friendica-github/include/poller.php

517 lines
15 KiB
PHP
Raw Normal View History

2010-07-19 03:49:54 +00:00
<?php
2011-04-16 06:40:43 +00:00
require_once("boot.php");
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);
};
2010-08-17 03:47:40 +00:00
2011-04-16 06:40:43 +00:00
require_once('include/session.php');
require_once('include/datetime.php');
require_once('library/simplepie/simplepie.inc');
2010-09-02 07:31:11 +00:00
require_once('include/items.php');
2010-09-09 03:14:17 +00:00
require_once('include/Contact.php');
2011-04-16 06:40:43 +00:00
require_once('include/email.php');
2010-09-09 03:14:17 +00:00
load_config('config');
load_config('system');
2010-08-17 03:47:40 +00:00
$a->set_baseurl(get_config('system','url'));
2010-07-19 03:49:54 +00:00
2011-04-26 11:39:27 +00:00
load_hooks();
2010-11-26 02:31:33 +00:00
logger('poller: start');
2010-11-26 02:22:54 +00:00
// run queue delivery process in the background
2010-11-22 23:30:52 +00:00
2011-02-23 23:16:12 +00:00
proc_run('php',"include/queue.php");
2011-01-31 16:16:35 +00:00
2011-03-16 00:31:49 +00:00
// once daily run expire in background
$d1 = get_config('system','last_expire_day');
$d2 = intval(datetime_convert('UTC','UTC','now','d'));
if($d2 != intval($d1)) {
set_config('system','last_expire_day',$d2);
proc_run('php','include/expire.php');
}
2011-01-31 16:16:35 +00:00
// clear old cache
q("DELETE FROM `cache` WHERE `updated` < '%s'",
2011-01-31 16:16:35 +00:00
dbesc(datetime_convert('UTC','UTC',"now - 30 days")));
2010-11-22 23:30:52 +00:00
$manual_id = 0;
$generation = 0;
$hub_update = false;
$force = false;
$restart = false;
2010-11-05 03:47:44 +00:00
if(($argc > 1) && ($argv[1] == 'force'))
$force = true;
if(($argc > 1) && ($argv[1] == 'restart')) {
$restart = true;
$generation = intval($argv[2]);
if(! $generation)
killme();
}
2011-01-24 04:09:34 +00:00
if(($argc > 1) && intval($argv[1])) {
$manual_id = intval($argv[1]);
$force = true;
2011-01-24 04:09:34 +00:00
}
$sql_extra = (($manual_id) ? " AND `id` = $manual_id " : "");
2011-04-26 11:39:27 +00:00
reload_plugins();
2011-03-09 10:12:32 +00:00
$d = datetime_convert();
2011-04-26 11:39:27 +00:00
if(! $restart)
proc_run('php','include/cronhooks.php');
2011-03-09 10:12:32 +00:00
2011-08-24 01:17:35 +00:00
// 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.
2011-09-19 08:17:12 +00:00
$contacts = q("SELECT `contact`.`id` FROM `contact` LEFT JOIN `user` ON `user`.`uid` = `contact`.`uid`
WHERE ( `rel` = %d OR `rel` = %d ) AND `poll` != ''
2011-08-24 01:17:35 +00:00
AND `network` != '%s'
2011-01-24 04:09:34 +00:00
$sql_extra
2011-09-19 08:17:12 +00:00
AND `self` = 0 AND `contact`.`blocked` = 0 AND `contact`.`readonly` = 0
AND `user`.`account_expired` = 0 ORDER BY RAND()",
2011-08-07 23:15:54 +00:00
intval(CONTACT_IS_SHARING),
2011-08-24 01:17:35 +00:00
intval(CONTACT_IS_FRIEND),
dbesc(NETWORK_DIASPORA)
);
2010-07-19 03:49:54 +00:00
if(! count($contacts)) {
return;
}
2010-07-19 03:49:54 +00:00
foreach($contacts as $c) {
2010-07-19 03:49:54 +00:00
$res = q("SELECT * FROM `contact` WHERE `id` = %d LIMIT 1",
intval($c['id'])
);
if((! $res) || (! count($res)))
continue;
foreach($res as $contact) {
2011-07-01 05:00:08 +00:00
2011-04-16 06:40:43 +00:00
$xml = false;
if($manual_id)
$contact['last-update'] = '0000-00-00 00:00:00';
if($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;
}
2011-09-22 11:11:39 +00:00
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 = $r[0];
2010-07-19 03:49:54 +00:00
logger("poller: poll: IMPORTER: {$importer['name']}, CONTACT: {$contact['name']}");
2010-07-19 03:49:54 +00:00
$last_update = (($contact['last-update'] === '0000-00-00 00:00:00')
? datetime_convert('UTC','UTC','now - 30 days', ATOM_TIME)
: datetime_convert('UTC','UTC',$contact['last-update'], ATOM_TIME)
);
2010-07-19 03:49:54 +00:00
2011-04-16 06:40:43 +00:00
if($contact['network'] === NETWORK_DFRN) {
2010-09-09 12:25:01 +00:00
$idtosend = $orig_id = (($contact['dfrn-id']) ? $contact['dfrn-id'] : $contact['issued-id']);
2010-07-19 03:49:54 +00:00
if(intval($contact['duplex']) && $contact['dfrn-id'])
$idtosend = '0:' . $orig_id;
if(intval($contact['duplex']) && $contact['issued-id'])
2011-04-11 02:29:21 +00:00
$idtosend = '1:' . $orig_id;
// they have permission to write to us. We already filtered this in the contact query.
$perm = 'rw';
2010-09-13 04:25:37 +00:00
2011-03-30 22:04:18 +00:00
$url = $contact['poll'] . '?dfrn_id=' . $idtosend
. '&dfrn_version=' . DFRN_PROTOCOL_VERSION
2011-04-11 02:29:21 +00:00
. '&type=data&last_update=' . $last_update
. '&perm=' . $perm ;
2011-04-16 06:40:43 +00:00
$handshake_xml = fetch_url($url);
2010-09-13 04:25:37 +00:00
2011-04-16 06:40:43 +00:00
logger('poller: handshake with url ' . $url . ' returns xml: ' . $handshake_xml, LOGGER_DATA);
2010-09-02 07:31:11 +00:00
2011-04-16 06:40:43 +00:00
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);
2010-09-02 07:31:11 +00:00
// 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;
}
2011-04-16 06:40:43 +00:00
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;
}
2010-09-13 04:25:37 +00:00
2011-04-16 06:40:43 +00:00
$res = parse_xml_string($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
2010-09-13 04:25:37 +00:00
$r = q("UPDATE `contact` SET `last-update` = '%s' WHERE `id` = %d LIMIT 1",
dbesc(datetime_convert()),
intval($contact['id'])
);
2010-07-19 03:49:54 +00:00
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;
$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']);
2010-11-05 03:47:44 +00:00
}
2010-09-09 03:14:17 +00:00
$final_dfrn_id = substr($final_dfrn_id, 0, strpos($final_dfrn_id, '.'));
2010-07-19 03:49:54 +00:00
if(strpos($final_dfrn_id,':') == 1)
$final_dfrn_id = substr($final_dfrn_id,2);
2010-07-19 03:49:54 +00:00
if($final_dfrn_id != $orig_id) {
2011-04-11 02:29:21 +00:00
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;
2011-04-11 02:29:21 +00:00
$postvars['perm'] = 'rw';
2010-09-02 07:31:11 +00:00
$xml = post_url($contact['poll'],$postvars);
}
2011-04-16 06:40:43 +00:00
elseif(($contact['network'] === NETWORK_OSTATUS)
|| ($contact['network'] === NETWORK_DIASPORA)
|| ($contact['network'] === NETWORK_FEED) ) {
2010-09-02 07:31:11 +00:00
2011-04-16 06:40:43 +00:00
// Upgrading DB fields from an older Friendika version
// Will only do this once per notify-enabled OStatus contact
// or if relationship changes
2011-08-07 23:15:54 +00:00
$stat_writeable = ((($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($contact['id'])
);
}
2011-04-16 06:40:43 +00:00
// Are we allowed to import from this person?
2011-08-07 23:15:54 +00:00
if($contact['rel'] == CONTACT_IS_FOLLOWER || $contact['blocked'] || $contact['readonly'])
continue;
$xml = fetch_url($contact['poll']);
}
2011-04-16 06:40:43 +00:00
elseif($contact['network'] === NETWORK_MAIL) {
$mail_disabled = ((function_exists('imap_open') && (! get_config('system','imap_disabled'))) ? 0 : 1);
if($mail_disabled)
continue;
2011-04-18 22:16:51 +00:00
$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)
);
2011-04-16 06:40:43 +00:00
}
}
if($mbox) {
2011-04-18 06:27:11 +00:00
2011-04-16 06:40:43 +00:00
$msgs = email_poll($mbox,$contact['addr']);
2011-04-18 06:27:11 +00:00
2011-04-16 06:40:43 +00:00
if(count($msgs)) {
2011-04-16 15:45:08 +00:00
foreach($msgs as $msg_uid) {
$datarray = array();
$meta = email_msg_meta($mbox,$msg_uid);
2011-04-18 06:27:11 +00:00
$headers = email_msg_headers($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']) : '');
2011-04-16 15:45:08 +00:00
$datarray['uri'] = trim($meta->message_id,'<>');
2011-04-18 06:27:11 +00:00
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'];
2011-04-16 15:45:08 +00:00
// Have we seen it before?
$r = q("SELECT * FROM `item` WHERE `uid` = %d AND `uri` = '%s' LIMIT 1",
intval($importer_uid),
dbesc($datarray['uri'])
);
2011-04-18 06:27:11 +00:00
2011-04-16 15:45:08 +00:00
if(count($r)) {
if($meta->deleted && ! $r[0]['deleted']) {
q("UPDATE `item` SET `deleted` = 1, `changed` = '%s' WHERE `id` = %d LIMIT 1",
2011-04-16 15:45:08 +00:00
dbesc(datetime_convert()),
intval($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'];
2011-04-16 15:45:08 +00:00
$datarray['uid'] = $importer_uid;
$datarray['contact-id'] = $contact['id'];
2011-04-18 06:27:11 +00:00
if($datarray['parent-uri'] === $datarray['uri'])
$datarray['private'] = 1;
2011-08-11 08:52:31 +00:00
if(! get_pconfig($importer_uid,'system','allow_public_email_replies')) {
$datarray['private'] = 1;
2011-08-11 08:52:31 +00:00
$datarray['allow_cid'] = '<' . $contact['id'] . '>';
}
2011-04-16 15:45:08 +00:00
$datarray['author-name'] = $contact['name'];
$datarray['author-link'] = 'mailbox';
$datarray['author-avatar'] = $contact['photo'];
2011-04-18 06:27:11 +00:00
$stored_item = item_store($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($stored_item)
);
2011-04-16 15:45:08 +00:00
}
2011-04-16 06:40:43 +00:00
}
2011-04-18 22:16:51 +00:00
imap_close($mbox);
2011-04-16 06:40:43 +00:00
}
}
elseif($contact['network'] === NETWORK_FACEBOOK) {
2011-08-10 08:19:27 +00:00
// This is picked up by the Facebook plugin on a cron hook.
// Ignored here.
}
2010-09-13 04:25:37 +00:00
2011-04-16 06:40:43 +00:00
if($xml) {
logger('poller: received xml : ' . $xml, LOGGER_DATA);
2011-04-04 08:18:28 +00:00
2011-04-16 06:40:43 +00:00
if(! strstr($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;
}
2011-04-16 06:40:43 +00:00
consume_feed($xml,$importer,$contact,$hub,1, true);
2011-04-16 06:40:43 +00:00
// 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);
2011-09-22 11:11:39 +00:00
$hubmode = 'subscribe';
if($contact['network'] === NETWORK_DFRN || $contact['blocked'] || $contact['readonly'])
$hubmode = 'unsubscribe';
2011-04-16 06:40:43 +00:00
2011-09-22 11:11:39 +00:00
if((strlen($hub)) && ($hub_update) && ($contact['rel'] != CONTACT_IS_FOLLOWER)) {
logger('poller: hub ' . $hubmode . ' : ' . $hub . ' contact name : ' . $contact['name'] . ' local user : ' . $importer['name']);
2011-04-16 06:40:43 +00:00
$hubs = explode(',', $hub);
if(count($hubs)) {
foreach($hubs as $h) {
$h = trim($h);
if(! strlen($h))
continue;
2011-09-22 11:11:39 +00:00
subscribe_to_hub($h,$importer,$contact,$hubmode);
2011-04-16 06:40:43 +00:00
}
}
}
}
2011-09-22 11:11:39 +00:00
$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'])
);
2010-07-19 03:49:54 +00:00
// loop - next contact
}
}
2011-04-16 15:45:08 +00:00
2010-07-19 03:49:54 +00:00
return;
}
2010-07-19 03:49:54 +00:00
if (array_search(__file__,get_included_files())===0){
poller_run($argv,$argc);
killme();
}