require_once 'include/queue_fn.php';
load_config('config');
load_config('system');
$a->set_baseurl(get_config('system', 'url'));
load_hooks();
logger('onepoll: start');
$manual_id = 0;
$generation = 0;
$hub_update = false;
$force = false;
$restart = false;
if ($argc > 1 && intval($argv[1])) {
$contact_id = intval($argv[1]);
}
if (!$contact_id) {
logger('onepoll: no contact');
return;
}
$lockpath = get_lockpath();
if ($lockpath != '') {
$pidfile = new pidfile($lockpath, 'onepoll' . $contact_id);
if ($pidfile->is_already_running()) {
logger("onepoll: Already running for contact " . $contact_id);
if ($pidfile->running_time() > 9 * 60) {
$pidfile->kill();
logger("killed stale process");
}
exit;
}
}
$d = datetime_convert();
// Only poll from those with suitable relationships,
// and which have a polling address and ignore Diaspora since
// we are unable to match those posts with a Diaspora GUID and prevent duplicates.
$contacts = q("SELECT `contact`.* FROM `contact`\n\t\tWHERE ( `rel` = %d OR `rel` = %d ) AND `poll` != ''\n\t\tAND NOT `network` IN ( '%s', '%s', '%s' )\n\t\tAND `contact`.`id` = %d\n\t\tAND `self` = 0 AND `contact`.`blocked` = 0 AND `contact`.`readonly` = 0\n\t\tAND `contact`.`archive` = 0 LIMIT 1", intval(CONTACT_IS_SHARING), intval(CONTACT_IS_FRIEND), dbesc(NETWORK_DIASPORA), dbesc(NETWORK_FACEBOOK), dbesc(NETWORK_PUMPIO), intval($contact_id));
if (!count($contacts)) {
return;
}
$contact = $contacts[0];
$xml = false;
$t = $contact['last-update'];
if ($contact['subhub']) {
$poll_interval = get_config('system', 'pushpoll_frequency');
$contact['priority'] = $poll_interval !== false ? intval($poll_interval) : 3;
$hub_update = false;
if (datetime_convert('UTC', 'UTC', 'now') > datetime_convert('UTC', 'UTC', $t . " + 1 day")) {
$hub_update = true;
}
} else {
$hub_update = false;
}
$importer_uid = $contact['uid'];
$r = q("SELECT `contact`.*, `user`.`page-flags` FROM `contact` INNER JOIN `user` on `contact`.`uid` = `user`.`uid` WHERE `user`.`uid` = %d AND `contact`.`self` = 1 LIMIT 1", intval($importer_uid));
if (!count($r)) {
return;
}
$importer = $r[0];
logger("onepoll: poll: ({$contact['id']}) IMPORTER: {$importer['name']}, CONTACT: {$contact['name']}");
$last_update = $contact['last-update'] === '0000-00-00 00:00:00' ? datetime_convert('UTC', 'UTC', 'now - 7 days', ATOM_TIME) : datetime_convert('UTC', 'UTC', $contact['last-update'], ATOM_TIME);
if ($contact['network'] === NETWORK_DFRN) {
$idtosend = $orig_id = $contact['dfrn-id'] ? $contact['dfrn-id'] : $contact['issued-id'];
if (intval($contact['duplex']) && $contact['dfrn-id']) {
$idtosend = '0:' . $orig_id;
}
if (intval($contact['duplex']) && $contact['issued-id']) {
$idtosend = '1:' . $orig_id;
}
// they have permission to write to us. We already filtered this in the contact query.
$perm = 'rw';
// But this may be our first communication, so set the writable flag if it isn't set already.
if (!intval($contact['writable'])) {
q("update contact set writable = 1 where id = %d", intval($contact['id']));
}
$url = $contact['poll'] . '?dfrn_id=' . $idtosend . '&dfrn_version=' . DFRN_PROTOCOL_VERSION . '&type=data&last_update=' . $last_update . '&perm=' . $perm;
$handshake_xml = fetch_url($url);
$html_code = $a->get_curl_code();
logger('onepoll: handshake with url ' . $url . ' returns xml: ' . $handshake_xml, LOGGER_DATA);
if (!strlen($handshake_xml) || $html_code >= 400 || !$html_code) {
logger("poller: {$url} appears to be dead - marking for death ");
// dead connection - might be a transient event, or this might
// mean the software was uninstalled or the domain expired.
// Will keep trying for one month.
mark_for_death($contact);
// set the last-update so we don't keep polling
$r = q("UPDATE `contact` SET `last-update` = '%s' WHERE `id` = %d", dbesc(datetime_convert()), intval($contact['id']));
return;
}
if (!strstr($handshake_xml, '<')) {
logger('poller: response from ' . $url . ' did not contain XML.');
mark_for_death($contact);
$r = q("UPDATE `contact` SET `last-update` = '%s' WHERE `id` = %d", dbesc(datetime_convert()), intval($contact['id']));
return;
}
$res = parse_xml_string($handshake_xml);
if (intval($res->status) == 1) {
logger("poller: {$url} replied status 1 - marking for death ");
// we may not be friends anymore. Will keep trying for one month.
// set the last-update so we don't keep polling
$r = q("UPDATE `contact` SET `last-update` = '%s' WHERE `id` = %d", dbesc(datetime_convert()), intval($contact['id']));
mark_for_death($contact);
} else {
if ($contact['term-date'] != '0000-00-00 00:00:00') {
logger("poller: {$url} back from the dead - removing mark for death");
unmark_for_death($contact);
}
}
if (intval($res->status) != 0 || !strlen($res->challenge) || !strlen($res->dfrn_id)) {
return;
}
if ((double) $res->dfrn_version > 2.21 && $contact['poco'] == '') {
q("update contact set poco = '%s' where id = %d", dbesc(str_replace('/profile/', '/poco/', $contact['url'])), intval($contact['id']));
}
$postvars = array();
$sent_dfrn_id = hex2bin((string) $res->dfrn_id);
$challenge = hex2bin((string) $res->challenge);
$final_dfrn_id = '';
if ($contact['duplex'] && strlen($contact['prvkey'])) {
openssl_private_decrypt($sent_dfrn_id, $final_dfrn_id, $contact['prvkey']);
openssl_private_decrypt($challenge, $postvars['challenge'], $contact['prvkey']);
} else {
openssl_public_decrypt($sent_dfrn_id, $final_dfrn_id, $contact['pubkey']);
openssl_public_decrypt($challenge, $postvars['challenge'], $contact['pubkey']);
}
$final_dfrn_id = substr($final_dfrn_id, 0, strpos($final_dfrn_id, '.'));
if (strpos($final_dfrn_id, ':') == 1) {
$final_dfrn_id = substr($final_dfrn_id, 2);
}
if ($final_dfrn_id != $orig_id) {
logger('poller: ID did not decode: ' . $contact['id'] . ' orig: ' . $orig_id . ' final: ' . $final_dfrn_id);
// did not decode properly - cannot trust this site
return;
}
$postvars['dfrn_id'] = $idtosend;
$postvars['dfrn_version'] = DFRN_PROTOCOL_VERSION;
$postvars['perm'] = 'rw';
$xml = post_url($contact['poll'], $postvars);
} elseif ($contact['network'] === NETWORK_OSTATUS || $contact['network'] === NETWORK_DIASPORA || $contact['network'] === NETWORK_FEED) {
// Upgrading DB fields from an older Friendica version
// Will only do this once per notify-enabled OStatus contact
// or if relationship changes
$stat_writeable = $contact['notify'] && ($contact['rel'] == CONTACT_IS_FOLLOWER || $contact['rel'] == CONTACT_IS_FRIEND) ? 1 : 0;
if ($contact['network'] === NETWORK_OSTATUS && get_pconfig($importer_uid, 'system', 'ostatus_autofriend')) {
$stat_writeable = 1;
}
if ($stat_writeable != $contact['writable']) {
q("UPDATE `contact` SET `writable` = %d WHERE `id` = %d", intval($stat_writeable), intval($contact['id']));
}
// Are we allowed to import from this person?
if ($contact['rel'] == CONTACT_IS_FOLLOWER || $contact['blocked'] || $contact['readonly']) {
return;
}
$xml = fetch_url($contact['poll']);
} elseif ($contact['network'] === NETWORK_MAIL || $contact['network'] === NETWORK_MAIL2) {
logger("Mail: Fetching", LOGGER_DEBUG);
$mail_disabled = function_exists('imap_open') && !get_config('system', 'imap_disabled') ? 0 : 1;
if ($mail_disabled) {
return;
}
logger("Mail: Enabled", LOGGER_DEBUG);
$mbox = null;
$x = q("SELECT `prvkey` FROM `user` WHERE `uid` = %d LIMIT 1", intval($importer_uid));
$mailconf = q("SELECT * FROM `mailacct` WHERE `server` != '' AND `uid` = %d LIMIT 1", intval($importer_uid));
if (count($x) && count($mailconf)) {
$mailbox = construct_mailbox_name($mailconf[0]);
$password = '';
openssl_private_decrypt(hex2bin($mailconf[0]['pass']), $password, $x[0]['prvkey']);
$mbox = email_connect($mailbox, $mailconf[0]['user'], $password);
unset($password);
logger("Mail: Connect to " . $mailconf[0]['user']);
if ($mbox) {
q("UPDATE `mailacct` SET `last_check` = '%s' WHERE `id` = %d AND `uid` = %d", dbesc(datetime_convert()), intval($mailconf[0]['id']), intval($importer_uid));
logger("Mail: Connected to " . $mailconf[0]['user']);
} else {
logger("Mail: Connection error " . $mailconf[0]['user'] . " " . print_r(imap_errors()));
}
}
if ($mbox) {
$msgs = email_poll($mbox, $contact['addr']);
if (count($msgs)) {
logger("Mail: Parsing " . count($msgs) . " mails for " . $mailconf[0]['user'], LOGGER_DEBUG);
$metas = email_msg_meta($mbox, implode(',', $msgs));
if (count($metas) != count($msgs)) {
logger("onepoll: for " . $mailconf[0]['user'] . " there are " . count($msgs) . " messages but received " . count($metas) . " metas", LOGGER_DEBUG);
} else {
$msgs = array_combine($msgs, $metas);
foreach ($msgs as $msg_uid => $meta) {
logger("Mail: Parsing mail " . $msg_uid, LOGGER_DATA);
$datarray = array();
$datarray['verb'] = ACTIVITY_POST;
$datarray['object-type'] = ACTIVITY_OBJ_NOTE;
// $meta = email_msg_meta($mbox,$msg_uid);
// $headers = email_msg_headers($mbox,$msg_uid);
$datarray['uri'] = msgid2iri(trim($meta->message_id, '<>'));
// Have we seen it before?
$r = q("SELECT * FROM `item` WHERE `uid` = %d AND `uri` = '%s' LIMIT 1", intval($importer_uid), dbesc($datarray['uri']));
if (count($r)) {
logger("Mail: Seen before " . $msg_uid . " for " . $mailconf[0]['user'] . " UID: " . $importer_uid . " URI: " . $datarray['uri'], LOGGER_DEBUG);
// Only delete when mails aren't automatically moved or deleted
if ($mailconf[0]['action'] != 1 and $mailconf[0]['action'] != 3) {
if ($meta->deleted && !$r[0]['deleted']) {
q("UPDATE `item` SET `deleted` = 1, `changed` = '%s' WHERE `id` = %d", dbesc(datetime_convert()), intval($r[0]['id']));
}
}
switch ($mailconf[0]['action']) {
case 0:
logger("Mail: Seen before " . $msg_uid . " for " . $mailconf[0]['user'] . ". Doing nothing.", LOGGER_DEBUG);
break;
case 1:
logger("Mail: Deleting " . $msg_uid . " for " . $mailconf[0]['user']);
imap_delete($mbox, $msg_uid, FT_UID);
break;
case 2:
logger("Mail: Mark as seen " . $msg_uid . " for " . $mailconf[0]['user']);
imap_setflag_full($mbox, $msg_uid, "\\Seen", ST_UID);
break;
case 3:
logger("Mail: Moving " . $msg_uid . " to " . $mailconf[0]['movetofolder'] . " for " . $mailconf[0]['user']);
imap_setflag_full($mbox, $msg_uid, "\\Seen", ST_UID);
if ($mailconf[0]['movetofolder'] != "") {
imap_mail_move($mbox, $msg_uid, $mailconf[0]['movetofolder'], FT_UID);
}
break;
}
continue;
}
// look for a 'references' or an 'in-reply-to' header and try to match with a parent item we have locally.
// $raw_refs = ((x($headers,'references')) ? str_replace("\t",'',$headers['references']) : '');
$raw_refs = property_exists($meta, 'references') ? str_replace("\t", '', $meta->references) : '';
if (!trim($raw_refs)) {
$raw_refs = property_exists($meta, 'in_reply_to') ? str_replace("\t", '', $meta->in_reply_to) : '';
}
$raw_refs = trim($raw_refs);
// Don't allow a blank reference in $refs_arr
if ($raw_refs) {
$refs_arr = explode(' ', $raw_refs);
if (count($refs_arr)) {
for ($x = 0; $x < count($refs_arr); $x++) {
$refs_arr[$x] = "'" . msgid2iri(str_replace(array('<', '>', ' '), array('', '', ''), dbesc($refs_arr[$x]))) . "'";
}
}
$qstr = implode(',', $refs_arr);
$r = q("SELECT `uri` , `parent-uri` FROM `item` WHERE `uri` IN ( {$qstr} ) AND `uid` = %d LIMIT 1", intval($importer_uid));
if (count($r)) {
$datarray['parent-uri'] = $r[0]['parent-uri'];
}
// Set the parent as the top-level item
// $datarray['parent-uri'] = $r[0]['uri'];
}
// Decoding the header
$subject = imap_mime_header_decode($meta->subject);
$datarray['title'] = "";
foreach ($subject as $subpart) {
if ($subpart->charset != "default") {
$datarray['title'] .= iconv($subpart->charset, 'UTF-8//IGNORE', $subpart->text);
} else {
$datarray['title'] .= $subpart->text;
}
}
$datarray['title'] = notags(trim($datarray['title']));
//$datarray['title'] = notags(trim($meta->subject));
$datarray['created'] = datetime_convert('UTC', 'UTC', $meta->date);
// Is it a reply?
$reply = (substr(strtolower($datarray['title']), 0, 3) == "re:" or substr(strtolower($datarray['title']), 0, 3) == "re-" or $raw_refs != "");
// Remove Reply-signs in the subject
$datarray['title'] = RemoveReply($datarray['title']);
// If it seems to be a reply but a header couldn't be found take the last message with matching subject
if (!x($datarray, 'parent-uri') and $reply) {
$r = q("SELECT `uri` , `parent-uri` FROM `item` WHERE `title` = \"%s\" AND `uid` = %d ORDER BY `created` DESC LIMIT 1", dbesc(protect_sprintf($datarray['title'])), intval($importer_uid));
if (count($r)) {
$datarray['parent-uri'] = $r[0]['parent-uri'];
}
}
if (!x($datarray, 'parent-uri')) {
$datarray['parent-uri'] = $datarray['uri'];
}
$r = email_get_msg($mbox, $msg_uid, $reply);
if (!$r) {
logger("Mail: can't fetch msg " . $msg_uid . " for " . $mailconf[0]['user']);
continue;
}
$datarray['body'] = escape_tags($r['body']);
$datarray['body'] = limit_body_size($datarray['body']);
logger("Mail: Importing " . $msg_uid . " for " . $mailconf[0]['user']);
// some mailing lists have the original author as 'from' - add this sender info to msg body.
// todo: adding a gravatar for the original author would be cool
if (!stristr($meta->from, $contact['addr'])) {
$from = imap_mime_header_decode($meta->from);
$fromdecoded = "";
foreach ($from as $frompart) {
if ($frompart->charset != "default") {
$fromdecoded .= iconv($frompart->charset, 'UTF-8//IGNORE', $frompart->text);
} else {
$fromdecoded .= $frompart->text;
}
}
$fromarr = imap_rfc822_parse_adrlist($fromdecoded, $a->get_hostname());
$frommail = $fromarr[0]->mailbox . "@" . $fromarr[0]->host;
if (isset($fromarr[0]->personal)) {
$fromname = $fromarr[0]->personal;
} else {
$fromname = $frommail;
}
//$datarray['body'] = "[b]".t('From: ') . escape_tags($fromdecoded) . "[/b]\n\n" . $datarray['body'];
$datarray['author-name'] = $fromname;
$datarray['author-link'] = "mailto:" . $frommail;
$datarray['author-avatar'] = $contact['photo'];
$datarray['owner-name'] = $contact['name'];
$datarray['owner-link'] = "mailto:" . $contact['addr'];
$datarray['owner-avatar'] = $contact['photo'];
} else {
$datarray['author-name'] = $contact['name'];
$datarray['author-link'] = 'mailbox';
$datarray['author-avatar'] = $contact['photo'];
}
$datarray['uid'] = $importer_uid;
$datarray['contact-id'] = $contact['id'];
if ($datarray['parent-uri'] === $datarray['uri']) {
$datarray['private'] = 1;
}
if ($contact['network'] === NETWORK_MAIL && !get_pconfig($importer_uid, 'system', 'allow_public_email_replies')) {
$datarray['private'] = 1;
$datarray['allow_cid'] = '<' . $contact['id'] . '>';
}
$stored_item = item_store($datarray);
q("UPDATE `item` SET `last-child` = 0 WHERE `parent-uri` = '%s' AND `uid` = %d", dbesc($datarray['parent-uri']), intval($importer_uid));
q("UPDATE `item` SET `last-child` = 1 WHERE `id` = %d", intval($stored_item));
switch ($mailconf[0]['action']) {
case 0:
logger("Mail: Seen before " . $msg_uid . " for " . $mailconf[0]['user'] . ". Doing nothing.", LOGGER_DEBUG);
break;
case 1:
logger("Mail: Deleting " . $msg_uid . " for " . $mailconf[0]['user']);
imap_delete($mbox, $msg_uid, FT_UID);
break;
case 2:
logger("Mail: Mark as seen " . $msg_uid . " for " . $mailconf[0]['user']);
imap_setflag_full($mbox, $msg_uid, "\\Seen", ST_UID);
break;
case 3:
logger("Mail: Moving " . $msg_uid . " to " . $mailconf[0]['movetofolder'] . " for " . $mailconf[0]['user']);
imap_setflag_full($mbox, $msg_uid, "\\Seen", ST_UID);
if ($mailconf[0]['movetofolder'] != "") {
imap_mail_move($mbox, $msg_uid, $mailconf[0]['movetofolder'], FT_UID);
}
break;
}
}
}
} else {
logger("Mail: no mails for " . $mailconf[0]['user']);
}
logger("Mail: closing connection for " . $mailconf[0]['user']);
imap_close($mbox);
}
} elseif ($contact['network'] === NETWORK_FACEBOOK) {
// This is picked up by the Facebook plugin on a cron hook.
// Ignored here.
} elseif ($contact['network'] === NETWORK_PUMPIO) {
// This is picked up by the pump.io plugin on a cron hook.
// Ignored here.
}
if ($xml) {
logger('poller: received xml : ' . $xml, LOGGER_DATA);
if (!strstr($xml, '<')) {
logger('poller: post_handshake: response from ' . $url . ' did not contain XML.');
$r = q("UPDATE `contact` SET `last-update` = '%s' WHERE `id` = %d", dbesc(datetime_convert()), intval($contact['id']));
return;
}
consume_feed($xml, $importer, $contact, $hub, 1, 1);
// do it twice. Ensures that children of parents which may be later in the stream aren't tossed
consume_feed($xml, $importer, $contact, $hub, 1, 2);
$hubmode = 'subscribe';
if ($contact['network'] === NETWORK_DFRN || $contact['blocked'] || $contact['readonly']) {
$hubmode = 'unsubscribe';
}
if (($contact['network'] === NETWORK_OSTATUS || $contact['network'] == NETWORK_FEED) && !$contact['hub-verify']) {
$hub_update = true;
}
if (strlen($hub) && $hub_update && ($contact['rel'] != CONTACT_IS_FOLLOWER || $contact['network'] == NETWORK_FEED)) {
logger('poller: hub ' . $hubmode . ' : ' . $hub . ' contact name : ' . $contact['name'] . ' local user : '******'name']);
$hubs = explode(',', $hub);
if (count($hubs)) {
foreach ($hubs as $h) {
$h = trim($h);
if (!strlen($h)) {
continue;
}
subscribe_to_hub($h, $importer, $contact, $hubmode);
}
}
}
}
$updated = datetime_convert();
$r = q("UPDATE `contact` SET `last-update` = '%s', `success_update` = '%s' WHERE `id` = %d", dbesc($updated), dbesc($updated), intval($contact['id']));
// load current friends if possible.
if ($contact['poco']) {
$r = q("SELECT count(*) as total from glink\n\t\t\twhere `cid` = %d and updated > UTC_TIMESTAMP() - INTERVAL 1 DAY", intval($contact['id']));
}
if (count($r)) {
if (!$r[0]['total']) {
poco_load($contact['id'], $importer_uid, 0, $contact['poco']);
}
}
return;
}
function poller_run($argv, $argc)
{
global $a, $db;
if (is_null($a)) {
$a = new App();
}
if (is_null($db)) {
@(include ".htconfig.php");
require_once "dba.php";
$db = new dba($db_host, $db_user, $db_pass, $db_data);
unset($db_host, $db_user, $db_pass, $db_data);
}
require_once 'include/session.php';
require_once 'include/datetime.php';
require_once 'library/simplepie/simplepie.inc';
require_once 'include/items.php';
require_once 'include/Contact.php';
require_once 'include/email.php';
require_once 'include/socgraph.php';
load_config('config');
load_config('system');
$a->set_baseurl(get_config('system', 'url'));
load_hooks();
logger('poller: start');
// run queue delivery process in the background
proc_run('php', "include/queue.php");
// expire any expired accounts
q("UPDATE user SET `account_expired` = 1 where `account_expired` = 0 \n\t\tAND `account_expires_on` != '0000-00-00 00:00:00' \n\t\tAND `account_expires_on` < UTC_TIMESTAMP() ");
$abandon_days = intval(get_config('system', 'account_abandon_days'));
if ($abandon_days < 1) {
$abandon_days = 0;
}
// once daily run birthday_updates and then expire in background
$d1 = get_config('system', 'last_expire_day');
$d2 = intval(datetime_convert('UTC', 'UTC', 'now', 'd'));
if ($d2 != intval($d1)) {
update_contact_birthdays();
update_suggestions();
set_config('system', 'last_expire_day', $d2);
proc_run('php', 'include/expire.php');
}
// clear old cache
Cache::clear();
$manual_id = 0;
$generation = 0;
$hub_update = false;
$force = false;
$restart = false;
if ($argc > 1 && $argv[1] == 'force') {
$force = true;
}
if ($argc > 1 && $argv[1] == 'restart') {
$restart = true;
$generation = intval($argv[2]);
if (!$generation) {
killme();
}
}
if ($argc > 1 && intval($argv[1])) {
$manual_id = intval($argv[1]);
$force = true;
}
$sql_extra = $manual_id ? " AND `id` = {$manual_id} " : "";
reload_plugins();
$d = datetime_convert();
if (!$restart) {
proc_run('php', 'include/cronhooks.php');
}
// Only poll from those with suitable relationships,
// and which have a polling address and ignore Diaspora since
// we are unable to match those posts with a Diaspora GUID and prevent duplicates.
$abandon_sql = $abandon_days ? sprintf(" AND `user`.`login_date` > UTC_TIMESTAMP() - INTERVAL %d DAY ", intval($abandon_days)) : '';
$contacts = q("SELECT `contact`.`id` FROM `contact` LEFT JOIN `user` ON `user`.`uid` = `contact`.`uid` \n\t\tWHERE ( `rel` = %d OR `rel` = %d ) AND `poll` != ''\n\t\tAND NOT `network` IN ( '%s', '%s' )\n\t\t{$sql_extra} \n\t\tAND `self` = 0 AND `contact`.`blocked` = 0 AND `contact`.`readonly` = 0 \n\t\tAND `user`.`account_expired` = 0 {$abandon_sql} ORDER BY RAND()", intval(CONTACT_IS_SHARING), intval(CONTACT_IS_FRIEND), dbesc(NETWORK_DIASPORA), dbesc(NETWORK_FACEBOOK));
if (!count($contacts)) {
return;
}
foreach ($contacts as $c) {
$res = q("SELECT * FROM `contact` WHERE `id` = %d LIMIT 1", intval($c['id']));
if (!$res || !count($res)) {
continue;
}
foreach ($res as $contact) {
$xml = false;
if ($manual_id) {
$contact['last-update'] = '0000-00-00 00:00:00';
}
if ($contact['network'] === NETWORK_DFRN || $contact['network'] === NETWORK_OSTATUS) {
$contact['priority'] = 2;
}
if ($contact['priority'] || $contact['subhub']) {
$hub_update = true;
$update = false;
$t = $contact['last-update'];
// We should be getting everything via a hub. But just to be sure, let's check once a day.
// (You can make this more or less frequent if desired by setting 'pushpoll_frequency' appropriately)
// This also lets us update our subscription to the hub, and add or replace hubs in case it
// changed. We will only update hubs once a day, regardless of 'pushpoll_frequency'.
if ($contact['subhub']) {
$interval = get_config('system', 'pushpoll_frequency');
$contact['priority'] = $interval !== false ? intval($interval) : 3;
$hub_update = false;
if (datetime_convert('UTC', 'UTC', 'now') > datetime_convert('UTC', 'UTC', $t . " + 1 day") || $force) {
$hub_update = true;
}
} else {
$hub_update = false;
}
/**
* Based on $contact['priority'], should we poll this site now? Or later?
*/
switch ($contact['priority']) {
case 5:
if (datetime_convert('UTC', 'UTC', 'now') > datetime_convert('UTC', 'UTC', $t . " + 1 month")) {
$update = true;
}
break;
case 4:
if (datetime_convert('UTC', 'UTC', 'now') > datetime_convert('UTC', 'UTC', $t . " + 1 week")) {
$update = true;
}
break;
case 3:
if (datetime_convert('UTC', 'UTC', 'now') > datetime_convert('UTC', 'UTC', $t . " + 1 day")) {
$update = true;
}
break;
case 2:
if (datetime_convert('UTC', 'UTC', 'now') > datetime_convert('UTC', 'UTC', $t . " + 12 hour")) {
$update = true;
}
break;
case 1:
default:
if (datetime_convert('UTC', 'UTC', 'now') > datetime_convert('UTC', 'UTC', $t . " + 1 hour")) {
$update = true;
}
break;
}
if (!$update && !$force) {
continue;
}
}
// Check to see if we are running out of memory - if so spawn a new process and kill this one
$avail_memory = return_bytes(ini_get('memory_limit'));
$memused = memory_get_peak_usage(true);
if (intval($avail_memory)) {
if ($memused / $avail_memory > 0.95) {
if ($generation + 1 > 10) {
logger('poller: maximum number of spawns exceeded. Terminating.');
killme();
}
logger('poller: memory exceeded. ' . $memused . ' bytes used. Spawning new poll.');
proc_run('php', 'include/poller.php', 'restart', (string) $generation + 1);
killme();
}
}
$importer_uid = $contact['uid'];
$r = q("SELECT * FROM `contact` WHERE `uid` = %d AND `self` = 1 LIMIT 1", intval($importer_uid));
if (!count($r)) {
continue;
}
$importer = $r[0];
logger("poller: poll: ({$contact['id']}) IMPORTER: {$importer['name']}, CONTACT: {$contact['name']}");
$last_update = $contact['last-update'] === '0000-00-00 00:00:00' ? datetime_convert('UTC', 'UTC', 'now - 30 days', ATOM_TIME) : datetime_convert('UTC', 'UTC', $contact['last-update'], ATOM_TIME);
if ($contact['network'] === NETWORK_DFRN) {
$idtosend = $orig_id = $contact['dfrn-id'] ? $contact['dfrn-id'] : $contact['issued-id'];
if (intval($contact['duplex']) && $contact['dfrn-id']) {
$idtosend = '0:' . $orig_id;
}
if (intval($contact['duplex']) && $contact['issued-id']) {
$idtosend = '1:' . $orig_id;
}
// they have permission to write to us. We already filtered this in the contact query.
$perm = 'rw';
$url = $contact['poll'] . '?dfrn_id=' . $idtosend . '&dfrn_version=' . DFRN_PROTOCOL_VERSION . '&type=data&last_update=' . $last_update . '&perm=' . $perm;
$handshake_xml = fetch_url($url);
logger('poller: handshake with url ' . $url . ' returns xml: ' . $handshake_xml, LOGGER_DATA);
if (!$handshake_xml) {
logger("poller: {$url} appears to be dead - marking for death ");
// dead connection - might be a transient event, or this might
// mean the software was uninstalled or the domain expired.
// Will keep trying for one month.
mark_for_death($contact);
// set the last-update so we don't keep polling
$r = q("UPDATE `contact` SET `last-update` = '%s' WHERE `id` = %d LIMIT 1", dbesc(datetime_convert()), intval($contact['id']));
continue;
}
if (!strstr($handshake_xml, '<?xml')) {
logger('poller: response from ' . $url . ' did not contain XML.');
$r = q("UPDATE `contact` SET `last-update` = '%s' WHERE `id` = %d LIMIT 1", dbesc(datetime_convert()), intval($contact['id']));
continue;
}
$res = parse_xml_string($handshake_xml);
if (intval($res->status) == 1) {
logger("poller: {$url} replied status 1 - marking for death ");
// we may not be friends anymore. Will keep trying for one month.
// set the last-update so we don't keep polling
$r = q("UPDATE `contact` SET `last-update` = '%s' WHERE `id` = %d LIMIT 1", dbesc(datetime_convert()), intval($contact['id']));
mark_for_death($contact);
} else {
if ($contact['term-date'] != '0000-00-00 00:00:00') {
logger("poller: {$url} back from the dead - removing mark for death");
unmark_for_death($contact);
}
}
if (intval($res->status) != 0 || !strlen($res->challenge) || !strlen($res->dfrn_id)) {
continue;
}
if ((double) $res->dfrn_version > 2.21 && $contact['poco'] == '') {
q("update contact set poco = '%s' where id = %d limit 1", dbesc(str_replace('/profile/', '/poco/', $contact['url'])), intval($contact['id']));
}
$postvars = array();
$sent_dfrn_id = hex2bin((string) $res->dfrn_id);
$challenge = hex2bin((string) $res->challenge);
$final_dfrn_id = '';
if ($contact['duplex'] && strlen($contact['prvkey'])) {
openssl_private_decrypt($sent_dfrn_id, $final_dfrn_id, $contact['prvkey']);
openssl_private_decrypt($challenge, $postvars['challenge'], $contact['prvkey']);
} else {
openssl_public_decrypt($sent_dfrn_id, $final_dfrn_id, $contact['pubkey']);
openssl_public_decrypt($challenge, $postvars['challenge'], $contact['pubkey']);
}
$final_dfrn_id = substr($final_dfrn_id, 0, strpos($final_dfrn_id, '.'));
if (strpos($final_dfrn_id, ':') == 1) {
$final_dfrn_id = substr($final_dfrn_id, 2);
}
if ($final_dfrn_id != $orig_id) {
logger('poller: ID did not decode: ' . $contact['id'] . ' orig: ' . $orig_id . ' final: ' . $final_dfrn_id);
// did not decode properly - cannot trust this site
continue;
}
$postvars['dfrn_id'] = $idtosend;
$postvars['dfrn_version'] = DFRN_PROTOCOL_VERSION;
$postvars['perm'] = 'rw';
$xml = post_url($contact['poll'], $postvars);
} elseif ($contact['network'] === NETWORK_OSTATUS || $contact['network'] === NETWORK_DIASPORA || $contact['network'] === NETWORK_FEED) {
// Upgrading DB fields from an older Friendika version
// Will only do this once per notify-enabled OStatus contact
// or if relationship changes
$stat_writeable = $contact['notify'] && ($contact['rel'] == CONTACT_IS_FOLLOWER || $contact['rel'] == CONTACT_IS_FRIEND) ? 1 : 0;
if ($stat_writeable != $contact['writable']) {
q("UPDATE `contact` SET `writable` = %d WHERE `id` = %d LIMIT 1", intval($stat_writeable), intval($contact['id']));
}
// Are we allowed to import from this person?
if ($contact['rel'] == CONTACT_IS_FOLLOWER || $contact['blocked'] || $contact['readonly']) {
continue;
}
$xml = fetch_url($contact['poll']);
} elseif ($contact['network'] === NETWORK_MAIL || $contact['network'] === NETWORK_MAIL2) {
$mail_disabled = function_exists('imap_open') && !get_config('system', 'imap_disabled') ? 0 : 1;
if ($mail_disabled) {
continue;
}
$mbox = null;
$x = q("SELECT `prvkey` FROM `user` WHERE `uid` = %d LIMIT 1", intval($importer_uid));
$mailconf = q("SELECT * FROM `mailacct` WHERE `server` != '' AND `uid` = %d LIMIT 1", intval($importer_uid));
if (count($x) && count($mailconf)) {
$mailbox = construct_mailbox_name($mailconf[0]);
$password = '';
openssl_private_decrypt(hex2bin($mailconf[0]['pass']), $password, $x[0]['prvkey']);
$mbox = email_connect($mailbox, $mailconf[0]['user'], $password);
unset($password);
if ($mbox) {
q("UPDATE `mailacct` SET `last_check` = '%s' WHERE `id` = %d AND `uid` = %d LIMIT 1", dbesc(datetime_convert()), intval($mailconf[0]['id']), intval($importer_uid));
}
}
if ($mbox) {
$msgs = email_poll($mbox, $contact['addr']);
if (count($msgs)) {
foreach ($msgs as $msg_uid) {
$datarray = array();
$meta = email_msg_meta($mbox, $msg_uid);
$headers = email_msg_headers($mbox, $msg_uid);
// look for a 'references' header and try and match with a parent item we have locally.
$raw_refs = x($headers, 'references') ? str_replace("\t", '', $headers['references']) : '';
$datarray['uri'] = trim($meta->message_id, '<>');
if ($raw_refs) {
$refs_arr = explode(' ', $raw_refs);
if (count($refs_arr)) {
for ($x = 0; $x < count($refs_arr); $x++) {
$refs_arr[$x] = "'" . str_replace(array('<', '>', ' '), array('', '', ''), dbesc($refs_arr[$x])) . "'";
}
}
$qstr = implode(',', $refs_arr);
$r = q("SELECT `uri` , `parent-uri` FROM `item` WHERE `uri` IN ( {$qstr} ) AND `uid` = %d LIMIT 1", intval($importer_uid));
if (count($r)) {
$datarray['parent-uri'] = $r[0]['uri'];
}
}
if (!x($datarray, 'parent-uri')) {
$datarray['parent-uri'] = $datarray['uri'];
}
// Have we seen it before?
$r = q("SELECT * FROM `item` WHERE `uid` = %d AND `uri` = '%s' LIMIT 1", intval($importer_uid), dbesc($datarray['uri']));
if (count($r)) {
if ($meta->deleted && !$r[0]['deleted']) {
q("UPDATE `item` SET `deleted` = 1, `changed` = '%s' WHERE `id` = %d LIMIT 1", dbesc(datetime_convert()), intval($r[0]['id']));
}
continue;
}
$datarray['title'] = notags(trim($meta->subject));
$datarray['created'] = datetime_convert('UTC', 'UTC', $meta->date);
$r = email_get_msg($mbox, $msg_uid);
if (!$r) {
continue;
}
$datarray['body'] = escape_tags($r['body']);
// some mailing lists have the original author as 'from' - add this sender info to msg body.
// todo: adding a gravatar for the original author would be cool
if (!stristr($meta->from, $contact['addr'])) {
$datarray['body'] = t('From: ') . escape_tags($meta->from) . "\n\n" . $datarray['body'];
}
$datarray['uid'] = $importer_uid;
$datarray['contact-id'] = $contact['id'];
if ($datarray['parent-uri'] === $datarray['uri']) {
$datarray['private'] = 1;
}
if ($contact['network'] === NETWORK_MAIL && !get_pconfig($importer_uid, 'system', 'allow_public_email_replies')) {
$datarray['private'] = 1;
$datarray['allow_cid'] = '<' . $contact['id'] . '>';
}
$datarray['author-name'] = $contact['name'];
$datarray['author-link'] = 'mailbox';
$datarray['author-avatar'] = $contact['photo'];
$stored_item = item_store($datarray);
q("UPDATE `item` SET `last-child` = 0 WHERE `parent-uri` = '%s' AND `uid` = %d", dbesc($datarray['parent-uri']), intval($importer_uid));
q("UPDATE `item` SET `last-child` = 1 WHERE `id` = %d LIMIT 1", intval($stored_item));
}
}
imap_close($mbox);
}
} elseif ($contact['network'] === NETWORK_FACEBOOK) {
// This is picked up by the Facebook plugin on a cron hook.
// Ignored here.
}
if ($xml) {
logger('poller: received xml : ' . $xml, LOGGER_DATA);
if (!strstr($xml, '<?xml')) {
logger('poller: post_handshake: response from ' . $url . ' did not contain XML.');
$r = q("UPDATE `contact` SET `last-update` = '%s' WHERE `id` = %d LIMIT 1", dbesc(datetime_convert()), intval($contact['id']));
continue;
}
consume_feed($xml, $importer, $contact, $hub, 1, 1);
// do it twice. Ensures that children of parents which may be later in the stream aren't tossed
consume_feed($xml, $importer, $contact, $hub, 1, 2);
$hubmode = 'subscribe';
if ($contact['network'] === NETWORK_DFRN || $contact['blocked'] || $contact['readonly']) {
$hubmode = 'unsubscribe';
}
if (strlen($hub) && $hub_update && $contact['rel'] != CONTACT_IS_FOLLOWER) {
logger('poller: hub ' . $hubmode . ' : ' . $hub . ' contact name : ' . $contact['name'] . ' local user : '******'name']);
$hubs = explode(',', $hub);
if (count($hubs)) {
foreach ($hubs as $h) {
$h = trim($h);
if (!strlen($h)) {
continue;
}
subscribe_to_hub($h, $importer, $contact, $hubmode);
}
}
}
}
$updated = datetime_convert();
$r = q("UPDATE `contact` SET `last-update` = '%s', `success_update` = '%s' WHERE `id` = %d LIMIT 1", dbesc($updated), dbesc($updated), intval($contact['id']));
// load current friends if possible.
if ($contact['poco']) {
$r = q("SELECT count(*) as total from glink \n\t\t\t\t\twhere `cid` = %d and updated > UTC_TIMESTAMP() - INTERVAL 1 DAY", intval($contact['id']));
}
if (count($r)) {
if (!$r[0]['total']) {
poco_load($contact['id'], $importer_uid, $contact['poco']);
}
}
// loop - next contact
}
}
return;
}
function probe_url($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_guid = '';
$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);
$lastfm = strpos($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)) {
$has_lrdd = true;
logger('probe_url: found lrdd links: ' . print_r($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) {
$diaspora = false;
}
}
// Status.Net can have more than one profile URL. We need to match the profile URL
// to a contact on incoming messages to prevent spam, and we won't know which one
// to match. So in case of two, one of them is stored as an alias. Only store URL's
// and not webfinger user@host aliases. If they've got more than two non-email style
// aliases, let's hope we're lucky and get one that matches the feed author-uri because
// otherwise we're screwed.
foreach ($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']);
}
}
}
}
// 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'] = $ret['photo'];
}
}
}
if ($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 && $network !== NETWORK_MAIL) {
if ($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'] = $addr_parts[0];
}
/* if($twitter) {
logger('twitter: setup');
$tid = basename($url);
$tapi = 'https://api.twitter.com/1/statuses/user_timeline.rss';
if(intval($tid))
$poll = $tapi . '?user_id=' . $tid;
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'];
}
}
$check_feed = false;
if (stristr($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'];
}
require_once 'library/simplepie/simplepie.inc';
$feed = new SimplePie();
$xml = fetch_url($poll);
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);
// Don't try and parse an empty string
$feed->set_raw_data($xml ? $xml : '<?xml version="1.0" encoding="utf-8" ?><xml></xml>');
$feed->init();
if ($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();
}
$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'], '@'));
}
$email = unxmlify($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) {
$elems = $rawtags[0]['child'][SIMPLEPIE_NAMESPACE_ATOM_10];
if (x($elems, 'link') && $elems['link'][0]['attribs']['']['rel'] === 'photo') {
$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 = $feed->get_item(0);
if ($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'], '@'));
}
$email = unxmlify($author->get_email());
if (!$profile && $author->get_link()) {
$profile = trim(unxmlify($author->get_link()));
}
}
if (!$vcard['photo']) {
$rawmedia = $item->get_item_tags('http://search.yahoo.com/mrss/', 'thumbnail');
if ($rawmedia && $rawmedia[0]['attribs']['']['url']) {
$vcard['photo'] = unxmlify($rawmedia[0]['attribs']['']['url']);
}
}
if (!$vcard['photo']) {
$rawtags = $item->get_item_tags(SIMPLEPIE_NAMESPACE_ATOM_10, 'author');
if ($rawtags) {
$elems = $rawtags[0]['child'][SIMPLEPIE_NAMESPACE_ATOM_10];
if (x($elems, 'link') && $elems['link'][0]['attribs']['']['rel'] === 'photo') {
$vcard['photo'] = $elems['link'][0]['attribs']['']['href'];
}
}
}
}
}
// Workaround for misconfigured Friendica servers
if ($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 = $noscrapedata["dfrn-poll"];
}
}
}
if (!$vcard['photo'] && strlen($email)) {
$vcard['photo'] = avatar_img($email);
}
if ($poll === $profile) {
$lnk = $feed->get_permalink();
}
if (isset($lnk) && strlen($lnk)) {
$profile = $lnk;
}
if (!$network) {
$network = NETWORK_FEED;
// If it is a feed, don't take the author name as feed name
unset($vcard['fn']);
}
if (!x($vcard, 'fn')) {
$vcard['fn'] = notags($feed->get_title());
}
if (!x($vcard, 'fn')) {
$vcard['fn'] = notags($feed->get_description());
}
if (strpos($vcard['fn'], 'Twitter / ') !== false) {
$vcard['fn'] = substr($vcard['fn'], strpos($vcard['fn'], '/') + 1);
$vcard['fn'] = trim($vcard['fn']);
}
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'] = $a->get_baseurl() . '/images/person-175.jpg';
}
if (!$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'] = $url;
}
if ($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, "/");
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['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 ($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_PHANTOM) {
Cache::set("probe_url:" . $mode . ":" . $url, serialize($result), CACHE_DAY);
}
return $result;
}
