[00:04:32] <MatmaRex>	 hmmm
[00:04:44] <MatmaRex>	 actually, nevermind.
[00:07:31] <gerrit-wm>	 New patchset: Brian Wolff; "Add {{#filemetadata:..}} to insert file metadata on to a wiki page." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/67047
[00:11:17] <gerrit-wm>	 New patchset: Brian Wolff; "Add {{#filemetadata:..}} to insert file metadata on to a wiki page." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/67047
[00:11:55] <gerrit-wm>	 New review: Brian Wolff; "PS10: Change getIndependantMetaArray to getCommonMetaArray at suggestion of Ori." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/67047
[00:18:29] <gerrit-wm>	 New review: Brian Wolff; "(1 comment)" [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/70446
[01:14:31] <gerrit-wm>	 New patchset: Ori.livneh; "Load Puppet-managed settings before user-managed ones; update README" [mediawiki/vagrant] (master) - https://gerrit.wikimedia.org/r/72057
[06:06:18] <gerrit-wm>	 New review: Parent5446; "Responded. Will fix some things once Brian replies." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/65176
[06:19:04] <gerrit-wm>	 New review: Krinkle; "Need to make sure that this is supported on the client side and/or worked around correctly." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/58668
[06:21:24] <gerrit-wm>	 New review: Krinkle; "OK. Let's make it more generic in that case and add a few more at once (not just wgUser)." [mediawiki/tools/codesniffer] (master) - https://gerrit.wikimedia.org/r/69090
[06:23:11] <gerrit-wm>	 New patchset: Santhosh; "Load messages from a fallback chain" [mediawiki/extensions/UniversalLanguageSelector] (master) - https://gerrit.wikimedia.org/r/72065
[06:41:13] <gerrit-wm>	 New review: Tim Starling; "In production testing, it exits after a while with" [mediawiki/core] (master) C: -1;  - https://gerrit.wikimedia.org/r/72067
[07:17:57] <mooey>	 hm is anyone around by any chance with a Firefox 22.0 up for a quick test for me?
[07:19:15] <turkentor>	 what do you need mooey?
[07:20:36] <mooey>	 VisualEditor doesn't appear for me on firefox 22.0 (but does on Chrome) ..
[07:21:17] <mooey>	 if you happen to have firefix 22.0 on, can you check if you have VisualEditor on Wikipedia (or MediaWiki) articles? The "Edit" and "Edit Source" tabs? I only see "Edit" without the 'edit source' and it doesn't load VE...
[07:21:30] <mooey>	 but it *does* work in chrome. I'm wondering if it's some fluke in my local machine or if it's a bug
[07:21:38] <Reedy>	 Check your error console?
[07:21:38] <gerrit-wm>	 New review: Wikinaut; "(1 comment)" [mediawiki/extensions/OpenID] (master) - https://gerrit.wikimedia.org/r/71961
[07:22:07] <mooey>	 hang on, I started restarting it
[07:22:12] * mooey  awaits its reloading

[07:22:49] <mooey>	 hmmmmm ok this is very odd. I do have it working on MediaWiki.org
[07:22:54] <mooey>	 but not on Wikipedia
[07:24:03] <turkentor>	 works for me mooey
[07:24:11] <turkentor>	 on wikipedia
[07:24:15] <turkentor>	 and others
[07:24:21] <turkentor>	 w/ ff22
[07:24:22] <mooey>	 http://en.wikipedia.org/wiki/Circus --> can anyone check if they can start VE on this article inside Firefox 22.0 ?
[07:24:40] <turkentor>	 yeap, works
[07:24:41] <mooey>	 .. that's weird. it's not Firefox, because it DOES work in MediaWiki.org pages
[07:24:47] <mooey>	 meh
[07:25:03] <mooey>	 my console has tons of warnings but none VE-related
[07:25:36] <mooey>	 AH.
[07:25:48] <mooey>	 okay, well, I deleted cache and now it works.
[07:25:57] * mooey  shakes her fist at firefox cache

[07:26:17] <mooey>	 thanks for the help :)
[07:28:24] <turkentor>	 anytime
[07:36:00] <gerrit-wm>	 New patchset: Raimond Spekking; "add prefix linkedwiki- in the i18n files" [mediawiki/extensions/LinkedWiki] (master) - https://gerrit.wikimedia.org/r/71926
[07:38:15] <gerrit-wm>	 New review: Raimond Spekking; "Thanks Karima for prefixing the message keys. I made some further tweaks for consistency with core a..." [mediawiki/extensions/LinkedWiki] (master) - https://gerrit.wikimedia.org/r/71926
[07:48:43] <gerrit-wm>	 New patchset: Parent5446; "changed the remote ext path to Annotator instead of prototyping inline comments" [mediawiki/extensions/Annotator] (master) - https://gerrit.wikimedia.org/r/70801
[07:49:13] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/extensions/Annotator] (master) - https://gerrit.wikimedia.org/r/70801
[07:49:46] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/extensions/Annotator] (master) - https://gerrit.wikimedia.org/r/71278
[09:03:44] <gerrit-wm>	 New review: Raimond Spekking; "PNG: no idea why the conversion script touches the png. but it looks ok." [mediawiki/extensions/LinkedWiki] (master) - https://gerrit.wikimedia.org/r/71926
[09:03:53] <gerrit-wm>	 New patchset: Tim Starling; "Add --list parameter to showJobs.php" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72067
[09:04:36] <gerrit-wm>	 New review: Tim Starling; "Maybe something like that would work. Lightly tested." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72067
[09:33:02] <Nikerabbit>	 hashar: hmm any idea why this hasn't been merged yet: https://gerrit.wikimedia.org/r/#/c/72074/ ?
[09:33:27] <hashar>	 Nikerabbit: i have no clue
[09:33:38] <hashar>	 ask santhosh ? :-)
[09:34:08] <hashar>	 ohh hh
[09:34:12] <hashar>	 damn context
[09:34:26] <hashar>	 Nikerabbit: so yeah hmm you voted +2 on the change but it did not get merged despite the test passing
[09:34:41] <hashar>	 which means there is a bug somewhere, most probably jenkins is not allowed to merge on that repo
[09:35:04] * apergos  looks for AzaThoth to thank him

[09:35:07] <apergos>	 not here...
[09:35:35] <apergos>	 also apparently gitweb redirects aren't happening yet...
[09:36:23] <Nikerabbit>	 hashar: yeah sorry for being vague
[09:37:03] <hashar>	 the stupid zuul gives a stacktrace
[09:37:08] <hashar>	 but does not give Gerrit error message :(
[09:37:18] <hashar>	 I am pretty sure it is a user permission problem
[09:45:12] <Nikerabbit>	 hashar: has there been any chnages to user permissions recently?
[09:46:35] <hashar>	 git fetch refs/meta/config
[09:46:38] <hashar>	 err
[09:49:24] <hashar>	 Nikerabbit: error: rule error: Cannot submit draft patch sets
[09:49:24] <hashar>	 one or more approvals failed; review output above
[09:49:30] <hashar>	 so hmm there is an issue somewhere :-]
[09:49:41] <hashar>	 which is that it is a draft -]
[09:49:50] <hashar>	 that was "hard" to spot hehe
[09:50:11] <Nikerabbit>	 hashar: really?
[09:50:22] <hashar>	 yup
[09:50:33] <hashar>	 I replayed the command in production:  jenkins@gallium:~$ ssh -i /var/lib/jenkins/.ssh/id_rsa -p 29418 jenkins-bot@gerrit.wikimedia.org gerrit review --project mediawiki/extensions/UniversalLanguageSelector --message '"Hapilly merging"'  --verified 2 --submit --code-review 0 72074,3
[09:50:36] <hashar>	 that gave me that message
[09:50:42] <hashar>	 and indeed patchset 3 is a draft
[09:50:49] <gerrit-wm>	 New patchset: Santhosh; "Make anything with class 'uls-settings-trigger' work like ULS trigger" [mediawiki/extensions/UniversalLanguageSelector] (master) - https://gerrit.wikimedia.org/r/72077
[09:50:53] <hashar>	 ideally Zuul should not even attempt to run gate-and-submit on a draft
[09:51:00] <Nikerabbit>	 it is not shown as draft to me!
[09:51:01] <hashar>	 I am not sure it can be configured to do so though
[09:51:09] <Nikerabbit>	 ah there is a small text
[09:52:43] <gerrit-wm>	 New patchset: Amire80; "Allow disabling of IME by a jQuery selector" [mediawiki/extensions/UniversalLanguageSelector] (master) - https://gerrit.wikimedia.org/r/68339
[09:55:59] <Nikerabbit>	 I'll ask Pau
[10:04:55] <gerrit-wm>	 New patchset: Santhosh; "jquery.i18n messagestore for ULS" [mediawiki/extensions/UniversalLanguageSelector] (master) - https://gerrit.wikimedia.org/r/71990
[10:08:14] <gerrit-wm>	 New review: Pginer; "The outline for the cog should be only visible when accessed using the keyboard. This can be achieve..." [mediawiki/extensions/UniversalLanguageSelector] (master) - https://gerrit.wikimedia.org/r/71784
[10:30:22] <gerrit-wm>	 New patchset: Amire80; "Allow disabling of IME by a jQuery selector" [mediawiki/extensions/UniversalLanguageSelector] (master) - https://gerrit.wikimedia.org/r/68339
[10:33:21] <Nemo_bis>	 apergos: are you able to review apache rewrite patches then? there's one for gitweb in review :)
[10:34:29] <apergos>	 it's pretty random on how useful I am for review of a given patch
[10:44:53] <Nemo_bis>	 apergos: https://gerrit.wikimedia.org/r/#/c/70370/
[10:46:10] <apergos>	 I admit I would need to setup and do testing for this before I could sign off
[10:46:35] <apergos>	 that can't happen today (I am behind on stuff I want done so I can run en wp dumps out of eqiad)
[10:54:07] <gerrit-wm>	 New review: Nemo bis; "That was filed as bug 50793" [mediawiki/extensions/UniversalLanguageSelector] (master) - https://gerrit.wikimedia.org/r/71784
[10:59:05] <MatmaRex>	 @notify James_F
[10:59:05] <wm-bot>	 I'll let you know when I see James_F around here
[11:05:27] <gerrit-wm>	 New patchset: Matmarex; "Provide month names via mediawiki.language.months RL module" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/36391
[11:28:15] <gerrit-wm>	 New review: Amire80; "(1 comment)" [mediawiki/extensions/UniversalLanguageSelector] (master) - https://gerrit.wikimedia.org/r/72065
[11:38:24] <gerrit-wm>	 New review: Santhosh; "(1 comment)" [mediawiki/extensions/UniversalLanguageSelector] (master) - https://gerrit.wikimedia.org/r/72065
[11:42:41] <qchris>	 hashar: About draft status of patch sets and running gate-and-submit jobs. I also think that it's not (easily) possible to detect a patch set's (!= change's) draft status for Jenkins
[11:42:53] <qchris>	 hashar: Would we want something like that?
[11:43:24] <qchris>	 Oh ... hashar's no longer here :-)
[11:49:51] <gerrit-wm>	 New review: Matmarex; "(1 comment)" [mediawiki/core] (master) C: -1;  - https://gerrit.wikimedia.org/r/60952
[12:07:24] <gerrit-wm>	 New review: Matmarex; "I assume that mw.ui is supposed to require a CSS3-capable browser. (For the :not() selector used here.)" [mediawiki/core] (master) C: 2;  - https://gerrit.wikimedia.org/r/66044
[12:09:53] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/66044
[12:10:44] <gerrit-wm>	 New review: Jpond; "Re: Comments." [mediawiki/extensions/OpenID] (master) - https://gerrit.wikimedia.org/r/71961
[12:16:23] <gerrit-wm>	 New patchset: Siebrand; "jquery.i18n message store for ULS" [mediawiki/extensions/UniversalLanguageSelector] (master) - https://gerrit.wikimedia.org/r/71990
[12:36:30] <gerrit-wm>	 New review: Yuvipanda; "I had fixed this but apparently failed to push." [apps/android/commons] (master); V: 2 C: 2;  - https://gerrit.wikimedia.org/r/71866
[12:36:31] <gerrit-wm>	 Change merged: Yuvipanda; [apps/android/commons] (master) - https://gerrit.wikimedia.org/r/71866
[13:35:26] <gerrit-wm>	 New patchset: Hashar; "tie qunit jobs to slaves" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/72083
[13:36:28] <gerrit-wm>	 New patchset: Hashar; "Jenkins job validation (DO NOT SUBMIT)" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/70646
[13:37:31] <gerrit-wm>	 New patchset: Nikerabbit; "Allow disabling of IME by a jQuery selector" [mediawiki/extensions/UniversalLanguageSelector] (master) - https://gerrit.wikimedia.org/r/68339
[13:38:04] <gerrit-wm>	 New review: Nikerabbit; "Tested in Chrome/Linux with custom rule "input"." [mediawiki/extensions/UniversalLanguageSelector] (master) C: 2;  - https://gerrit.wikimedia.org/r/68339
[13:38:41] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/extensions/UniversalLanguageSelector] (master) - https://gerrit.wikimedia.org/r/68339
[13:39:30] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/72083
[13:43:08] <gerrit-wm>	 New patchset: Rjain; "Added the create class" [mediawiki/extensions/Annotator] (master) - https://gerrit.wikimedia.org/r/71364
[13:49:47] <gerrit-wm>	 New patchset: Santhosh; "Apply webfonts to the IME selector too" [mediawiki/extensions/UniversalLanguageSelector] (master) - https://gerrit.wikimedia.org/r/72087
[13:55:13] <gerrit-wm>	 New patchset: Shirayuki; "Fix the capitalization" [mediawiki/extensions/EducationProgram] (master) - https://gerrit.wikimedia.org/r/72088
[14:13:32] <gerrit-wm>	 New patchset: Anomie; "Improve disable_functions handling in LuaStandalone" [mediawiki/extensions/Scribunto] (master) - https://gerrit.wikimedia.org/r/72089
[14:25:43] <gerrit-wm>	 New patchset: Santhosh; "Update jquery.ime from upstream" [mediawiki/extensions/UniversalLanguageSelector] (master) - https://gerrit.wikimedia.org/r/72093
[14:26:26] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/71272
[14:29:24] <gerrit-wm>	 New patchset: Aude; "Add Wikivoyage site link group message for Wikidata" [mediawiki/extensions/WikimediaMessages] (master) - https://gerrit.wikimedia.org/r/72095
[14:40:57] <James_F>	 matanya: Ping?
[14:41:02] <James_F>	 Gah.
[14:41:05] <James_F>	 Sorry, mis-ping.
[14:41:57] <YuviPanda>	 clearly, IRC is useless for collaboration, and we should all use Facebook Chat
[14:47:48] <gerrit-wm>	 New review: Anomie; "Code seems to work in a quick test. I'm not too sure about the specific error response though." [mediawiki/extensions/AbuseFilter] (master) - https://gerrit.wikimedia.org/r/71945
[15:02:56] <thedj[work]>	 YuviPanda: no.... the Facebookification of chat dze horror
[15:11:25] <gerrit-wm>	 New patchset: Cmcmahon; "Turn off VE tests for IEs < 9" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/72100
[15:13:58] <gerrit-wm>	 New patchset: Cmcmahon; "Turn off VE tests for IEs < 9" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/72100
[15:14:53] <gerrit-wm>	 Change merged: Zfilipin; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/72100
[15:23:13] <wmf-selenium-bot>	 Yippie, build fixed!
[15:23:14] <wmf-selenium-bot>	 Project _debug-browsertests-template build #188: FIXED in 1 min 7 sec: https://wmf.ci.cloudbees.com/job/_debug-browsertests-template/188/
[15:29:51] <gerrit-wm>	 New review: CSteipp; "(1 comment)" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/71997
[15:29:58] <gerrit-wm>	 New patchset: Orsagi; "opening" [apps/android/translate] (master) - https://gerrit.wikimedia.org/r/72103
[15:31:14] <bawolff>	 csteipp: thanks for looking at it :)
[15:31:30] <csteipp>	 bawolff: no problem!
[15:32:08] <bawolff>	 The real thing I want feedback on that patch though is if people thing the random selection method is actually random enough
[15:32:18] <bawolff>	 but no one has ventured an opinion on that yet
[15:32:41] * bawolff  should probably do some stats to try and justify himself

[15:36:00] <gerrit-wm>	 New review: Anomie; "(1 comment)" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72064
[15:42:43] <gerrit-wm>	 Change merged: Orsagi; [apps/android/translate] (master) - https://gerrit.wikimedia.org/r/72103
[15:46:32] <Nemo_bis>	 bawolff: oh, that would be immensely useful
[15:46:54] <Nemo_bis>	 especially when TS gets killed and we lose erwin's tool
[15:47:41] <bawolff>	 The downside is, that the method I use is of questionable randomness
[15:47:52] <bawolff>	 but I feel its better than nothing
[15:50:45] <bawolff>	 Nemo_bis: In regards to bug 46918, have you seen http://www.wikihow.com/Special:CommunityDashboard
[15:55:25] <Nemo_bis>	 bawolff: commented on bug https://bugzilla.wikimedia.org/show_bug.cgi?id=25931#c40
[15:56:08] <bawolff>	 yeah, that's pretty much correct
[15:56:28] <YuviPanda>	 bawolff: extract the randomness stuff from that patch and run on toollabs?
[15:57:01] <YuviPanda>	 (or the internal analytics slaves, if you want EXPLAIN)
[15:57:02] <bawolff>	 The worst case would be if you have some sort of maintenance category, where 1000 articles were added on one ay in 2007, and 1 article was added in 2013
[15:57:19] <bawolff>	 99% of the time it would go for the 2013 article
[15:57:38] <bawolff>	 Well maybe not, since it fudges it a bit, but would still be highly biased towards that one
[15:57:41] <YuviPanda>	 if you do the 'flip > and < randomly' thing?
[15:58:14] <bawolff>	 It would make it so it hit the 2013 article about 50% of the time
[15:58:22] <YuviPanda>	 better than 99% :)
[15:58:27] <bawolff>	 true
[15:58:59] <gerrit-wm>	 New patchset: Zfilipin; "Use temp folder to manage temporary test images" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/71516
[15:59:15] * bawolff  does not have access to "internal analytics slaves". OTOH I have pretty good idea that performance wise the patch is fine, the open question is if it gives good enough answers

[15:59:42] <Nemo_bis>	 merge merge merge
[16:00:08] <YuviPanda>	 bawolff: perhaps call it 'pseudorandom' or 'time based random' or something like that?
[16:00:10] <Nemo_bis>	 bawolff: no, never saw that dashboard before, it seems to want me login :)
[16:00:14] <YuviPanda>	 to set expectations?
[16:00:27] <Nemo_bis>	 isn't everything pseudorandom
[16:00:31] <gerrit-wm>	 New review: Zfilipin; "Chris and I paired on this. I have changed the commit so it saves the image to a temp folder. We hav..." [qa/browsertests] (master) C: 2;  - https://gerrit.wikimedia.org/r/71516
[16:00:34] <gerrit-wm>	 Change merged: Zfilipin; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/71516
[16:00:42] <bawolff>	 special:ArbitraryMemberOfCategoryThatIsNotReallyRandom
[16:00:51] <YuviPanda>	 legit
[16:00:52] <bawolff>	 I put a warning on the form view
[16:01:05] * bawolff  worries about making the page name too long

[16:05:06] <parent5446>	 Just merge it. There's no way to achieve better randomness without making a separate table to manage the distribution.
[16:05:56] * bawolff  tends to agree

[16:06:05] <gerrit-wm>	 New review: coren; "I'm no expert, but this appears to be correct." [integration/jenkins-job-builder-config] (master) C: 1;  - https://gerrit.wikimedia.org/r/71245
[16:06:37] * bawolff  goes and does the flip direction thingy

[16:06:55] * YuviPanda  also tends to agree

[16:07:26] <YuviPanda>	 bawolff: plus, if it is used for 'backlog clearing' your '1000 things in 2007, 1 thing in 2013' situation should become better as soon as the 2013 thing is removed from teh category
[16:07:40] <bawolff>	 lol, true
[16:07:49] <YuviPanda>	 it will be  a problem for 'random browsing' but not backlog clearing
[16:08:12] <YuviPanda>	 it might even be desirable behavior for backlogs
[16:14:41] <gerrit-wm>	 New review: Anomie; "Code seems sane, works in a quick test with TimedMediaHandler. Two minor comments, only the one on l..." [mediawiki/core] (master) C: -1;  - https://gerrit.wikimedia.org/r/60858
[16:22:15] <Nemo_bis>	 YuviPanda: the problem is if the 2013 item is particularly hard to solve
[16:22:36] <YuviPanda>	 all the better to throw it at more people :P
[16:23:55] <bawolff>	 Meh, if the users don't like their new toy, we could always take it away from them :P
[16:24:02] <YuviPanda>	 :P
[16:24:19] <YuviPanda>	 you give them something they hate you. you take away something they hate you..
[16:24:21] <gerrit-wm>	 New patchset: Brian Wolff; "Add Special:RandomInCategory." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/71997
[16:24:41] <bawolff>	 You sit there and do nothing, they hate you
[16:24:51] <bawolff>	 You really can't win with enwiki
[16:25:04] <YuviPanda>	 heh :)
[16:26:13] <bawolff>	 meh, I did that wrong
[16:26:33] <bawolff>	 no wait, i didn't
[16:29:08] <sumanah>	 bawolff: so there's a bit in the Bhagavad Gita that is relevant
[16:29:28] * bawolff  actually did probably do it wrong...

[16:29:34] <YuviPanda>	 'do or do not, there is no try?'
[16:29:36] <YuviPanda>	 oh wait
[16:29:38] <sumanah>	 where Krishna reminds Arjuna that we must live according to our duty and learn not to desire the fruit (outcome) of our actions
[16:30:32] <Nemo_bis>	 right, that's how I survived so far in Wikimedia
[16:30:35] <Nemo_bis>	 never hope anything
[16:30:39] * bawolff  was all expecting a part of the Bhagavad Gita that applied to db optimization :)

[16:31:37] <sumanah>	 bawolff: I would not be surprised if there's something about that in the Mahabharata as a whole. there is a LOT in the Mahabharata
[16:32:12] <sumanah>	 basically any time some random person made up a tale they could say "this is a story one Pandava told another Pandava during the 12 years of exile"
[16:32:19] <sumanah>	 bam
[16:32:29] <sumanah>	 it's worse than the Marvel continuity
[16:32:34] <gerrit-wm>	 New review: MarkTraceur; "c.f. bug 49988" [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/71581
[16:32:44] <YuviPanda>	 haha
[16:32:48] <bawolff>	 lol
[16:32:50] <gerrit-wm>	 New patchset: Orsagi; "added .gitreview file" [apps/android/translate] (master) - https://gerrit.wikimedia.org/r/72109
[16:33:25] <YuviPanda>	 UploadWizard: I didn't know you could type 'bug <number>' on a gerrit comment and it would link!
[16:33:29] <marktraceur>	 Hah, Krishna said "doodie"
[16:33:33] <marktraceur>	 damn it YuviPanda
[16:33:50] <YuviPanda>	 :D
[16:34:17] <sumanah>	 I find I have to re-remind myself of that particular lesson from the Gita every once in a while
[16:34:27] <gerrit-wm>	 Change merged: Orsagi; [apps/android/translate] (master) - https://gerrit.wikimedia.org/r/72109
[16:34:37] <sumanah>	 both parts are important: about doing my duty faithfully, and about renouncing the fruit of my action
[16:34:54] <YuviPanda>	 congratulations, orsa :)
[16:35:05] <orsa>	 thanks.. :)
[16:35:30] <Nemo_bis>	 sumanah: sounds like cc-by-sa
[16:35:42] <bawolff>	 I generally agree
[16:35:46] <sumanah>	 however I am not such a huge fan of the casteism, or of spending 18 days battling at Kurukshetra trying to defeat my cousins
[16:36:05] <YuviPanda>	 ^ +\infinity
[16:36:22] <sumanah>	 Nemo_bis: I think you probably see things differently from me but this probably isn't the best medium for hashing it out :)
[16:36:29] <orsa>	 YuviPanda: btw i cant see our repo in https://git.wikimedia.org/project/apps
[16:36:30] * YuviPanda  finds it hard to pick and choose 'good' things from a sea of horrible things / mediocre things

[16:37:16] <YuviPanda>	 orsa: https://git.wikimedia.org/summary/?r=apps/android/translate.git
[16:37:35] <bawolff>	 Fun fact: Guess what the smallest category on enwikipedia is?
[16:37:51] <orsa>	 then it is not indexed
[16:37:52] <bawolff>	 Answer: Welsh Americans, with a whopping -338 pages in it
[16:37:56] <sumanah>	 but anyway - we gotta do our duty, act with compassion and rigor, act according to our values, including outreach to users. and then renounce the fruit
[16:38:07] <YuviPanda>	 orsa: should fix itself soon, I suppose
[16:38:13] <bawolff>	 (that's 338 pages less then an empty category)
[16:38:17] <orsa>	 alrigtht
[16:38:17] <YuviPanda>	 bawolff: ... I... wasn't expecting that
[16:38:32] * YuviPanda  waits for someone to write an editorial about how enwiki is hostile to welsh americans

[16:39:00] <bawolff>	 I guess race conditions when updating functionally dependent db fields are actually important ;)
[16:39:13] <sumanah>	 ;)
[16:46:25] <gerrit-wm>	 New review: Brion VIBBER; "Needs to document dependencies:" [mediawiki/extensions/UploadWizard] (master) C: -1;  - https://gerrit.wikimedia.org/r/70446
[16:47:09] <parent5446>	 Are there any further objections to the RandomInCategory merge?
[16:48:23] <gerrit-wm>	 New review: Brion VIBBER; "Sample data can't be saved." [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/70446
[16:50:04] <gerrit-wm>	 New patchset: Brian Wolff; "Add Special:RandomInCategory." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/71997
[16:50:26] <bawolff>	 Not with my next commit
[16:51:12] <gerrit-wm>	 New review: Brian Wolff; "PS5: If there's a big gap, favour both sides of the gap." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/71997
[16:51:33] * bawolff  would now be happy to see it merged

[16:52:22] <gerrit-wm>	 New review: Anomie; "(1 comment)" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/71997
[16:53:04] <gerrit-wm>	 New patchset: SuchABot; "Redo campaigns backend to use ContentHandler" [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/70446
[16:55:03] <YuviPanda>	 Is there a standard way to detect if an extension has loaded?
[16:55:14] <YuviPanda>	 other than gross hacks like checking for variables it defines?
[16:55:50] <bawolff>	 no...
[16:55:52] <parent5446>	 PHP extension or MW extension?
[16:55:56] <YuviPanda>	 mw
[16:56:03] <parent5446>	 Yeah, no...
[16:56:06] <parent5446>	 :P
[16:56:13] <gerrit-wm>	 New review: CSteipp; "Allowing these tags in the first k of a file will lead many browsers to determine that these files a..." [mediawiki/core] (master) C: -1;  - https://gerrit.wikimedia.org/r/71991
[16:56:56] <marktraceur>	 YuviPanda: What's the state of that patch? Should I review again?
[16:59:47] <YuviPanda>	 marktraceur: brion is reviewing it right now, and it now has schema validation. And yes, I would like you to review it again so it can be merged :)
[17:02:47] <marktraceur>	 YuviPanda: Can has maintenance scripts? :D
[17:03:03] <YuviPanda>	 marktraceur: do you really really want it to be in the same commit?
[17:03:05] <YuviPanda>	 I'm working on it now
[17:03:10] <YuviPanda>	 but would ideally be a separate commit
[17:03:12] <YuviPanda>	 but if you insist...
[17:03:27] <YuviPanda>	 I'll add it to this one
[17:03:47] <YuviPanda>	 can you assume it exists for now and review it? I am okay with it being merged only after a maintenance script, as long as I get CR for the rest of it now
[17:04:06] <marktraceur>	 No, I just want it to exist and work so I can make sure Commons won't break
[17:04:08] <parent5446>	 Just curious, what patch are you guys talking about?
[17:04:17] <marktraceur>	 I _really_ don't want Commons to break.
[17:04:30] <marktraceur>	 parent5446: 70446
[17:04:45] <gerrit-wm>	 New patchset: Aaron Schulz; "Fixed return type of getContentNamespaces()" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72111
[17:04:56] <parent5446>	 Thanks
[17:04:58] <YuviPanda>	 marktraceur: me neither
[17:06:21] <brion>	 YuviPanda: maint script is like one file, so it's isolated and doesn't really make the commit any uglier to review
[17:06:57] <YuviPanda>	 brion: true, but ideally I'd have wanted even this to be split across 4-5 patches. Either way, will just add it to this one now.
[17:08:41] <brion>	 hmmm
[17:08:55] <brion>	 http://hawkeye.local/core/index.php/Special:UploadWizard?campaign=Sample_campaign <- works
[17:08:58] <brion>	 http://hawkeye.local/core/index.php/Special:UploadWizard?campaign=sample_campaign <- doesn't
[17:09:06] <brion>	 "there is no campaign with name "sample_campaign""
[17:09:32] <brion>	 current campaign names appear to all have lowercase names except LandtagsprojectAT
[17:09:50] <gerrit-wm>	 New review: Parent5446; "Funny we actually have a unit test for an incorrect return type." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72111
[17:11:28] <gerrit-wm>	 New review: Brion VIBBER; "Campaign names appear to be case-sensitive for the first letter. If naming a campaign "Sample campai..." [mediawiki/extensions/UploadWizard] (master) C: -1;  - https://gerrit.wikimedia.org/r/70446
[17:12:00] <marktraceur>	 brion: Shouldn't this be tied to whether the campaign namespace is case-sensitive? I think there's a wg for that...
[17:12:13] <brion>	 marktraceur: well it's case-insensitive by default
[17:12:17] <brion>	 so it should, you know, work
[17:12:28] <brion>	 the default of looking up a title via a string should already take care of that
[17:12:38] <brion>	 i wonder if it's loading things in a weird way
[17:13:11] <marktraceur>	 wgCaseInsensitive
[17:13:56] <ori-l>	 not to be confused with wgcaseinsensitive
[17:13:58] <marktraceur>	 And wgCapitalLinkOverrides
[17:14:13] <marktraceur>	 ori-l: Well, you have to enable the former to use the latter
[17:14:30] <marktraceur>	 Er, I'm sorry
[17:14:34] <marktraceur>	 wgCapitalLinks
[17:14:42] <brion>	  +    $success = $dbw->upsert(
[17:14:45] <marktraceur>	 I dunno what I'm smoking
[17:14:49] <brion>	 upsert? ugh when'd we add that :P
[17:15:00] * Nemo_bis  read something about capitalism and success

[17:15:26] <parent5446>	 Was added a couple of weeks ago I believe.
[17:15:37] <parent5446>	 Now we can finally use ON DUPLICATE KEY UPDATE.
[17:15:38] <YuviPanda>	 brion: marktraceur I guess it should have teh same rules as the page casing
[17:16:15] <YuviPanda>	 brion: yeah, it is 1.22 :)
[17:19:13] <bawolff>	 anomie: I still get Notice: Undefined index on php 5 for undefined array indexes
[17:19:31] <parent5446>	 csteipp: Are you still working on 70747? Because the Annotator GSoC extension is going to need it.
[17:19:52] <YuviPanda>	 marktraceur: brion not sure if campaigns should be case insensitive tho.
[17:20:14] <brion>	 YuviPanda: well existing campaigns are lowercase except one, so existing links are likely ?campaign=wlm-at etc
[17:20:15] <csteipp>	 parent5446: Yep, I need it
[17:20:21] <parent5446>	 OK awesome. Just checking.
[17:20:29] <brion>	 if that suddenly breaks, then people have to go around changing their links
[17:20:42] <YuviPanda>	 brion: hmm, so migration script will need to make it to Campaign:wlm-at
[17:20:44] <YuviPanda>	 and that should work
[17:20:45] <YuviPanda>	 got it
[17:21:00] <brion>	 YuviPanda: well… that or let either way work :)
[17:21:00] <anomie>	 bawolff: Weird, I don't.
[17:21:19] <YuviPanda>	 brion: true but the end goal is 'do not break incoming links!'
[17:21:31] <YuviPanda>	 which I heartily agree on
[17:21:34] <brion>	 right
[17:21:48] <brion>	 so letting campaign=wlm work even if it's "become" Wlm would resolve that
[17:21:51] <bawolff>	 anomie: The docs say "Attempting to access an array key which has not been defined is the same as accessing any other undefined variable: an E_NOTICE-level error message will be issued, and the result will be NULL. "
[17:22:04] <anomie>	 bawolff: But inside unset()?
[17:22:12] <brion>	 ok so uw_campaign_conf table is gone, but uw_campaigns remains?
[17:22:17] <YuviPanda>	 brion: indeed.
[17:22:30] <anomie>	 bawolff: (of course it does for the general case)
[17:22:38] <brion>	 not sure what uw_campaigns gets us that page table doesn't
[17:22:40] <YuviPanda>	 brion: uw_campaigns will host 'queryable' columns, so that we don't have to load the entire json to query for things
[17:22:44] <brion>	 is it just the campaign_enabled option?
[17:22:44] <bawolff>	 oh, sorry, I thought you were talking about the general case
[17:22:45] <YuviPanda>	 brion: such as 'enabled' status
[17:22:53] <brion>	 mm ok
[17:23:00] <YuviPanda>	 brion: and more can be added later (geo-coordinates, start/end dates, etc)
[17:23:41] <bawolff>	 anomie: you're right, no warning, I'll change the code
[17:23:57] <brion>	 so campaign_id will become obsolete, no longer used for anything? or will it still be used for new tables potentially?
[17:24:00] <gerrit-wm>	 New review: Aaron Schulz; "Here is some "difficult" code to work around exceptions (tested in eval.php by sabotaging the redis ..." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72067
[17:24:10] <YuviPanda>	 brion: I don't see it being used for anything at all, at least not publicly.
[17:24:12] <brion>	 if the main data's in the page, which is linked via name
[17:24:14] <brion>	 ok
[17:24:35] <bawolff>	 anomie: Sorry, its been a while since I looked at this patch, and I didn't even look at what the line after the if statement was doing
[17:26:11] <gerrit-wm>	 New patchset: Brian Wolff; "Allow handler specific parameters in <gallery> (page number, etc)" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/60858
[17:26:23] <gerrit-wm>	 New review: Yuvipanda; "Sample data now works with updated patchset." [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/70446
[17:29:24] <gerrit-wm>	 New patchset: Brian Wolff; "Allow handler specific parameters in <gallery> (page number, etc)" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/60858
[17:29:47] <gerrit-wm>	 New review: Brian Wolff; "PS2: per anomie" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/60858
[17:29:48] <gerrit-wm>	 New patchset: Rtdwivedi; "Added unit tests for Page pages." [mediawiki/extensions/ProofreadPage] (pagePagesRefactoring) - https://gerrit.wikimedia.org/r/71172
[17:32:48] <marktraceur>	 YuviPanda: So my initial question was intended to gauge whether, in your opinion, if I reviewed it now I would want to merge it or if you think there are still issues to fix
[17:33:18] <YuviPanda>	 brion has identified issues to fix and I'm fixing them now
[17:33:29] <marktraceur>	 Right, so I will hack on something else.
[17:33:33] <YuviPanda>	 marktraceur: but yeah, if you review it and I fix the things that come up in review, I'd want to merge it.
[17:33:35] <bawolff>	 anomie: Thanks :)
[17:33:42] <anomie>	 bawolff: np
[17:33:51] <YuviPanda>	 marktraceur: they're minor issues, so I'll poke you again when I've fixed them :P it's like a 3-4 line patch...
[17:33:56] <marktraceur>	 *nod*
[17:34:44] <yurik>	 anomie, around?
[17:34:55] <anomie>	 yurik: busy right now
[17:35:39] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/60858
[17:36:01] <gerrit-wm>	 New review: Brion VIBBER; "Several inline comments above." [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/70446
[17:36:06] <brion>	 added some more comments
[17:36:10] <YuviPanda>	 looking
[17:36:29] <YuviPanda>	 grr Gerrit
[17:37:56] <YuviPanda>	 brion: re: fetching from table than Campaign: namespace, the rationale was that it is faster to check if it is enabled that way.
[17:38:04] <YuviPanda>	 brion: unsure if that is a tradeoff worth making. Thoughts?
[17:38:17] <brion>	 yeah that doesn't feel like a critical path
[17:38:22] <brion>	 if it is enabled you're going to load the data anyway
[17:38:30] <brion>	 if it isn't, you've hit the rare case of following a stale campaign link
[17:38:31] <YuviPanda>	 right
[17:38:47] <gerrit-wm>	 New review: Brian Wolff; "subtle bump... :)" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/59974
[17:38:50] <YuviPanda>	 brion: I had forgotten the id1, id2 url things. let me fix
[17:39:04] <yurik>	 anomie, ok, ping me when you have time, i had some more ideas and wanted to discuss
[17:39:17] <gerrit-wm>	 New review: Aaron Schulz; "That said, ideally the iterators would distinguish not finding the ID (e.g. the race condition) but ..." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72067
[17:39:23] <YuviPanda>	 brion: does the rest of the code look okay to you? I'm fixing the stuff you pointed out
[17:39:39] <brion>	 overall nothing that's freaking me out just looking at it :)
[17:39:47] <brion>	 but i'll put it through its paces a bit more and see :D
[17:40:08] <YuviPanda>	 brion: :)
[17:40:23] <YuviPanda>	 brion: there is, for one, absolutely no explicit caching anywhere. is thjat... okay/
[17:40:23] <YuviPanda>	 ?
[17:40:42] <brion>	 page data is cached, you're using regular channels to load it, i think that's fine :D
[17:41:19] <YuviPanda>	 ah, that's what I was trying to figure out
[17:41:28] <brion>	 yeah, the revision data anyway
[17:43:20] <YuviPanda>	 marktraceur: offtopic, but I remember your feelings against async: false in code. https://bugzilla.wikimedia.org/show_bug.cgi?id=49935
[17:44:42] <marktraceur>	 Oh my god.
[17:44:46] <YuviPanda>	 yeah
[17:44:49] <marktraceur>	 Haven't even looked at the bug yet
[17:44:52] <marktraceur>	 And I'm already scared
[17:44:57] <gerrit-wm>	 New review: CSteipp; "(1 comment)" [mediawiki/extensions/OAuth] (master) C: 2;  - https://gerrit.wikimedia.org/r/71156
[17:45:04] <RoanKattouw>	 I just heard about it and I headdesked
[17:45:06] <YuviPanda>	 marktraceur: async: false load everytime every wiki page loads,apparently.
[17:45:21] * marktraceur  hugs MatmaRex

[17:45:39] <marktraceur>	 Dumb dumb dumb.
[17:46:17] <marktraceur>	 I can vouch for "maybe rewriting half the library", I had to do that for Nischayn22's patch
[17:46:28] <YuviPanda>	 I did that for the old Wikipedia app, too.
[17:46:37] <YuviPanda>	 (+ vouch for that)
[17:48:27] <gerrit-wm>	 New review: Nilesh; "Sorry about the misleading commit. You can see the deleted part. The added part went in https://gerr..." [mediawiki/extensions/WikidataEntitySuggester] (master) - https://gerrit.wikimedia.org/r/71772
[17:48:55] <gerrit-wm>	 New review: Tpt; "Some points for the regexs." [mediawiki/extensions/ProofreadPage] (pagePagesRefactoring) C: -1;  - https://gerrit.wikimedia.org/r/71172
[17:58:23] <gerrit-wm>	 New patchset: Rjain; "Added the create class" [mediawiki/extensions/Annotator] (master) - https://gerrit.wikimedia.org/r/71364
[18:05:58] <gerrit-wm>	 New patchset: Nilesh; "Added prototype PHP client" [mediawiki/extensions/WikidataEntitySuggester] (master) - https://gerrit.wikimedia.org/r/72116
[18:07:28] <gerrit-wm>	 New patchset: Siebrand; "Consistency update in group name" [mediawiki/extensions/Comments] (master) - https://gerrit.wikimedia.org/r/72117
[18:08:06] <gerrit-wm>	 Change merged: CSteipp; [mediawiki/extensions/OAuth] (master) - https://gerrit.wikimedia.org/r/71156
[18:16:44] <thedj_>	 MatmaRex: yikes
[18:17:45] <thedj_>	 async:false is never a good idea.
[18:20:17] <gerrit-wm>	 New review: Brian Wolff; "re" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/65176
[18:22:46] <MatmaRex>	 tell me aobut it, eh.
[18:35:34] <gerrit-wm>	 New patchset: Siebrand; "Localisation updates from http://translatewiki.net." [mediawiki/extensions/WikimediaMessages] (master) - https://gerrit.wikimedia.org/r/72120
[18:47:17] * YuviPanda  grumbles about the php doc's search being unusable

[18:48:10] <Nemo_bis>	 mal comune mezzo gaudio
[18:49:09] <liangent>	 do we have some good function to escape summary?
[18:52:21] <YuviPanda>	 if I get a WikiPage object, and not 'null', can I assume that the page exists?
[18:52:25] <YuviPanda>	 or should I still explicitly check for it?
[18:53:03] <brion>	 YuviPanda: well there's an exists() method on WikiPage
[18:53:10] <YuviPanda>	 and on Title too
[18:53:12] <brion>	 so i suppose it depends how you got it
[18:53:23] <YuviPanda>	 ::factory
[18:53:26] <YuviPanda>	 which... doesn't say anything
[18:53:32] <YuviPanda>	 neither does Title::makeTitleSafe
[18:54:15] <brion>	 YuviPanda: yeah looks like it'll give you a page object as long as the title's valid
[18:54:18] <brion>	 so…. check exists () :D
[18:54:24] <YuviPanda>	 yeah
[18:54:27] <YuviPanda>	 did that now
[18:58:44] <YuviPanda>	 I... clearly need a new machine
[18:59:24] * brion  emails YuviPanda some more RAM

[18:59:36] * YuviPanda  downloads the RAM, asks for a better CPU

[19:00:18] <brion>	 a core 2 duo ought be enough for anyone
[19:01:48] <gerrit-wm>	 New patchset: Siebrand; "Update autonym/interlanguage description for Hakka" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72126
[19:01:56] <gerrit-wm>	 New review: Brian Wolff; "An alternative approach could be to add X amount of whitespace to the top of any xml file that fails..." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/71991
[19:02:09] <apergos>	 oohhh
[19:02:11] <apergos>	 more ram?
[19:02:22] <apergos>	 8gb turns out that with gnome-shell it is *not* enough
[19:02:24] <apergos>	 stupid thing
[19:03:15] <gerrit-wm>	 New patchset: SuchABot; "Redo campaigns backend to use ContentHandler" [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/70446
[19:04:16] <gerrit-wm>	 New review: Parent5446; "Needs a rebase." [mediawiki/core] (master) C: -1;  - https://gerrit.wikimedia.org/r/59974
[19:06:02] <YuviPanda>	 brion: fixed most of the things you mentioned, now just the required and idField things
[19:06:08] <gerrit-wm>	 New patchset: Brian Wolff; "Do not allow sorting Special:Listfiles by img_name when filtering by username" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/59974
[19:06:10] <brion>	 \o/
[19:06:24] <gerrit-wm>	 New review: Brian Wolff; "PS7: rebase" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/59974
[19:06:47] <gerrit-wm>	 New patchset: SuchABot; "Redo campaigns backend to use ContentHandler" [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/70446
[19:06:56] <YuviPanda>	 brion: also kicked out some redundant sql update files with it
[19:07:00] <YuviPanda>	 more deletions!
[19:07:03] <brion>	 \o/
[19:17:34] * legoktm  pokes csteipp about https://gerrit.wikimedia.org/r/#/c/48608/ :)

[19:19:19] <csteipp>	 Ah, that one... :)
[19:28:27] <gerrit-wm>	 New review: CSteipp; "Works in my testing, and seems sane. thanks!" [mediawiki/extensions/GlobalBlocking] (master) C: 2;  - https://gerrit.wikimedia.org/r/48608
[19:28:35] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/extensions/GlobalBlocking] (master) - https://gerrit.wikimedia.org/r/48608
[19:28:55] <MatmaRex>	 well
[19:28:59] <gerrit-wm>	 New patchset: Siebrand; "Localisation updates from http://translatewiki.net." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72131
[19:29:06] <MatmaRex>	 it turns out that Opera doesn't support [].slice correctly.
[19:29:10] <MatmaRex>	 https://bugzilla.wikimedia.org/show_bug.cgi?id=50385#c6
[19:29:14] <MatmaRex>	 isn't it just amazing
[19:29:24] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/59974
[19:41:22] <MatmaRex>	 hey, it's superm401.
[19:41:39] <MatmaRex>	 superm401: how about some trivial code review? https://gerrit.wikimedia.org/r/#/c/71795/ and its dependency are waiting :)
[19:41:49] <superm401>	 Hey MatmaRex
[19:43:07] <gerrit-wm>	 New review: Tpt; "Currently no browser supports x3d out of the box so, to allow view of x3d file, I'm writing an exten..." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/71991
[19:43:16] <gerrit-wm>	 New patchset: Mwjames; "Extend \SMW\Test\MockObjectBuilder" [mediawiki/extensions/SemanticMediaWiki] (master) - https://gerrit.wikimedia.org/r/72132
[19:46:22] <superm401>	 MatmaRex, were you able to confirm what they said about the old code working in debug mode?
[19:47:55] <MatmaRex>	 superm401: hm, i didn't check that, actually
[19:48:08] <MatmaRex>	 but idid confirm that regular mode didn't work correctly on IE9
[19:48:20] <MatmaRex>	 and apparently the person confirms my patch fixes it (see bug)
[19:48:33] <MatmaRex>	 which is weird, but oh well, browser bugs
[19:50:39] <MatmaRex>	 superm401: i just checked and i'm confirming.
[19:50:48] <MatmaRex>	 works with &debug=1. funny stuff, eh.
[19:51:29] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/extensions/SemanticMediaWiki] (master) - https://gerrit.wikimedia.org/r/72132
[19:52:05] <superm401>	 MatmaRex, hmm, then there's a ResourceLoader bug.
[19:52:44] <MatmaRex>	 superm401: nah, i'm fairly sure it's a IE bug
[19:52:50] <MatmaRex>	 especially since it works on other browsers
[19:52:56] <superm401>	 MatmaRex, it shouldn't behave differently in debug mode.
[19:52:57] <MatmaRex>	 and eve nworks on older IEs
[19:53:02] <MatmaRex>	 superm401: yes, i know
[19:53:22] <superm401>	 That means there's a problem with the minification.
[19:53:24] <MatmaRex>	 pseudoselectors should neither, but they doin IE<9
[19:53:28] <MatmaRex>	 do in*
[19:53:34] <MatmaRex>	 because it requires a space after them
[19:53:50] <gerrit-wm>	 New patchset: Bsitu; "Remove unused i18n messages" [mediawiki/extensions/Thanks] (master) - https://gerrit.wikimedia.org/r/70750
[19:53:50] <MatmaRex>	 as in, p:first-letter<space here>{ color: red; }
[19:54:16] <MatmaRex>	 or maybe IE<8, whatever.
[19:54:33] <gerrit-wm>	 New patchset: CSteipp; "Basic MediaWiki integration, library cleanup" [mediawiki/extensions/OAuth] (master) - https://gerrit.wikimedia.org/r/70316
[19:55:08] <MatmaRex>	 so they don't work when minified
[19:56:58] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/extensions/OpenStackManager] (master) - https://gerrit.wikimedia.org/r/65269
[19:57:10] <RoanKattouw>	 MatmaRex: OMFG
[19:57:27] <MatmaRex>	 superm401: oh come on. you think  anybody going to work on https://bugzilla.wikimedia.org/show_bug.cgi?id=50830 ? :P
[19:57:36] <MatmaRex>	 RoanKattouw: huh?
[19:57:48] <superm401>	 MatmaRex, IE9 is definitely still supported.
[19:57:52] <RoanKattouw>	 MatmaRex: Re minified selectors not working in IE
[19:57:54] <superm401>	 It's a bug whether or not anyone takes the time to fix it.
[19:58:12] <superm401>	 MatmaRex, if you have any more info (e.g. from looking at the inspector), please add it.
[19:58:14] <MatmaRex>	 RoanKattouw: i think this is even reported as bug
[19:58:31] <MatmaRex>	 superm401: i don't have ie9, i just tested with a screenshot service :)
[19:58:57] <MatmaRex>	 (as in, mwbug)
[19:59:24] <MatmaRex>	 hm, maybe it's not.
[19:59:40] <MatmaRex>	 i remember trying to work around that in some patchset i later redid or abandoned
[19:59:50] <superm401>	 MatmaRex, the pseudo-selector one?
[19:59:58] <superm401>	 If you can't find it, can you file that too?
[20:00:12] <gerrit-wm>	 New review: CSteipp; "PS7: Rebase on top of Aaron's work" [mediawiki/extensions/OAuth] (master) - https://gerrit.wikimedia.org/r/70316
[20:01:09] <gerrit-wm>	 New patchset: Bsitu; "Remove unused i18n messages for corresponding HTML email patch" [mediawiki/extensions/Thanks] (master) - https://gerrit.wikimedia.org/r/70750
[20:02:08] <MatmaRex>	 superm401: RoanKattouw: it's ie6-only, apparnetly
[20:02:10] <MatmaRex>	 http://reference.sitepoint.com/css/pseudoelement-firstletter
[20:02:15] <MatmaRex>	 second-to-last paragraph
[20:02:37] <MatmaRex>	 i can hardly imagine anyone caring about that.
[20:05:20] <superm401>	 IE6 used to be a "grade B" browser, but now that's kind of up in the air.
[20:05:23] <superm401>	 Still going to file, though.
[20:07:18] <wmf-selenium-bot>	 Yippie, build fixed!
[20:07:19] <wmf-selenium-bot>	 Project browsertests-en.wikipedia.org-windows-internet_explorer_10 build #31: FIXED in 5 min 37 sec: https://wmf.ci.cloudbees.com/job/browsertests-en.wikipedia.org-windows-internet_explorer_10/31/
[20:09:37] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/vagrant] (master) - https://gerrit.wikimedia.org/r/72057
[20:14:35] <gerrit-wm>	 Change merged: Raimond Spekking; [mediawiki/extensions/Comments] (master) - https://gerrit.wikimedia.org/r/72117
[20:16:34] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/extensions/WikimediaMessages] (master) - https://gerrit.wikimedia.org/r/72120
[20:17:05] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72131
[20:17:41] <gerrit-wm>	 New review: Mattflaschen; "(1 comment)" [mediawiki/core] (master) C: -1;  - https://gerrit.wikimedia.org/r/71966
[20:18:31] <gerrit-wm>	 New patchset: Parent5446; "Moved account creation throttle to ping limiter" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/65867
[20:19:44] <gerrit-wm>	 New patchset: SuchABot; "Redo campaigns backend to use ContentHandler" [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/70446
[20:20:10] <gerrit-wm>	 Change merged: Parent5446; [mediawiki/extensions/ThrottleOverride] (master) - https://gerrit.wikimedia.org/r/67627
[20:25:19] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72126
[20:34:51] <gerrit-wm>	 New patchset: Legoktm; "testing gerrit-wm [do not merge]" [pywikibot/core] (master) - https://gerrit.wikimedia.org/r/72141
[20:36:44] <MatmaRex>	 superm401: could you do https://gerrit.wikimedia.org/r/#/c/71780/ as well? it's a dependency tothe other one
[20:38:45] <superm401>	 MatmaRex, did you test it?
[20:39:33] <MatmaRex>	 superm401: there's nothing to test there apart from CSS syntax
[20:39:42] <MatmaRex>	 i removed selectors for elements which no longer appear in page HTML
[20:39:57] <MatmaRex>	 they do that because i removed them myself in i think september
[20:39:59] <superm401>	 I know, I just don't feel comfortable merging a JS change if no one's tested it.
[20:40:07] <MatmaRex>	 it's CSS, not JS D:
[20:40:08] <superm401>	 Will do.
[20:40:10] <ori-l>	 someone called?
[20:40:11] <superm401>	 It's both.
[20:40:12] <ori-l>	 heh
[20:40:12] <MatmaRex>	 ah, well
[20:40:14] <MatmaRex>	 technically JS
[20:40:22] <MatmaRex>	 with changes only in jQuery selectors
[20:40:35] <MatmaRex>	 which are technically CSS, too.
[20:40:42] <MatmaRex>	 ori-l: https://gerrit.wikimedia.org/r/#/c/71780/ :)
[20:42:38] <ori-l>	 lgtm
[20:42:41] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/extensions/Vector] (master) - https://gerrit.wikimedia.org/r/71795
[20:42:41] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/extensions/Vector] (master) - https://gerrit.wikimedia.org/r/71780
[20:46:17] <superm401>	 MatmaRex, works fine.  There is also a h5 selector there in core, though.
[20:46:29] <gerrit-wm>	 New patchset: Cmcmahon; "Test for VE Transclusion UI" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/72145
[20:47:06] <YuviPanda>	 UploadWizard: Review time!
[20:47:07] <YuviPanda>	 brion: ^
[20:47:12] <YuviPanda>	 updated
[20:47:47] <gerrit-wm>	 New patchset: SuchABot; "Redo campaigns backend to use ContentHandler" [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/70446
[20:48:29] <YuviPanda>	 now to work on the maintanence script...
[20:49:00] <MatmaRex>	 superm401: there shouldn't be, then
[20:49:06] <MatmaRex>	 i though i killed them already
[20:49:11] <brion>	 whee
[20:49:21] <marktraceur>	 YuviPanda: If you're not careful I'll add a hilight exception for your nick
[20:49:29] <YuviPanda>	 marktraceur: fine fine I'll stop doing that
[20:49:32] <marktraceur>	 Heh.
[20:49:57] <YuviPanda>	 marktraceur: it's actually easier to just highlight your nick, since I haz <tab>
[20:50:24] <marktraceur>	 Exactly.
[20:51:03] <YuviPanda>	 marktraceur: what IRC client are you on, btw?
[20:51:13] <marktraceur>	 irssi
[20:51:20] <YuviPanda>	 with notifications via?
[20:51:26] <YuviPanda>	 an extension?
[20:51:28] * marktraceur  teaches YuviPanda the magic of /ctcp version marktraceur

[20:51:29] <marktraceur>	 No
[20:51:37] <marktraceur>	 I just visit my IRC window pretty frequently
[20:51:43] <YuviPanda>	 hmm
[20:51:47] <YuviPanda>	 that's... very frequent
[20:51:50] <YuviPanda>	 multiple monitors?
[20:51:51] <marktraceur>	 Yes
[20:51:56] <marktraceur>	 Not usually, not right now
[20:52:09] <YuviPanda>	 it's like a polling loop that's so fast you think it is actually event driven
[20:52:14] <marktraceur>	 Heh.
[20:52:32] <marktraceur>	 It would probably be a good idea to put in notifications somehow
[20:52:40] <marktraceur>	 I had an SSH command that did it for a while
[20:52:47] <YuviPanda>	 oh, you don't have a bouncer?
[20:52:51] <YuviPanda>	 just ssh to irssi somewhere?
[20:52:52] <marktraceur>	 But having two SSH connections open constantly seems like a massive waste
[20:52:54] <marktraceur>	 Yeah
[20:53:04] <YuviPanda>	 interesting use of the word 'massive
[20:53:04] <YuviPanda>	 '
[20:53:34] <marktraceur>	 *shrug* considering I'm never the only person on my connection, I'd say it's a pretty accurate use
[20:53:54] <YuviPanda>	 :)
[20:54:02] <YuviPanda>	 '100% increase!'
[20:56:51] <MatmaRex>	 https://en.wikipedia.org/w/index.php?title=Wikipedia:Village_pump_(technical)&diff=prev&oldid=563018581 anybody feel like filing the bug?
[20:59:15] <gerrit-wm>	 New patchset: QChris; "Prohibit uploading draft patch sets to non-draft changes" [gerrit] (wmf) - https://gerrit.wikimedia.org/r/72149
[20:59:15] <superm401>	 MatmaRex, no, not killed yet: https://git.wikimedia.org/blob/mediawiki%2Fcore.git/HEAD/skins%2Fvector%2Fscreen.css#L494
[21:01:10] <MatmaRex>	 superm401: hm. will you merge if i kill it? :)
[21:01:31] <superm401>	 Yeah
[21:02:04] <MatmaRex>	 let me just grep to see if there's anything else
[21:02:27] <MatmaRex>	 yeahhh, quite a bunch. bleh.
[21:06:32] <gerrit-wm>	 New review: Mattflaschen; "(1 comment)" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/60952
[21:07:06] <gerrit-wm>	 New patchset: Catrope; "Update VisualEditor to master" [mediawiki/core] (wmf/1.22wmf9) - https://gerrit.wikimedia.org/r/72150
[21:07:56] <gerrit-wm>	 Change merged: Catrope; [mediawiki/core] (wmf/1.22wmf9) - https://gerrit.wikimedia.org/r/72150
[21:07:56] <gerrit-wm>	 New patchset: Matmarex; "Remove unnecessary h5 selectors" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72151
[21:07:59] <MatmaRex>	 superm401: ^ (untested)
[21:08:03] <gerrit-wm>	 New patchset: Catrope; "Update VisualEditor to master" [mediawiki/core] (wmf/1.22wmf8) - https://gerrit.wikimedia.org/r/72152
[21:08:50] <superm401>	 Matmarex, what happens if you use :not in an older browser?
[21:09:00] <superm401>	 Will it just drop the :not part, or the whole selector?
[21:11:16] <MatmaRex>	 superm401: the whole selector
[21:11:36] <MatmaRex>	 (and even other sub-selectors separated by commas)
[21:13:35] <gerrit-wm>	 New patchset: Raimond Spekking; "Add new special page to alias file" [mediawiki/extensions/ThrottleOverride] (master) - https://gerrit.wikimedia.org/r/72154
[21:13:49] <superm401>	 That's a pain.  We may have to revisit that at some point.
[21:14:52] <MatmaRex>	 that's backwards-compatibility
[21:15:03] <gerrit-wm>	 New patchset: Jqnatividad; "Added try catch block around MongoClient call to suppress MW error message that shows the entire MongoDB connection string when a connection can't be established." [mediawiki/extensions/ExternalData] (master) - https://gerrit.wikimedia.org/r/72156
[21:15:21] <MatmaRex>	 and gracefuldegradation
[21:15:23] <MatmaRex>	 a whole class of IE6 hacks depends on this ;)
[21:16:35] <gerrit-wm>	 New review: Raimond Spekking; "i18n review" [mediawiki/extensions/GlobalBlocking] (master) - https://gerrit.wikimedia.org/r/48608
[21:20:17] <brion>	 IE6 hacks should die at some point
[21:20:31] <brion>	 I remember when we killed Netscape 4 style hacks
[21:20:36] <brion>	 that was a happy day
[21:20:56] <superm401>	 brion, yeah, James_F marked the overly ambitious browser compatibilty section as historical.
[21:21:01] <superm401>	 But we don't have a replacement yet.
[21:21:43] <James_F>	 superm401: Well, I supplied the suggestion blessed by EMGT, and was shouted at in wikitech-l.
[21:21:47] <James_F>	 superm401: So… eh.
[21:21:51] <YuviPanda>	 superm401: RFC time, perhaps :)
[21:21:54] * James_F  has had other things to do. :-)

[21:22:15] <YuviPanda>	 brion: any luck on CR?
[21:22:20] <James_F>	 YuviPanda: I'm not sure an RFC is appropriate; this was about what WMF will pay to ensure works.
[21:22:26] <superm401>	 Yeah, an RFC seems appropriate, especially with the new process to actually get them resolved.
[21:22:31] <brion>	 YuviPanda: nothing new to complain about yet :)
[21:22:36] <YuviPanda>	 brion: that's good!
[21:22:42] <YuviPanda>	 i'll get you a maintenance script shortly
[21:22:47] <brion>	 yayy
[21:23:17] <YuviPanda>	 hehe, at superm401 and James_F's consecutive but divergent views on suitability of RFC to this
[21:23:34] * James_F  grins.

[21:23:59] <gerrit-wm>	 Change merged: Parent5446; [mediawiki/extensions/ThrottleOverride] (master) - https://gerrit.wikimedia.org/r/72154
[21:24:09] <superm401>	 James_F, well, even though a lot of the MW developers are WMF, not all are.
[21:24:21] <superm401>	 The new RFC process basically has the architects making the final decision.
[21:24:32] <superm401>	 Though listening to other devs' feedback, which seems to make sense here.
[21:24:36] <James_F>	 superm401: Oh, absolutely, but pages claiming that the Platform team ensure that we support IE6 are just a lie.
[21:24:51] <James_F>	 superm401: "Architect"s. :-)
[21:25:22] <ori-l>	 starchitects
[21:26:03] <brion>	 B)
[21:26:06] <YuviPanda>	 can't we have a cooler name?
[21:26:16] <YuviPanda>	 BDFL is way cooler than 'Architecht'
[21:26:21] <YuviPanda>	 Architect sounds so javaey.
[21:26:31] <MatmaRex>	 superm401: re https://gerrit.wikimedia.org/r/#/c/60952/ - hm, interesting
[21:26:58] <superm401>	 MatmaRex, are there really h3s in the cactions?
[21:27:02] <superm401>	 Where/when are they?
[21:27:29] <MatmaRex>	 sup
[21:27:32] <brion>	 Architect architect = ArchitectFactory.defaultFactory().createArchitect("brion");
[21:27:33] <MatmaRex>	 superm401: there's one, hidden
[21:27:52] <YuviPanda>	 brion: you forgot the try catch around that
[21:27:55] <brion>	 no no that's too easy. need some URLs in there, like JDBC
[21:28:05] <brion>	 maybe an xml config file
[21:28:17] <YuviPanda>	 brion: yeah, that too. 'defaultFactory()' doesn't actually sound like it
[21:28:56] <brion>	 ArchitectForURI(URI.parse("arch://com.brionv"))
[21:29:07] <YuviPanda>	 that sounds too sane
[21:29:32] <superm401>	 MatmaRex, oh, I see, the text is hidden, but the arrow is animated.
[21:30:34] <YuviPanda>	 brion: should the maintenance script delete things from the database after it converts them?
[21:30:37] <YuviPanda>	 (I think so)
[21:30:57] <brion>	 YuviPanda: if you're adventurous you could drop the table when done :)
[21:31:03] <brion>	 but we've tended to just leave the old data there
[21:31:10] <brion>	 which is often good in case it fucks up in production :)
[21:31:23] <YuviPanda>	 brion: sounds good, but then we might want to make this idempotent
[21:31:28] <YuviPanda>	 but that's gonna be icky
[21:31:33] <brion>	 yep
[21:31:47] <brion>	 dropping the table makes it very idempotent -- it won't have to run the conversion if the table's gone
[21:31:51] <YuviPanda>	 :P
[21:31:54] <brion>	 do we have it backed up in the dumps?
[21:32:02] <YuviPanda>	 yeah but I don't want to go there
[21:32:06] <MatmaRex>	 what tables you guys droppin'?
[21:32:16] <brion>	 looks like no
[21:32:19] <apergos>	 uh?
[21:32:20] <brion>	 upload wizard campaign fields
[21:32:27] <brion>	 i forget the table name
[21:32:33] <brion>	 we're moving it into json content pages
[21:32:34] <YuviPanda>	 uw_campaign_conf i think
[21:32:45] <apergos>	 yeah well I don't have em. maybe db snaps
[21:32:48] <YuviPanda>	 we shouldn't drop it in the maintenance page.
[21:32:57] <YuviPanda>	 err, maintenance script
[21:33:06] <gerrit-wm>	 New review: Matmarex; "(1 comment)" [mediawiki/core] (master) C: 2;  - https://gerrit.wikimedia.org/r/60952
[21:33:11] <apergos>	 keep is good
[21:33:11] <brion>	 well at least somebody should snapshot the data before we deploy :D
[21:33:12] <MatmaRex>	 superm401: ^
[21:33:23] <ori-l>	 YuviPanda: i can do it
[21:33:26] <brion>	 oh there's also that upgrade info table now isn't there
[21:33:38] <YuviPanda>	 ori-l: snapshotting?
[21:33:50] <superm401>	 MatmaRex, thanks.
[21:33:51] <ori-l>	 yeah; i presume the table is pretty small
[21:33:54] <YuviPanda>	 yeah, deployment of this would require support from someone who regularly does deploys
[21:33:56] <YuviPanda>	 it's not too big yeah
[21:34:12] <YuviPanda>	 MatmaRex: part of this largeish patch https://gerrit.wikimedia.org/r/#/c/70446
[21:34:38] <gerrit-wm>	 New patchset: Bsitu; "Update email copy and remove unused messages" [mediawiki/extensions/PageTriage] (master) - https://gerrit.wikimedia.org/r/70761
[21:34:45] <brion>	 -- A table to log updates, one text key row per update.
[21:34:46] <brion>	 CREATE TABLE /*_*/updatelog (
[21:34:48] <MatmaRex>	 YuviPanda: +886, -1815
[21:34:51] <MatmaRex>	 i like it. :P
[21:34:56] <brion>	 ^ that looks like a great place to say "we've done this update, don't run it again by accident"
[21:35:08] <YuviPanda>	 MatmaRex: it was ~-1800 +600, brion and marktraceur are making me increase the +
[21:35:20] <YuviPanda>	 brion: where's that from?
[21:35:25] <brion>	 tables.sql
[21:35:27] <brion>	 core!
[21:35:35] <brion>	 it's real, i didn't just imagine it
[21:35:39] <YuviPanda>	 :P
[21:35:55] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/60952
[21:36:21] <YuviPanda>	 MatmaRex: reviews welcome! It also uses a pattern I'll try to move into core in a week or so... (JsonContent + validation)
[21:36:34] <ori-l>	 the tables are 192kb and 16kb, I think I can manage
[21:36:40] <ori-l>	 i'll just use ZMODEM in case i need to restart the transfer
[21:36:56] <YuviPanda>	 ori-l: hmm, so should I drop the rows from the table after I've migrated them?
[21:37:22] <ori-l>	 YuviPanda: no; the fact that the tables are small makes retention *less* problematic, not more :)
[21:37:24] <YuviPanda>	 that sounds like the 'right thing' to do to me
[21:37:26] <YuviPanda>	 ah
[21:37:37] <YuviPanda>	 more like 'what happens if someone runs that again'?
[21:37:45] <YuviPanda>	 but... why am I trying to guard things against people holding an idiot ball?
[21:37:52] <MatmaRex>	 YuviPanda: i've had enough code for today
[21:37:52] <YuviPanda>	 on the server...
[21:37:53] <YuviPanda>	 ok
[21:38:10] <YuviPanda>	 MatmaRex: heh :) VE on opera?
[21:38:13] <MatmaRex>	 YuviPanda: have i already told you why VE would eat categories in Opera?
[21:38:17] <MatmaRex>	 yeah.
[21:38:26] <YuviPanda>	 MatmaRex: no, something horrible with js compatibility?
[21:38:28] <MatmaRex>	 because Opera can't implement Array.prototype.splice right
[21:38:36] <MatmaRex>	 in 20-fscking-13.
[21:38:39] <YuviPanda>	 oh, so they have an implementation but it's just... wrong?
[21:38:43] <MatmaRex>	 IE6 supports that better than theydo
[21:38:50] <MatmaRex>	 it eats all elements after the 256th one
[21:38:52] <MatmaRex>	 not kidding.
[21:38:52] <YuviPanda>	 MatmaRex: all this should get better when they move properly to blink/webkit
[21:38:55] <YuviPanda>	 ....
[21:38:56] <YuviPanda>	 wow
[21:39:20] <MatmaRex>	 YuviPanda: https://bugzilla.wikimedia.org/show_bug.cgi?id=50385#c6
[21:39:22] <YuviPanda>	 well, this entire HTML / JS / CSS 'application' stack is just one layer of hacks on top of another....
[21:39:32] <YuviPanda>	 so yeah, this kinda shit will be around for a long time :)
[21:39:34] <ori-l>	 YuviPanda: can you rename the table upon completion?
[21:39:47] <ori-l>	 uw_campaigns -> uw_campaigns_migrated?
[21:39:52] <YuviPanda>	 hmm, that would work
[21:40:01] <YuviPanda>	 but is there really no precedent for this?
[21:40:07] <YuviPanda>	 we've never killed a table before?
[21:40:36] <YuviPanda>	 MatmaRex: I once had ajax causing segfaults in a WP7 IE engine....
[21:40:47] <ori-l>	 kill it after the migration is done and the change has been live for a bit
[21:41:10] <ori-l>	 there's little benefit and plenty of risk with automating the drop
[21:41:44] <YuviPanda>	 true
[21:41:54] <MatmaRex>	 YuviPanda: i remember someone complaining about a segfault in safari related to <audio> or something like that
[21:42:05] <YuviPanda>	 MatmaRex: yeah, I vaguely remember that too....
[21:42:12] <YuviPanda>	 MatmaRex: mobile browsers are an order of magnitude worse than desktop ones
[21:42:23] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72151
[21:42:25] <YuviPanda>	 ori-l: so I'll just read from the slaves and not do any writes
[21:42:30] <MatmaRex>	 paraphrased, "javascript has no way to check if the browser segfaults when i do that" :P
[21:42:53] <YuviPanda>	 I remember reading a very similar phrase, absolutely forgot where
[21:43:30] <YuviPanda>	 ori-l: and co-ordinate with renaming / dropping / whatever while deploying
[21:44:00] <brion>	 YuviPanda: the chief advantage of leaving it in place is that an emergency revert won't explode going "where's my table"
[21:44:20] <brion>	 can always manually kill it after :D
[21:45:01] <YuviPanda>	 brion: true. so i'll do that
[21:45:07] <brion>	 whee
[21:45:21] <YuviPanda>	 brion: any other bad things you see?
[21:45:42] <gerrit-wm>	 New patchset: Cmcmahon; "Make login test and method sane and fast" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/72210
[21:48:17] <brion>	 YuviPanda: i'll give it a more thorough workout when the script lands :)
[21:48:27] <YuviPanda>	 brion: oh c'mon :P
[21:48:38] <brion>	 :)
[21:48:50] <YuviPanda>	 brion: there's a lot of stuff here :)
[21:48:57] <gerrit-wm>	 New patchset: PleaseStand; "Unbreak cascading protection" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72211
[21:49:27] <brion>	 well, i want to make sure the script works on the existing campaign data
[21:49:38] <brion>	 and that the updated campaigns work :)
[21:49:41] <YuviPanda>	 brion: sure but that is orthogonal to, say, the js being bad. or something like that
[21:49:50] <ori-l>	 YuviPanda: stat1:/home/olivneh/campaign-tables-5-jul-2013.tar.gz
[21:50:01] * YuviPanda  ssh's

[21:50:11] <brion>	 nothing's sticking out at me from looking over the code
[21:50:14] <ori-l>	 has SQL and CSV dumps of each of uw_campaigns and uw_campaign_conf from commonswiki
[21:50:19] <brion>	 but i do want to exercise more of the code paths with real data
[21:50:23] <YuviPanda>	 brion: I... think that is just what I was going to do :D
[21:50:27] <brion>	 ori-l: i can has .sql?
[21:50:31] <YuviPanda>	 brion: agreed!
[21:50:35] <brion>	 yayyy
[21:50:50] <ori-l>	 brion: sure, how do you want it?
[21:50:53] <ori-l>	 the file on stat1 is world-readable
[21:51:15] <brion>	 ori-l: gimme the path and i can pull it from stat1 sure
[21:51:21] <brion>	 oh path is up there :D
[21:51:23] <brion>	 thanks ;)
[21:51:54] <ori-l>	 np. note that the SQL files each have DROP TABLE IF EXISTS
[21:52:14] <brion>	 little bobby tables
[21:52:19] <ori-l>	 heh
[21:53:39] <gerrit-wm>	 Change merged: Catrope; [mediawiki/core] (wmf/1.22wmf8) - https://gerrit.wikimedia.org/r/72152
[21:54:54] <gerrit-wm>	 New patchset: Cmcmahon; "wait until button is clickable" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/72213
[21:59:11] <gerrit-wm>	 New patchset: Andrew Bogott; "Add a script to unit test puppet modules." [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/72215
[22:03:59] <gerrit-wm>	 New patchset: Legoktm; "Minor updates to the i18n messages from d9a99c6" [mediawiki/extensions/GlobalBlocking] (master) - https://gerrit.wikimedia.org/r/72217
[22:05:54] <gerrit-wm>	 New patchset: MaxSem; "Add debug logging" [mediawiki/extensions/GeoData] (master) - https://gerrit.wikimedia.org/r/72218
[22:06:10] <gerrit-wm>	 Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/72145
[22:06:34] <YuviPanda>	 brion: distracted for a little bit explaining things to the WLM channel. will get back to the script in a little bit
[22:06:47] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/extensions/GlobalBlocking] (master) - https://gerrit.wikimedia.org/r/72217
[22:09:02] <gerrit-wm>	 Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/72213
[22:09:14] <gerrit-wm>	 Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/72210
[22:09:37] <gerrit-wm>	 New patchset: Yaron Koren; "Reverted one change, that led to default values being used for page edits" [mediawiki/extensions/SemanticForms] (master) - https://gerrit.wikimedia.org/r/72220
[22:11:42] <YuviPanda>	 ori-l: do you know how to calculate 'when this has to be merged and tested if it needs to be deployed before end of month'?
[22:11:50] <YuviPanda>	 greg-g_: ^ ?
[22:11:59] <YuviPanda>	 (asking for UploadWizard changes I'm making)
[22:12:08] <gerrit-wm>	 Change merged: Yaron Koren; [mediawiki/extensions/SemanticForms] (master) - https://gerrit.wikimedia.org/r/72220
[22:12:33] <MatmaRex>	 YuviPanda: look at the roadmap
[22:12:39] <MatmaRex>	 !wp mw:MediaWiki 1.22/Roadmap
[22:12:39] <wm-bot>	 https://en.wikipedia.org/?title=mw%3aMediaWiki+1.22%2fRoadmap
[22:12:43] <YuviPanda>	 ah
[22:12:44] <MatmaRex>	 or get it backported
[22:13:08] <YuviPanda>	 well this is going to be on an extension
[22:13:09] <YuviPanda>	 not core
[22:13:17] <YuviPanda>	 no core changes necessary for the first time
[22:13:22] <ori-l>	 YuviPanda: greg-g_ is pretty flexible when it comes to feature teams that manage the feature. If you can be relied on to fix any breakage you might be introducing, you can schedule things with fairly short notice
[22:13:27] <MatmaRex>	 new wmf branches get the master of extensions
[22:13:30] <MatmaRex>	 so this still applies
[22:13:36] <MatmaRex>	 unless you get it backported, of course :)
[22:13:52] <YuviPanda>	 MatmaRex: automatically? I thought they have to be explicitly updated
[22:15:08] <YuviPanda>	 ori-l: ah, sweet! So I'll just need to convince marktraceur to allocate some time to help get this deployed (since he is the UW maintainer)
[22:16:12] <MatmaRex>	 YuviPanda: well, unless i'm wrong about this
[22:16:17] <MatmaRex>	 but i'm pretty sure i'm right
[22:16:23] <MatmaRex>	 Reedy would know
[22:16:54] <Reedy>	 Most extensions are branched from master
[22:17:10] <Reedy>	 It's only a minority that track other branches and stuff
[22:17:29] <gerrit-wm>	 New patchset: Andrew Bogott; "Turn on puppet module unit tests." [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/72221
[22:17:51] <YuviPanda>	 hmm, okay then
[22:17:53] <ori-l>	 MatmaRex: https://github.com/wikimedia/mediawiki-tools-release/blob/master/make-wmf-branch/make-wmf-branch
[22:18:12] <ori-l>	 I don't think they track HEAD of master, though, right? You git clone the extension repository from the current branch
[22:18:22] <ori-l>	 which makes the clone be at the same commit
[22:18:33] <YuviPanda>	 ah, an accurate wording of the question I've actually been trying to ask :)
[22:18:38] <YuviPanda>	 do they track HEAD of master?
[22:18:43] <Reedy>	 No
[22:18:48] <gerrit-wm>	 New review: Andrew Bogott; "I'm not sure about this -- I want the test warn but not prevent merges at the moment, and don't know..." [integration/jenkins-job-builder-config] (master) C: -1;  - https://gerrit.wikimedia.org/r/72221
[22:18:49] <ori-l>	 https://github.com/wikimedia/mediawiki-tools-release/blob/master/make-wmf-branch/make-wmf-branch#L109
[22:18:55] <Reedy>	 That would be continuous deployment ;)
[22:18:56] <YuviPanda>	 so someone has to explicitly update them to HEAD (or something)
[22:18:59] <Reedy>	 Yes
[22:19:08] <YuviPanda>	 indeed :)
[22:19:12] <YuviPanda>	 and a little bit scary ;)
[22:19:16] <YuviPanda>	 (at this point, anyway)
[22:19:35] <ori-l>	 YuviPanda: the beta cluster auto-updates to HEAD I *think*
[22:19:48] <Reedy>	 yeaah
[22:20:05] <Reedy>	 Well, what mediawiki/extensions.git says
[22:20:06] <YuviPanda>	 yeah, though I guess it gets 'stuck' sometimes
[22:20:10] <Reedy>	 Some extensions are brokenededed
[22:20:17] <chrismcmahon>	 ori-l: YuviPanda yes, beta labs has whatever is merged to master minutes after the merge.
[22:20:37] <chrismcmahon>	 for some value of "whatever"
[22:20:51] <YuviPanda>	 heh
[22:21:40] <chrismcmahon>	 which is sometimes great and sometimes annoying.  we're settling the annoying bits one by one by one by one though
[22:22:23] <YuviPanda>	 :)
[22:22:40] <gerrit-wm>	 New review: Anomie; "Good catch. But personally I'd rather see it fixed the other way, by changing sysop?editprotected an..." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72211
[22:22:42] <MatmaRex>	 Reedy: uh, so if i, say, changed something in mediawiki/extensions/Vector, someone has to manually included tht in new wmf branch?
[22:22:49] <Reedy>	 No
[22:23:09] <Reedy>	 It's the master of the extension at the point of making the wmf branch
[22:23:13] <Reedy>	 unless specified othewise
[22:23:30] <MatmaRex>	 yeah, alright. so just like i said earlier?
[22:26:14] <gerrit-wm>	 New review: Spage; "We could do that as well. If a wiki can't update to 1.22 maybe it won't update to a new minor releas..." [mediawiki/extensions/Agora] (master) - https://gerrit.wikimedia.org/r/69458
[22:27:00] <ori-l>	 MatmaRex: someone has to manually bump the ref.
[22:27:31] <MatmaRex>	 ori-l: on new wmf branches oron existing wmf branches?
[22:27:43] <MatmaRex>	 you guys are not agreeing on this, that sounds bad :P
[22:27:44] <ori-l>	 either on the existing branch, before the new branch is cut, or on the new branch
[22:28:00] <MatmaRex>	 [00:23] <Reedy> It's the master of the extension at the point of making the wmf branch
[22:28:01] <gerrit-wm>	 New patchset: Bsitu; "Email copy update + adjustment to HTML email patch" [mediawiki/extensions/PageTriage] (master) - https://gerrit.wikimedia.org/r/72222
[22:28:50] <gerrit-wm>	 New patchset: PleaseStand; "Update 'right-protect', 'right-autoconfirmed' messages" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/72224
[22:29:28] <wmf-selenium-bot>	 Yippie, build fixed!
[22:29:29] <wmf-selenium-bot>	 Project browsertests-test2.wikipedia.org-linux-firefox build #393: FIXED in 18 min: https://wmf.ci.cloudbees.com/job/browsertests-test2.wikipedia.org-linux-firefox/393/
[22:29:41] <gerrit-wm>	 New patchset: Bsitu; "Email copy update + adjustment to HTML email patch" [mediawiki/extensions/PageTriage] (master) - https://gerrit.wikimedia.org/r/72222
[22:31:20] <gerrit-wm>	 New review: Mattflaschen; "I would actually like to consider Extension:Agora killed.  We don't need to delete the repo from Ger..." [mediawiki/extensions/Agora] (master) - https://gerrit.wikimedia.org/r/69458
[22:32:51] <ori-l>	 MatmaRex: look at https://git.wikimedia.org/commit/mediawiki%2Fcore/b0455df95d072bbf17f850b211d16a75d48f6fb1
[22:33:08] <ori-l>	 take, for example, extensions/Math, which was cut at 233d285
[22:35:06] <YuviPanda>	 bleh, network failure
[22:35:07] <YuviPanda>	 MatmaRex: was it clarified how the branching into wmf works?
[22:35:12] <ori-l>	 https://gerrit.wikimedia.org/r/#/c/69886/ was merged on Jun 22nd but was not included
[22:35:46] <YuviPanda>	 so someone has to manually update refs
[22:36:21] <MatmaRex>	 hm.
[22:36:32] <MatmaRex>	 that's not too cool, then.
[22:36:48] <ori-l>	 I'm actually no longer certain
[22:36:51] <MatmaRex>	 let me take this opportunity to ask for someone to update E:Vector with next wmf branch, then.
[22:37:30] <ori-l>	 MatmaRex: submit a patch
[22:38:05] <MatmaRex>	 submodules are dirty, i don't wanna touch them :(
[22:38:25] * YuviPanda  staples his network in place

[22:39:52] * Reedy  patches ori-l

[22:41:55] <jorm>	 i need some jquery help.
[22:42:05] <MatmaRex>	 whatcha messing with, jorm?
[22:42:06] <AzaToth>	 jorm: what's your problem?
[22:42:32] <jorm>	 okay. let's do this here.
[22:42:44] <jorm>	 i have a dom object, a table, with rows.
[22:42:53] <jorm>	 each row has an attribute, datestamp
[22:42:58] <jorm>	 i want to sort them by that.
[22:43:08] <MatmaRex>	 okay, the first question is, why would you do that :P
[22:43:19] <jorm>	 all my normal methods don't work; it seems to sort them without actually applying changes.
[22:43:31] <jorm>	 matmarex: in my flow prototype, i'm building "fake" histories.
[22:43:42] <MatmaRex>	 jorm: stuff that text in data-sortkey attribute and slap $.tablesorter on it?
[22:43:43] <MatmaRex>	 ah
[22:43:51] <MatmaRex>	 so without UI
[22:43:56] <jorm>	 *topics* have histories, and then so do *posts*.  and when i combine them, they're out of order.
[22:43:59] <MatmaRex>	 jorm: you'd have to insert them back in the correct order after sorting
[22:44:16] <jorm>	 so clone and then remove and then append.
[22:44:20] <jorm>	 awesomesauce.
[22:44:30] <MatmaRex>	 clone? noneed
[22:44:36] <MatmaRex>	 and noneed to remove either, actually :P
[22:44:59] <MatmaRex>	 (a node can only exist once, if you insert an existingnode somewhere it will be removed from its previous position)
[22:45:15] <MatmaRex>	 no need*
[22:45:19] <MatmaRex>	 i need to fix my spacebar.
[22:48:21] <gerrit-wm>	 New patchset: Cmcmahon; "Fix for Chrome see https://code.google.com/p/selenium/issues/detail?id=2766" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/72229
[22:48:51] <gerrit-wm>	 Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/72229
[22:50:11] <jorm>	 ah, fuck it. it's a prototype.  no need for super-fidelity.
[23:04:23] <mwalker>	 GorillaWarfare: *waves* anything new happening?
[23:05:21] <gerrit-wm>	 New patchset: Mwjames; "SMW\SQLStore\PropertiesCollector" [mediawiki/extensions/SemanticMediaWiki] (master) - https://gerrit.wikimedia.org/r/70041
[23:06:06] <mwalker>	 ah; probably not given that it is late on a friday -- I'll poke you on monday :)
[23:06:27] <GorillaWarfare>	 mwalker: Not true :P
[23:06:33] <GorillaWarfare>	 mwalker: I'm here, working on the JSON validation
[23:06:59] <YuviPanda>	 GorillaWarfare: if that helps, the large UploadWizard patch does JSON validation now via a Schema
[23:07:07] <YuviPanda>	 not too differnet from EventLogging tho
[23:07:08] <YuviPanda>	 so
[23:07:34] <mwalker>	 YuviPanda: is that deployed?
[23:07:48] <YuviPanda>	 mwalker: it's about to be merged
[23:08:04] <YuviPanda>	 is a bit of a pain, being -1800 +600
[23:08:15] <YuviPanda>	 and being one commit
[23:08:18] <YuviPanda>	 *cringe*
[23:08:28] <YuviPanda>	 my local history is actually currently at 51 commits
[23:08:38] <YuviPanda>	 which if I rebased to how it should be would be something like 45.
[23:08:52] <YuviPanda>	 having that be *one* commit on UW is just ridiculous.
[23:09:01] <YuviPanda>	 (and also reduces my github stats! wah wah wah!)
[23:09:06] <mwalker>	 ouch!
[23:09:20] * YuviPanda  stabs gerrit

[23:09:22] <mwalker>	 can you throw a link in here? I'm curious to see how you did it
[23:09:26] <YuviPanda>	 sure
[23:09:39] <YuviPanda>	 mwalker: https://github.com/yuvipanda/mediawiki-extensions-UploadWizard/tree/campaign-content-handler/includes
[23:09:46] <YuviPanda>	 mwalker: look at CampaignSchema.php for the schema itself
[23:09:55] <YuviPanda>	 mwalker: and validate function in CampaignContent for the validation logic
[23:10:00] <YuviPanda>	 well, not the 'logic' but just the call
[23:10:18] <YuviPanda>	 mwalker: this uses JsonSchema.php which robla wrote, and which should eventually make its way into core soon...
[23:10:25] <YuviPanda>	 (but I'm just directly calling it from EL)
[23:10:40] <mwalker>	 shiney
[23:11:08] <YuviPanda>	 indeed
[23:11:40] <YuviPanda>	 mwalker: if you like pain (or looking at large patches in gerrit), https://gerrit.wikimedia.org/r/#/c/70446 :)
[23:12:04] <YuviPanda>	 (this github -> gerrit sync service brought to you by SuchaBot, poke me if you want it for your repos!)
[23:12:20] <mwalker>	 GorillaWarfare: sorry -- got side tracked by the panda's awesomeness -- what I really wanted to poke you about was to say don't let the metadata discussion slow you down -- I got to thinking that worst case you just build in a feature later that allows you to link .book files to Index: pages using some sort of wiki specific transform rules
[23:12:51] <mwalker>	 YuviPanda: you can transfer comments too?
[23:12:54] <GorillaWarfare>	 mwalker: Yeah, I decided it wouldn't be too wise to wait for any concrete decision. Best to just test the waters, get some ideas, and keep on working
[23:12:56] <YuviPanda>	 mwalker: yes!
[23:13:01] <YuviPanda>	 mwalker: though currently only gerrit to GitHub
[23:13:09] <YuviPanda>	 mwalker: but github to gerrit coming soon! as soon as next week, hopefully
[23:13:35] <YuviPanda>	 look at the comments for https://github.com/wikimedia/mediawiki-extensions-UploadWizard/pull/3, for example.
[23:14:15] <mwalker>	 GorillaWarfare: *thumbs up* anything sticky that I can help with then?
[23:14:39] <YuviPanda>	 (or all of those at https://github.com/wikimedia/apps-android-commons/pulls?direction=desc&page=1&sort=created&state=closed - commons app on android uses SuchABot almost exclusviely)
[23:14:44] <GorillaWarfare>	 So far so good... right now I'm fiddling with an issue where it sometimes accepts properties of the wrong type
[23:14:50] <gerrit-wm>	 Change merged: jenkins-bot; [mediawiki/extensions/SemanticMediaWiki] (master) - https://gerrit.wikimedia.org/r/70041
[23:15:02] <mwalker>	 YuviPanda: egads this is a noisy pull
[23:15:11] <YuviPanda>	 mwalker: yeah, I need to do something about jenkins
[23:15:22] <mwalker>	 I recommend stabbing it with a spork
[23:15:22] <YuviPanda>	 probably only report it if the status changes from last time
[23:15:34] * YuviPanda  stabs gerrit with a large nuke

[23:15:36] <YuviPanda>	 whoops
[23:15:42] <mwalker>	 GorillaWarfare: it might have to do with php type casting?
[23:16:09] <mwalker>	 where it does stupid things like "12345" becomes 12345
[23:16:25] <GorillaWarfare>	 Oh goodness.
[23:16:41] <mwalker>	 YuviPanda: not entirely sure how you can stab with a nuke... bludgeon though... you can bludgeon with a nuke
[23:17:16] <YuviPanda>	 mwalker: step 1. take nuke in hand, step 2. poke gerrit with nuke, step 3. no more gerrit!
[23:17:37] <YuviPanda>	 mwalker: just make sure the nuke is pointy enough
[23:18:26] <mwalker>	 GorillaWarfare: aye... it's a 'feature' PHP; sometimes it's useful; other times its not
[23:18:33] <mwalker>	 *feature of
[23:18:38] <GorillaWarfare>	 Haha, yeah..
[23:18:49] <GorillaWarfare>	 Perhaps it's that, I'll poke some more
[23:19:21] <mwalker>	 depending on how you have things set up; Xdebug can make it easy to do inline debugging
[23:19:36] <mwalker>	 but you would need an IDE that supports it
[23:19:42] * YuviPanda  uses var_dump and die to debug

[23:19:52] * mwalker  *shudders*

[23:20:10] <mwalker>	 interactive debuggers for the win!
[23:20:17] <YuviPanda>	 :D true!
[23:20:20] <mwalker>	 if only php had memory breakpoints!
[23:20:37] <YuviPanda>	 it seems to be a habit I picked up from... somewhere. Unsure where
[23:20:46] <mwalker>	 embedded development?
[23:20:55] <YuviPanda>	 I only use interactive debuggers when the compile / run time of things is too slow (aka when I'm doing Java)
[23:21:11] <YuviPanda>	 mwalker: probably PhoneGap (our older, shittier android app), which didn't have much in the way of debugging.
[23:21:26] <YuviPanda>	 console.logs to find out where your syntax error is!
[23:21:56] <YuviPanda>	 mwalker: outrageous part was, however, that I later found out that it was so hard to debug simply because the PhoneGap code caught all exceptions, swallowed it, and printed a generic 'an error has occured' message, while the exception contained useful details
[23:22:59] <mwalker>	 oooh; we had something similar -- one of our libraries in donationinterface simply printed out "An exception has occurred -- details removed for security"
[23:23:09] <YuviPanda>	 hehe
[23:23:29] <YuviPanda>	 except this was 'details removed because, uh, ... hey that butterfly is shiny!'
[23:23:49] <YuviPanda>	 mwalker: the github gerrit sync bot will probably have two way comment sync done over the weekend. will poke you when that is done :)
[23:25:04] <csteipp>	 log the details, show the users something generic. please.
[23:25:17] <gerrit-wm>	 New patchset: Mwjames; "(Bug 50844) Special:Properties add properties search form" [mediawiki/extensions/SemanticMediaWiki] (master) - https://gerrit.wikimedia.org/r/72232
[23:40:55] <gerrit-wm>	 New patchset: SuchABot; "Redo campaigns backend to use ContentHandler" [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/70446
[23:46:24] <gerrit-wm>	 New patchset: Aaron Schulz; "jobqueue: improved performance of JobQueueGroup::getQueuesWithJobs()" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/71966
[23:46:25] <mwalker>	 csteipp: ah; I just saw your comment -- yes; but in these cases the framework is removing the details before we can even log them! which is totally useful
[23:59:40] <wmf-selenium-bot>	 Yippie, build fixed!
[23:59:40] <wmf-selenium-bot>	 Project browsertests-commons.wikimedia.beta.wmflabs.org-linux-chrome build #165: FIXED in 10 min: https://wmf.ci.cloudbees.com/job/browsertests-commons.wikimedia.beta.wmflabs.org-linux-chrome/165/