[07:34:21] <hashar>	 hello
[07:34:32] <Susan>	 Hi.
[07:34:38] <duh>	 o/
[09:26:35] <gerrit-wm>	 New patchset: Hashar; "triggers for PHP CodeSniffer" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/45066
[09:26:35] <gerrit-wm>	 New patchset: Hashar; "phpcs jobs are non voting for now" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/45065
[09:26:50] <gerrit-wm>	 New review: Hashar; "missed dots in the regex" [integration/zuul-config] (master); V: 0 C: 0;  - https://gerrit.wikimedia.org/r/45065
[09:27:07] <gerrit-wm>	 New review: Hashar; "PS2: rebase of dependent change" [integration/zuul-config] (master); V: 0 C: 0;  - https://gerrit.wikimedia.org/r/45066
[09:28:10] <gerrit-wm>	 New patchset: Hashar; "triggers for PHP CodeSniffer" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/45066
[09:28:10] <gerrit-wm>	 New patchset: Hashar; "phpcs jobs are non voting for now" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/45065
[09:28:22] <gerrit-wm>	 New review: Hashar; "rebased" [integration/zuul-config] (master); V: 2 C: 2;  - https://gerrit.wikimedia.org/r/45065
[09:28:22] <gerrit-wm>	 Change merged: Hashar; [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/45065
[09:28:30] <gerrit-wm>	 New review: Hashar; "rebased" [integration/zuul-config] (master); V: 2 C: 2;  - https://gerrit.wikimedia.org/r/45066
[09:28:30] <gerrit-wm>	 Change merged: Hashar; [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/45066
[09:33:09] <hashar>	 so hmm nobody listening
[09:33:15] <hashar>	 but I got some CodeSniffer report https://integration.mediawiki.org/ci/job/mediawiki-core-phpcs-HEAD/3/violations/file/maintenance/eval.php/? :-]
[09:33:16] <hashar>	 \O/
[09:33:47] <Nikerabbit>	 hashar: is there anyone in WMF who can help to solve nasty unit test related issues?
[09:34:12] <hashar>	 Nikerabbit: me ? :-]
[09:34:43] <Nikerabbit>	 hashar: well I already mentioned it to you: https://bugzilla.wikimedia.org/show_bug.cgi?id=43899
[09:34:49] <Nikerabbit>	 I just can't figure out what is going on here
[09:35:09] <Nikerabbit>	 hashar: nice for CodeSniffer... can we tune the checks to ignore some of them?
[09:36:14] <hashar>	 Nikerabbit: CodeSniffer currently run the default set of rules which do not pass for us.  I have a MediaWiki coding standard floating somewhere in a git repository, I need to slightly tweak it.
[09:36:33] <hashar>	 for DataValues, I have really no idea
[09:36:56] <hashar>	 Nikerabbit: the big issues with failing tests, ,is finding a way to reproduce it
[09:37:18] <hashar>	 since translatewiki loads various extensions and has lot of specific settings, I am not surprised some tests are failing
[09:37:25] <hashar>	 that is not an issue in translatewiki though
[09:37:32] <hashar>	 but more an issue about the tests not being well written
[09:38:45] <Nikerabbit>	 hashar: yes but something is causing exceptions to not get caught
[09:39:17] <Nikerabbit>	 there is nothing wrong in the tests as far as I can see
[09:39:45] <Nikerabbit>	 it's as if there is custom error handler that just ignores exceptions
[09:41:10] <Nikerabbit>	 sorry
[09:41:24] <Nikerabbit>	 the exception is thrown, but it is not caught by the try-catch block
[09:43:31] <Nikerabbit>	 hashar: really it has nothing to do with DataValues... it's just has tests which rely on catching exceptions with try-catch
[09:44:27] <ori-l>	 Nikerabbit: which line of which file are you commenting out? (https://bugzilla.wikimedia.org/show_bug.cgi?id=43899#c11)
[09:47:41] <Nikerabbit>	 ori-l: the throw new MWException in Title.php
[10:00:22] <ori-l>	 Nikerabbit: poking at it, nothing so far..
[10:06:32] <Nikerabbit>	 ori-l: yeah I've spent some hours on it
[10:07:28] <Nikerabbit>	 ori-l: protip: in MediaWikiTestCase::run, you can make it to echo the name of each test, and also assign numbers to it and skip some based on the number
[10:08:52] <pastakhov_>	 JeroenDeDauw: hi, you can create new branches in extensions/Maps for me?
[10:09:26] <pastakhov_>	 JeroenDeDauw: *one branch
[10:16:31] <JeroenDeDauw>	 pastakhov_: now has a pastakhov branch
[10:16:36] <JeroenDeDauw>	 made of last master
[10:17:35] <pastakhov_>	 JeroenDeDauw: thanks
[10:21:58] <ori-l>	 Nikerabbit: I'm hitting a wall so I'm going to drop this for now and look again tomorrow. I'll let you know if I make any headway.
[10:23:51] * hashar  hands ori-l a hammer for his wall

[10:24:12] <ori-l>	 hashar: it's 2:24 AM, I better not start banging :)
[10:24:18] <hashar>	 yeah
[10:24:28] <hashar>	 get a non-technical book and get some sleep I guess
[10:24:49] <ori-l>	 about to
[10:24:51] <hashar>	 I am not sure how you manage to get to bed sooo late and still wake up early in the morning
[10:24:57] <hashar>	 (or maybe you sleep in the morning haha)
[10:25:10] <hashar>	  /msg chanserv mode +b ori-l
[10:25:19] <hashar>	  /msg chanserv kick #wikimedia-dev ori-l
[10:26:08] <ori-l>	 :P good night.
[10:26:20] <Nikerabbit>	 ori-l: thanks, I know it is tricky.. especially if you can't reproduce
[10:27:04] <ori-l>	 Nikerabbit: yeah, sorry i couldn't come up with anything.
[12:04:03] <DanielK_WMDE>	 siebrand: you around? can you remove the -2 from https://gerrit.wikimedia.org/r/#/c/38906/ ? It has become baseless, I think.
[12:05:33] <siebrand>	 DanielK_WMDE: Changed to +2
[12:05:41] <DanielK_WMDE>	 siebrand: thanks
[12:05:56] * DanielK_WMDE  is not totally happy with the change as it is, but the -2 wqas blocking things

[12:06:25] <DanielK_WMDE>	 siebrand: triggering a special message based on the database name makes my skin crawl :P
[12:06:45] <siebrand>	 DanielK_WMDE: It doesn't deserve any prizes for elegance...
[12:06:54] <DanielK_WMDE>	 but it will do, i guess
[12:07:03] <DanielK_WMDE>	 still.
[12:07:16] <DanielK_WMDE>	 siebrand: anyway, thanks
[12:07:25] <siebrand>	 DanielK_WMDE: Shit, it needs amending...
[12:07:45] <siebrand>	 DanielK_WMDE: Terms of Use and https.
[12:07:49] <siebrand>	 DanielK_WMDE: Lemme do that...
[12:08:29] <DanielK_WMDE>	 well, with a +2, jenkins is already abotu to merge it, no?
[12:11:47] <siebrand>	 DanielK_WMDE: There's no auto check on that repo, I believe. Also, if you submit a new patch set, I think that will block Jenkins' auto merge for an earlier patch set, too. Pinging hashar to confirm the latter...
[12:12:19] <DanielK_WMDE>	 ah, right, no auto.merge for wikibase
[12:12:23] <DanielK_WMDE>	 (this is confusing)
[12:13:50] <siebrand>	 DanielK_WMDE: Merged.
[12:13:58] <DanielK_WMDE>	 thanks
[12:15:08] <hashar>	 siebrand: WikimediaMessages repo ?
[12:15:15] <siebrand>	 hashar: *nod*
[12:15:39] <hashar>	 I really have to audit all extensions and find out which one are missing
[12:16:13] <siebrand>	 hashar: The question is more like this: What does Jenkins do if it has "check and merge" orders for PS1 and PS2 is uploaded before the checks and merge has been performed. Will it merge PS1, or ignore PS1 results, and work on PS2?
[12:17:17] <hashar>	 I have zero idea
[12:17:31] <hashar>	 I guess it will attempt to submit patchset 1
[12:17:50] <hashar>	 and gerrit would complains because that is not the latest patchset
[12:17:53] <hashar>	 would have to try that out
[12:19:13] <gerrit-wm>	 New patchset: Hashar; "jobs for mw/ext/WikimediaMessages" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/45121
[12:19:16] * ^demon  waves to hashar and siebrand

[12:19:40] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/45064
[12:19:40] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/45121
[12:19:40] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/45063
[12:20:04] <hashar>	 hi chad :-]
[12:21:41] <^demon>	 How was your weekend?
[12:22:04] <hashar>	 debugged some python code
[12:22:05] <hashar>	 ranted about java
[12:22:18] <^demon>	 Don't hate on java :p
[12:22:20] <hashar>	 and cleaned a pile of stuff I have been accumulating in my house
[12:22:36] <hashar>	 so i can buy more stuff :-]
[12:23:09] <gerrit-wm>	 New patchset: Hashar; "triggers for mw/ext/WikimediaMessages" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/45122
[12:23:24] <gerrit-wm>	 Change merged: Hashar; [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/45122
[12:24:28] <hashar>	 ^demon: I think gerrit-wm is slightly broken
[12:24:33] <^demon>	 Oh?
[12:24:40] <hashar>	 ^demon: does not notify changes on mw/extensions in #mediawiki
[12:24:47] <hashar>	 maybe it has been silenced there or banned
[12:24:51] <^demon>	 Bleh. Was there a netsplit?
[12:24:58] <hashar>	 maybe
[12:24:58] <^demon>	 ircecho doesn't handle netsplits well.
[12:25:14] <hashar>	  https://gerrit.wikimedia.org/r/45123 onn /mediawiki/extensions/WikimediaMessages did not trigger a notification
[12:25:37] <^demon>	 I'll re-start the bot.
[12:25:39] <hashar>	 siebrand: you got tests for mediawiki/extensions/WikimediaMessages now
[12:26:47] <hashar>	 ^demon: I also made  mw/tools/* notifications to be sent  in #wikimedia-dev instead of #mediawiki  https://gerrit.wikimedia.org/r/#/c/45108/
[12:27:58] <^demon>	 +1
[12:34:45] * siebrand  thanks hashar 

[12:34:54] <hashar>	 yw
[12:35:08] <hashar>	 ^demon: I am working on a hack to finally get PHP CodeSniffer to run on our changes :D
[12:35:24] <^demon>	 oh?
[12:35:46] <hashar>	 bug is https://bugzilla.wikimedia.org/show_bug.cgi?id=35588
[12:35:52] <hashar>	 example output https://integration.mediawiki.org/ci/job/mediawiki-core-phpcs-HEAD/3/violations/file/maintenance/eval.php/
[12:36:02] <hashar>	 and https://integration.mediawiki.org/ci/job/mediawiki-core-phpcs-HEAD/3/violations/?
[12:36:14] <hashar>	 that is already in production for mediawiki/core
[12:36:28] <hashar>	 though a failure does not  V-1 the change (aka  any phpcs jobs are non voting)
[12:36:42] <hashar>	 https://integration.mediawiki.org/ci/job/mediawiki-core-phpcs-HEAD/11/violations/file/includes/parser/ParserOutput.php/?
[12:36:44] <hashar>	 sooo many failures
[12:37:02] <hashar>	 i am tweaking the configuration file right now
[12:38:09] <^demon>	 Yeah, some of those are wrong for us :p
[12:38:25] <^demon>	 "Spaces must be used to indent lines..."
[12:38:29] <^demon>	 I don't think so ;-)
[12:39:36] <hashar>	 :-]
[12:39:55] <hashar>	 PHP has some coding standards
[12:39:57] <hashar>	 https://github.com/php-fig/fig-standards/blob/master/accepted/PSR-2-coding-style-guide.md
[12:40:11] <hashar>	 I will let you read the first rules and scream :-]
[12:43:03] <^demon>	 Yeah, I've seen PSR.
[12:43:22] <^demon>	 It's basically what PEAR follows. Really, it's all arbitrary.
[12:43:29] <^demon>	 But so is our style.
[12:44:09] <Platonides>	 that's a group outside of php developers
[12:44:13] <Platonides>	 if that relieves you
[12:44:41] <^demon>	 I know.
[12:44:45] <^demon>	 Haha, "The keyword elseif SHOULD be used instead of else if so that all control keywords look like single words."
[12:44:52] <^demon>	 What silly things they think.
[12:45:45] <Platonides>	 well, I think we also prefer elseif
[12:45:58] <^demon>	 We do, but it's not "so they all look like single words"
[12:49:22] <hashar>	   Generic.WhiteSpace.DisallowSpaceIndent
[12:49:22] <hashar>	   Generic.WhiteSpace.DisallowTabIndent
[12:49:23] <hashar>	   Generic.WhiteSpace.ScopeIndent
[12:49:32] <hashar>	 pick one :-]
[12:50:03] <^demon>	 DisallowSpaceIndent.
[12:50:33] <^demon>	 Although I wonder if that will barf on files that have both tabs & spaces for vertical alignment, like Autoloader.
[12:50:39] <^demon>	 Oh well.
[12:52:42] <hashar>	 we will see
[13:09:31] <hashar>	 there are sooo many rules
[13:19:13] <hashar>	 lame code sniffer stuff is at https://www.mediawiki.org/wiki/Continuous_integration/PHP_CodeSniffer#run
[14:02:11] <hashar>	 ok so that is not java which is a failure
[14:02:33] <hashar>	 that is the open source authors figuring out they can hack stuff to fit their needs yet breaking for everyone else
[14:02:38] <hashar>	 and maintainer not giving clue
[14:02:40] <hashar>	 grr
[14:33:09] <gerrit-wm>	 New patchset: Hashar; "phpcs: use MediaWiki standard" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/45134
[14:34:38] <gerrit-wm>	 New review: Hashar; "I have rebuild the jobs so they use the MediaWiki standards:" [integration/jenkins-job-builder-config] (master); V: 2 C: 2;  - https://gerrit.wikimedia.org/r/45134
[14:34:38] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/45134
[14:37:52] <hashar>	 and here we have PHP_CodeSniffer
[14:37:53] <hashar>	 https://www.mediawiki.org/wiki/Continuous_integration/PHP_CodeSniffer
[14:52:01] <hashar>	 https://integration.mediawiki.org/ci/job/mediawiki-core-phpcs-HEAD/30/violations/file/maintenance/eval.php/?
[14:52:02] <hashar>	 lovely
[14:52:10] <hashar>	 show up that eval.php has too many blanklines
[14:53:28] <Nikerabbit>	 blank lines are like art!
[14:54:00] <^demon>	 They're only truly appreciated after the author is gone?
[14:57:01] <DanielK_WMDE>	 hashar: any idea what that means? https://integration.mediawiki.org/ci/job/mediawiki-core-phpcs-HEAD/25/console
[14:57:15] <hashar>	 DanielK_WMDE: about to write an announcement
[14:57:15] <hashar>	 DanielK_WMDE: you can ignore that
[14:57:41] <DanielK_WMDE>	 ok :)
[14:57:43] <hashar>	 DanielK_WMDE: basically that is an HIGLY experimental job to run PHP_CodeSniffer on mediawiki/core
[14:57:49] <hashar>	 the result is non voting / ignored
[14:57:55] <DanielK_WMDE>	 right
[15:21:45] <Nikerabbit>	 ^demon: good point :)
[15:50:01] <siebrand>	 hashar: still incomplete? https://integration.mediawiki.org/ci/job/mediawiki-core-phpcs-HEAD/33/console
[15:50:14] <hashar>	 siebrand: yeah
[15:50:23] <hashar>	 siebrand: was looking at it
[15:50:35] <hashar>	 I get the list of files with  git diff --name-only HEAD^
[15:50:36] <hashar>	 but that list deleted files as well hehe
[15:54:29] <hashar>	 git diff --name-status HEAD^ | egrep -v '^D'| cut -f2
[15:58:02] <gerrit-wm>	 New patchset: Hashar; "phpcs ran on deleted file" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/45143
[15:59:23] <gerrit-wm>	 New review: Hashar; "regenerated production jobs with PS1." [integration/jenkins-job-builder-config] (master); V: 0 C: 0;  - https://gerrit.wikimedia.org/r/45143
[15:59:42] <hashar>	 https://integration.mediawiki.org/ci/job/mediawiki-core-phpcs-HEAD/34/console
[15:59:45] <hashar>	 siebrand: that works :-]
[16:00:00] <hashar>	 the "nice" report is at https://integration.mediawiki.org/ci/job/mediawiki-core-phpcs-HEAD/34/violations/?
[16:00:23] <gerrit-wm>	 New review: Hashar; "worked :-) thanks siebrand!" [integration/jenkins-job-builder-config] (master); V: 2 C: 2;  - https://gerrit.wikimedia.org/r/45143
[16:00:41] <hashar>	 there are probably lot of checks we want to ignore for now
[16:00:46] <hashar>	 and or some we want to discuss about
[16:05:36] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/45143
[16:06:16] * hashar  waves

[16:06:20] <hashar>	 daughter duty
[21:35:40] <gerrit-wm>	 New review: Krinkle; "Thanks for the useful review, I'm very inexperienced with this language." [integration/jenkins-job-builder] (master); V: 1 C: 1;  - https://gerrit.wikimedia.org/r/44836
[21:36:12] <gerrit-wm>	 New review: Krinkle; "Constistent as in, the style it was written in since this is an upstream repo. Not our own." [integration/jenkins-job-builder] (master); V: 1 C: 1;  - https://gerrit.wikimedia.org/r/44836
[21:36:18] <Krinkle>	 ori-l: ^
[21:40:02] <Krinkle>	 ori-l: Look at the rest of that file, literally every single line I added in that commit is directly derived from an existing line.
[22:00:13] <gerrit-wm>	 New review: Ori.livneh; "Makes sense; I did not attach a score to my review for exactly that reason." [integration/jenkins-job-builder] (master); V: 0 C: 1;  - https://gerrit.wikimedia.org/r/44836
[22:52:39] <Ryan_Lane>	 can I get a review? https://gerrit.wikimedia.org/r/45258
[22:53:17] <^demon>	 merged.
[22:53:22] <Ryan_Lane>	 thanks
[22:53:26] <^demon>	 yw.
[23:04:24] <Krinkle>	 brion: "we're just writing to let you know that you've been automatically subscribed to a repository on GitHub."
[23:04:25] <Krinkle>	 brion: "wikimedia/Commons-iOS created by brion"
[23:04:30] <Krinkle>	 brion: Have you looked into WikiSnaps ?
[23:04:43] <brion>	 Krinkle: link it to me and i'll look it up
[23:04:46] <Krinkle>	 Created by <thedj>  at the amsterdam hackathon last year
[23:04:57] <Krinkle>	 https://www.mediawiki.org/wiki/WikiSnaps
[23:04:59] <Krinkle>	 brion: ^
[23:05:30] <Krinkle>	 it "worked" last I checked (was live demoed at the event)
[23:05:33] <Krinkle>	 Good luck :)