Import backlog on new subscription.

Danger is when importing a new feed that may be maliciously crafted
to contain a zillion entries.
This commit is contained in:
Mikael Nordfeldth 2017-04-21 09:31:27 +02:00
parent f51cb6fca9
commit e98bceec10
3 changed files with 33 additions and 3 deletions

View File

@ -77,7 +77,7 @@ class PushCallbackAction extends Action
/**
* Handler for GET verification requests from the hub.
*/
function handleGet()
public function handleGet()
{
$mode = $this->arg('hub_mode');
$topic = $this->arg('hub_topic');
@ -110,12 +110,21 @@ class PushCallbackAction extends Action
}
if ($mode == 'subscribe') {
if ($feedsub->sub_state == 'active') {
$renewal = ($feedsub->sub_state == 'active');
if ($renewal) {
common_log(LOG_INFO, __METHOD__ . ': sub update confirmed');
} else {
common_log(LOG_INFO, __METHOD__ . ': sub confirmed');
}
$feedsub->confirmSubscribe($lease_seconds);
if (!$renewal) {
// Kickstart the feed by importing its most recent backlog
// FIXME: Send this to background queue handling
common_log(LOG_INFO, __METHOD__ . ': Confirmed a new subscription, importing backlog...');
$feedsub->importFeed();
}
} else {
common_log(LOG_INFO, __METHOD__ . ": unsub confirmed; deleting sub record for $topic");
$feedsub->confirmUnsubscribe();

View File

@ -405,6 +405,7 @@ class FeedSub extends Managed_DataObject
}
$this->modified = common_sql_now();
common_debug(__METHOD__ . ': Updating sub state and metadata for '.$this->getUri());
return $this->update($original);
}
@ -463,6 +464,24 @@ class FeedSub extends Managed_DataObject
$this->receiveFeed($post);
}
/**
* All our feed URIs should be URLs.
*/
public function importFeed()
{
$feed_url = $this->getUri();
// Fetch the URL
try {
common_log(LOG_INFO, sprintf('Importing feed backlog from %s', $feed_url));
$feed_xml = HTTPClient::quickGet($feed_url, 'application/atom+xml');
} catch (Exception $e) {
throw new FeedSubException("Could not fetch feed from URL '%s': %s (%d).\n", $feed_url, $e->getMessage(), $e->getCode());
}
return $this->receiveFeed($feed_xml);
}
protected function receiveFeed($feed_xml)
{
// We're passed the XML for the Atom feed as $feed_xml,

View File

@ -53,9 +53,11 @@ if (!$sub) {
exit(1);
}
// XXX: This could maybe be replaced with $sub->importFeed()
// Fetch the URL
try {
$xml = HTTPClient::quickGet($feedurl, 'text/html,application/xhtml+xml');
$xml = HTTPClient::quickGet($feedurl, 'application/atom+xml');
} catch (Exception $e) {
echo sprintf("Could not fetch feedurl %s (%d).\n", $e->getMessage(), $e->getCode());
exit(1);