[09:06:10] Glaisher: review on https://gerrit.wikimedia.org/r/#/c/304692/ :) please! [09:06:31] I did leave a comment there [09:06:50] oh new patch [09:10:23] tonythomas: added some comments [09:10:35] Glaisher: checking now [09:57:47] Glaisher: was afk. pushed in new one. removed the else { part so that it would show up the original fatal error in https://gerrit.wikimedia.org/r/#/c/304692/3/includes/NewsletterEditPage.php [17:31:46] Glaisher: have time for one more review over here - https://gerrit.wikimedia.org/r/#/c/304692/ [17:38:41] tonythomas: done [17:41:36] Glaisher: wow. That was fast :) [17:42:14] I wasn't doing anything so I did have time :) [17:42:35] so this parsercache. I have seen it somewhere. Let me grep [17:44:17] ah. we have public function isParserCacheSupported() { return false; } in the NewsletterContenthandler page [17:44:27] *class [17:45:30] ah, okay cool [17:46:06] so your second comment (I can barely understand what it says) would not be an issue ? [17:49:32] I think so... cannot say for sure whether there are other things which would occur like that [17:56:09] Glaisher: alright -- and for the other one - https://gerrit.wikimedia.org/r/#/c/304692/2..3/includes/NewsletterEditPage.php -- do you think that returning Status::newFatal( 'newsletter-create-error' ); would be enough ? [17:56:28] or should we have a custom function that do the rollback, without adding in the logs or so ? [17:57:15] the db queries failed check will already happen in the if ( $newsletterCreated ) { } I guess ?