[00:36:04] kaldari: can you wave at Fabrice and ask him to switch on Skype or IRC? I'm on 3 and need to ask him something [00:36:46] I let him know [00:36:59] thx [00:59:21] if Title::setFragment() is deprecated, what's the correct method for adding a URL fragment to an existing Title object? [09:53:55] New patchset: Hashar; "check-voter pipeline, lint with verify+1" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/37198 [09:53:55] New patchset: Hashar; "check-voter for ops/puppet (v+1 on lint success)" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/37199 [09:54:38] Change merged: Hashar; [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/37199 [09:54:38] Change merged: Hashar; [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/37198 [10:06:44] New patchset: Hashar; "enable reporting on mediawiki/core.git" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/37200 [10:07:03] Change merged: Hashar; [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/37200 [10:57:48] New patchset: Hashar; "Get rid of the old InjectExtensions.php" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/37204 [10:58:11] Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/37204 [11:08:44] New patchset: Hashar; "jobs for mw/ext/Wikibase" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/37205 [11:08:58] Change merged: Hashar; [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/37205 [11:11:33] New patchset: Hashar; "add Wikibase mw extension" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/37207 [11:11:48] Change merged: Hashar; [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/37207 [17:49:48] AaronSchulz: hello :) [17:50:02] AaronSchulz: ends up my patch to the TranslateNotificationJob is not going to work :((( [17:51:35] hello hashar [17:53:06] I guess SanitizerTests are passing in jenkins... [17:53:28] Platonides: they should [17:53:42] well, some of them fail locally [17:54:33] merged 19 Nov: https://gerrit.wikimedia.org/r/17793 [17:55:05] might be due to some setting in your local install [17:55:14] or some global not being properly applied in the test [17:57:10] Sanitizer.php seems to properly check for the three globals [18:02:14] hi, i'm working on mediawiki+openldap configuration, i download LdapAuthentication but this plugin have a bug with postgresql, and i make a patch to fix, how to i submit the patch to resolv the problem? (sorry for my english) [18:04:49] no problem, dub_ [18:05:03] you can provide it on http://bugzilla.wikimedia.org/ [18:05:24] alternatively, you could get an account on our git server (gerrit) and submit it there [18:06:14] http://www.mediawiki.org/wiki/Developer_access [18:06:46] Platonides, excelent, when i have a any little free time i submit the patch, thanks! [18:07:15] you are welcome! [18:32:10] hashar, where is code-utils now? [18:32:55] Platonides: mediawiki/tools/code-utils ? [18:33:20] I don't have that folder [18:33:36] I may need to check which repos I miss [18:33:59] I can use the clone-all.php script which lives in code-util :) [18:34:05] Platonides: indeed https://gerrit.wikimedia.org/r/#/q/project:mediawiki/tools/code-utils,n,z [18:34:25] there is no jenkins job on this repo yet though [19:24:00] Is this the room for questions from the live streamed meeting? [19:43:11] blobaugh: i think that would be #wikimedia-office [19:44:41] DanielK_WMDE: got it. thanks [19:45:17] * hexmode waves at blobaugh  [19:45:18] :) [19:47:21] hey hexmode! [20:31:15] there is no brown bag meeting today? [20:32:44] metrics meeting is still going on [20:33:20] kaldari: thanks, so it is canceled? or just delayed? [20:33:26] not sure [20:42:19] I believe the weekly tech chat is NOT happening today. Next week instead [20:42:48] * DanielK_WMDE waves at sumanah [20:43:02] hi there DanielK_WMDE [20:43:23] DanielK_WMDE: http://ur1.ca/argdk for the notes just now from the monthly metrics meeting at WMF [20:43:28] how are you and your family doing DanielK_WMDE? [20:46:54] sumanah: the christmad madness is taking a grip :) other than that, pretty good, actually. [20:47:31] sumanah: wikidata is quite exciting. chad is currently setting up or first client wiki (just test2 for the moment, but still: the cross wiki change propagation code is going live) [20:48:40] aw yeah [21:06:44] chrismcmahonbrb: ping me when you're back [21:06:54] hi kaldari [21:07:23] I've been back, forgot to change my nick [21:09:14] as you've probably heard, the h3 bug was reported on several wikis since yesterday and is still persisting for a few people. I've deployed the new UW code, but I'm reluctant to turn it on until I know what's causing the 42452 problem. [21:10:37] understood [21:10:43] saw your comment on the bug [21:10:46] a least until the new JS has had a few days to sit [21:11:06] thanks for the heads up [21:13:40] chrismcmahon: I pinged Ops, but haven't heard anything from them. The next step would be to work on actually identifying a server that is serving outdated CSS/JS, but I don't have the bandwidth to work on this. [21:14:10] Hi Kaldari, got a minute for an Echo question? [21:14:15] sure [21:15:23] What is the easiest way to trigger my dev install to send me an email notification? [21:16:17] kaldari: How do you detect an outdated server? [21:16:24] well, there isn't really a super easy way, but one way is to leave a message on your talk page with another account [21:16:40] I have a plain install, and I get verify your email mails so I know email works, but I'm either not findng a preference for enabling notifications or I;m not runngn the job queue or something [21:16:48] Just wondering if you could make a nagios check for that ;-) [21:17:35] I have been trying by logging out and editing my own talk page [21:17:45] If the Nagios check is getting new messages it's looking at an old server [21:18:45] multichill: I don't know any automated way to do it. I was thinking, deploy some dummy CSS/JS, check what is getting served from various servers (you can see which servers are sending the content from the HTTP headers) and hope to find one that is misbehaving. [21:19:35] lwelling1: yes, you have to run your local job queue to get the emails... [21:20:09] kaldari: Something else, do we have a manual somewhere about the job queue? [21:20:09] manually triggering runJobs.php is not good enough? [21:20:33] no, that should work [21:20:53] multichill: not sure, if we do, please let me know so I can read it :) [21:21:41] Hmm, found it, for some reason I missed it earlier this week [21:21:48] https://www.mediawiki.org/wiki/Manual:Job_queue [21:22:19] thanks [21:23:06] lwelling1: You can also try disabling Echo and see if you get emails from talk page messages normally. [21:23:47] OK, thanks I'll try that and see if I can trace it back [21:25:04] lwelling1: In that case you'll need to make sure that 'E-mail me when my user talk page is changed' is on in your prefs, but I believe that is the default. [21:25:45] and I assume you've actually verified your email address [21:26:11] yes the verify email is the only one I've gotten so far [21:36:54] Thanks kaldari, I don't know waht was broken, but doign it all again slowly with a fresh user makes it work [21:37:11] yay [23:01:01] ori-l: are you guys still deploying? [23:01:46] spagewmf: ^ [23:02:17] oh yes [23:02:34] Are you Adrian? [23:02:51] spagewmf: ok - can you let me know when you guys are done? i have a MobileFrontend bugfix i need to push out [23:02:54] spagewmf: nope, arthur :) [23:05:13] awjr: you can go ahead [23:05:20] ori-l: thanks :) [23:05:23] we're not done but we need a little bit before we're ready for the next sync [23:05:45] ^ awjr [23:06:02] ori-l ok sounds good, this should be relatively quick [23:06:11] our deployment window is 1pm-3.... am, right? :-) [23:08:10] 1pm PST to 3pm UTC (+1) [23:18:40] 57 PHP Warning: Division by zero in /usr/local/apache/common-local/php-1.21wmf5/extensions/ParserFunctions/Expr.php on line 429 [23:18:42] That one is annoying me [23:32:48] Reedy: that's bizarre [23:33:31] yup [23:33:33] Reedy: array() [23:33:47] ori-l: null [23:34:01] Reedy: null == 0 [23:34:11] array() != 0 [23:34:14] so it can slip past the guard [23:34:41] i'm not sure if those values can actually make it there [23:35:15] exprtTests.txt should allow it to be tested [23:36:48] We don't track the sources of warnings in our logs [23:36:53] which makes attempting to debug a PitA [23:40:03] using either array() or null gives the warning [23:41:47] So just sticking || is_array( $right )... [23:42:12] i'd go with if ( (integer) $right === 0 ) [23:46:52] I think we get it on / too, so adding it there for luls [23:46:58] spagewmf: Having Gerrit trouble? :) [23:47:17] y u spy on me :) [23:47:42] ori-l: https://gerrit.wikimedia.org/r/#/c/37341/ thankyouplease [23:48:11] spagewmf: "spy" !== "paying attention in #mediawiki" [23:48:17] marktraceur, it seems if the patch author has comments on someone else's review, the only way to publish them is to click [Review]. [23:48:31] Reedy: === 0 on 415 would be cleaner, tho the net result is the same [23:48:38] spagewmf: I've often thought that it would be useful to have one-click "publish comments" button [23:48:44] Bah [23:48:47] Thought I'd added it [23:48:53] in one place, but not the other [23:49:00] yup [23:50:47] +2'd [23:51:36] thanks