[19:56:17] Reedy: Want me to merge https://gerrit.wikimedia.org/r/#/c/326507/ ? [19:56:43] (https://gerrit.wikimedia.org/r/#/c/326506/ has a comment from Daniel K. but it's not exactly a supported extension.) [19:57:11] heh [19:57:13] Please [19:58:13] https://gerrit.wikimedia.org/r/#/c/326506/ and https://gerrit.wikimedia.org/r/#/c/326480/ can go too [19:58:20] I mean there's already a patch for the unsupported extension, why not land it? :) [19:58:48] SOMEONE MAY CARE ABOUT MW < 1.21 [19:59:01] That's what branches are for [19:59:02] Well, I was giving Reedy the opportunity to say one of "OK, I'll fix it" or "caveat emptor". :-) [19:59:06] "This may result in $text being an object, or binary garbage, or silly hat." [19:59:15] * ostriches files bug "MediaWiki needs more silly hats" [19:59:23] ostriches: please do, I shall award tokens [19:59:26] ostriches: I think we call that one "Wikibase" [19:59:45] James_F: I'm making sure patches exist for all usages replacements [19:59:50] either personally, or sudo -u gci [20:01:22] .popupPanel .close { [20:01:22] background-image: url(/icons/overlay/close.png); [20:01:27] If that image doesn't exist [20:01:29] Should I just remove hte CSS? [20:01:49] It's causing a jenkins failure... [20:01:50] https://integration.wikimedia.org/ci/job/mwext-testextension-php55/29470/console [20:02:29] * James_F grins. [20:02:35] Yes. [20:03:01] There's so much shit in our repos [20:03:04] It's depressing [20:03:09] https://gerrit.wikimedia.org/r/326530 [20:08:55] Reedy: does it definitely not exist? the path starting with '/' is weird, but maybe that works in practice, only not in tests [20:09:19] MatmaRex: There's no images at all in popuppages [20:09:33] https://github.com/wikimedia/mediawiki-extensions-PopupPages/tree/master/resources [20:10:40] it's kind of weird to remove it, i feel like we're making the bug harder to fix. you can't git blame removals. comment it out? :D [20:11:07] MatmaRex: Fair point [20:11:39] not that anyone cares about that extension, probably. some of the other bits of css also look broken. but still [20:11:51] fixed [20:12:25] 20:08:28 Fatal error: Call to a member function canExist() on a non-object in /srv/jenkins-workspace/workspace/mwext-testextension-php55/src/includes/skins/Skin.php on line 212 [20:12:29] Why is Premoderation breaking that [20:29:16] MatmaRex: Find a revision before the removal, then use git blame --reverse? [20:29:40] anomie: In this case, I think commenting out does make more sense though ;) [21:38:52] bd808: any idea who could help with the ruby part of https://gerrit.wikimedia.org/r/#/c/326394/ ? [21:39:13] or should I just discard those tests? not like the rest of wmflib has any [22:34:44] tgr: marxarelli or zeljkof are probably your best bet for rspec help [22:35:18] thx [22:48:37] bd808: btw, any thoughts about https://gerrit.wikimedia.org/r/#/c/187874/ ? As far as I can tell remote debugging in PHPStorm is not possible without that (or ugly manual hacks) [22:49:37] tgr: can you put them in srv instead of the root? I think we have too much crap in the root [22:50:37] recurse in Puppet is gross but I don't think we have too many files in the docroot [23:37:16] done, tested [23:37:22] (19 files) [23:52:11] did something change in the vagrant webserver setup? it seems much more aggressive about dropping the connection when doing live debugging for longer than a few seconds [23:54:49] tgr: only whatever upstream bug fixes may have come in. Nothing systemic has changed in the apache config for mw-v in a long time.