[00:00:29] <csteipp>	 Does mediawiki 1.24 run on hhvm? I think we had hhvm on vagrant before last fall, but can't remember the timing..
[00:00:55] <legoktm>	 csteipp: the release notes say "beta support"
[00:01:20] <legoktm>	 https://github.com/wikimedia/mediawiki/blob/REL1_24/RELEASE-NOTES-1.24#L760 oh "experimental"
[00:35:20] <ori>	 AaronSchulz: https://gerrit.wikimedia.org/r/#/c/197449/
[01:16:51] <ori>	 Krinkle: what should we do about the remaining languages? I think we should combine the less-frequently-used ones into modules with, say, 20-30 languages each. That way, if you happen to land on a page that uses them (which is rare), you'd get some CSS bloat, but the common case (startup module, every req) would be optimized
[01:17:33] <Krinkle>	 ori: Hm.. is it loaded via moduleStyles or regular modules
[01:17:58] <Krinkle>	 it'd add a lot to the 5min/uncached requests for pages with one of the popular ones as it'd load 30 of them.
[01:18:15] <Krinkle>	 Which is its own problem.
[01:18:15] <ori>	 addModuleStyles
[01:18:25] <Krinkle>	 Roan and I started research into making addModuleStyles not uncached.
[01:18:29] <Krinkle>	 Which requires javascript
[01:18:33] <Krinkle>	 e.g. adding them to the top queue
[01:18:37] <Krinkle>	 with a <noscript> fallback
[01:18:50] <Krinkle>	 At the time we ran into the problem that some older browsers do not support <link> inside <noscript>
[01:19:00] <Krinkle>	 I think it was only Opera < 10
[01:19:05] <Krinkle>	 or < 11
[01:19:15] <Krinkle>	 uncached/unversioned
[01:21:33] <ori>	 this is 30 languages:
[01:21:36] <ori>	 https://bits.wikimedia.org/en.wikipedia.org/load.php?debug=false&lang=en&modules=ext.geshi.language.c%7Cext.geshi.language.cpp%7Cext.geshi.language.bash%7Cext.geshi.language.html4strict%7Cext.geshi.language.text%7Cext.geshi.language.java%7Cext.geshi.language.latex%7Cext.geshi.language.javascript%7Cext.geshi.language.python%7Cext.geshi.language.xml%7Cext.geshi.language.csharp%7Cext.geshi.language.php%7Cext.geshi.language.css%7Ce
[01:21:36] <ori>	 xt.geshi.language.asm%7Cext.geshi.language.sql%7Cext.geshi.language.pascal%7Cext.geshi.language.matlab%7Cext.geshi.language.html5%7Cext.geshi.language.haskell%7Cext.geshi.language.vb%7Cext.geshi.language.lisp%7Cext.geshi.language.ruby%7Cext.geshi.language.ada%7Cext.geshi.language.oracle11%7Cext.geshi.language.dos%7Cext.geshi.language.rsplus%7Cext.geshi.language.fortran%7Cext.geshi.language.d%7Cext.geshi.language.bnf%7Cext.geshi
[01:21:36] <ori>	 .language.ocaml&only=styles&skin=vector&*
[01:21:55] <ori>	 sorry, http://tinyurl.com/p5m9qcj 
[01:22:06] <ori>	 it's only 5k gzipped
[01:22:55] <ori>	 you could have the top 50 modules still defined as separate modules, that would cover 94.50% of usage
[01:23:28] <ori>	 and then the remaining 92 batched into three modules with ~31 languages each
[01:24:16] <ori>	 keep in mind that this is 94.50% of geshi usage; geshi is only used on a small minority of pages
[01:24:32] <ori>	 so the % of page views that would get the bloated CSS would be very, very small
[01:25:47] <ori>	 or have 20 language-specific modules, which would cover 81% of geshi usage
[01:26:59] <ori>	 and have 10 modules with 12 languages each
[01:27:24] <ori>	 that's 30 module defs in total, with acceptable bloat in a small minority of cases
[01:27:50] <ori>	 down from 215 modules now, which is absurd
[01:28:36] <Krinkle>	 ori: Yeah, I'd fine with any of those solutions.
[01:29:00] <Krinkle>	 ori: See also https://phabricator.wikimedia.org/T85796
[01:29:21] <Krinkle>	 ori: It really shouldn't need separated stylesheets in the first place. Though that requires rigging in the extensions logic itself
[01:29:48] <Krinkle>	 Upon furthe research I did find other highlighters that work this way, but initially Geshi was the only one I knew of that did this.
[01:30:17] <Krinkle>	 It seems quite unneeded and just showing of the poor quality in how it's written and maintained for more serious applications.
[01:32:21] <ori>	 not to mention the fact that although it supports blitzbasic, providex, intercal and gambas, it does not have support for rust
[01:32:34] <ori>	 and a host of other new-but-trending languages
[01:33:08] <Krinkle>	 I suspect AaronSchulz can look 2.5 minutes in the future.
[01:33:16] <Krinkle>	 Three times today now when I was about to ping him, he left 2.5 minutes ago.
[01:33:42] <ori>	 swift is another example
[01:33:46] <Krinkle>	 Yeah
[01:34:19] <ori>	 no swift, julia, nim, or rust
[01:34:45] <Krinkle>	 But hey, it supports CoffeeScript
[01:34:51] <ori>	 and dart
[01:35:02] <ori>	 no typescript
[01:35:28] <ori>	 but it does support typoscript, which is a real thing: http://docs.typo3.org/typo3cms/TyposcriptReference/
[01:35:46] <Krinkle>	 and Windows Registry file
[01:36:27] <ori>	 those HKEYs aren't going to highlight themselves
[01:37:25] <ori>	 https://github.com/github/linguist seems popular; is there no PHP parser of its grammar?
[01:38:26] <ori>	 i don't think maintaining a syntax highlighting service is practical
[01:38:33] <ori>	 in case you were about to propose 'syntaxoid' :P
[01:39:55] <ori>	 this looks ok: https://github.com/kzykhys/Pygments.php
[03:12:15] <wikibugs>	 6MediaWiki-Core-Team, 7HHVM, 7Upstream: HHVM with FastCGI does not support streaming output - https://phabricator.wikimedia.org/T91468#1127375 (10tstarling) Here is the patch as it stands:  {F99689}  It seems to work. I've tried a print loop and readfile('/dev/zero'), with and without ob_implicit_flush(), it...
[04:21:20] <ori>	 Krinkle: consolation: even google gets it wrong: http://www.google.com/makani/
[04:21:33] <ori>	 mousewheel event listener, i get 20-30 fps on scroll
[04:22:24] <ori>	 not to mention: Polaris-Bold.woff, Polaris-Book.woff, PolarisCondensed-Medium.woff
[04:26:40] <Krinkle>	 ori: Yeah, I looked at linguist, but doesn't seem very widely adopted (basically just that one implementation and github using it, which is cool, but not in PHP)
[04:26:54] <Krinkle>	 TextMate's grammer format seems quite widely supported, btu havent' found a PHP parser for it yet
[04:27:34] <ori>	 what about doing it entirely client-side?
[04:27:45] <Krinkle>	 ori: Possible yeah
[04:27:49] <Krinkle>	 If we find a js implementation
[04:27:52] <Krinkle>	 https://jira.appcelerator.org/browse/APSTUD-4905
[04:27:57] <Krinkle>	 Looks like there is a PHP implementation
[04:29:04] <ori>	 no, that's the parser for PHP code in Aptana studio
[04:29:24] <ori>	 lexer, really. and the bug reporter finds it inferior to textmate's.
[04:29:32] <Krinkle>	 Ah, right
[04:30:15] <ori>	 i wonder if we should revert <https://gerrit.wikimedia.org/r/#/c/196989/>
[04:30:27] <Krinkle>	 ori: Yeah
[04:30:52] <Krinkle>	 ori: I think some of it is definitely useful, but I'm not sure the queue improves things currently.
[04:31:01] <aspiecat>	 why not? :(
[04:31:04] <Krinkle>	 I take it you read the two loading-dev posts?
[04:31:47] <ori>	 aspiecat: TL;DR: browser devs got tired of waiting for front end devs to get the message, so modern browsers have a lookahead parser that loads resources anyway.
[04:32:33] <ori>	 Krinkle: I still need to go over those again. Before reverting I'd like the chance to poke at it while it's on beta.
[04:33:18] <Krinkle>	 ori: Hm.. I assumed you knew about the lookahead in browsers. I was thinking this intended to improve how inline scripts and external stylesheets factor in
[04:33:35] <Krinkle>	 somehow that the stylesheets needing to be applied affects something
[04:33:39] <ori>	 yes, that was the intent
[04:33:51] <ori>	 ok, let me read the emails now that i have a minute
[04:33:55] <Krinkle>	 k
[04:33:58] <Krinkle>	 I'm closing up
[04:34:01] <Krinkle>	 o/
[04:34:05] <ori>	 good night
[04:35:47] <ori>	 oh, interesting
[04:36:29] <ori>	 i should just patch https://code.google.com/p/chromium/codesearch#chromium/src/third_party/WebKit/Source/core/html/HTMLLinkElement.cpp&sq=package:chromium&type=cs&l=718&rcl=1425458788 and build chromium to see if it is blocking or not
[04:36:40] <ori>	 or use gdb if the build has debug symbols
[04:38:34] <ori>	 no symbols
[04:39:53] <ori>	 ubuntu has chromium-browser-dbg
[05:19:10] <wikibugs>	 6MediaWiki-Core-Team, 10SUL-Finalization: Don't list users who are now attached to a global account on Special:UsersWhoWillBeRenamed - https://phabricator.wikimedia.org/T93044#1127450 (10Legoktm) 3NEW a:3Legoktm
[05:20:46] <wikibugs>	 6MediaWiki-Core-Team, 7HHVM, 7Upstream: HHVM with FastCGI does not support streaming output - https://phabricator.wikimedia.org/T91468#1127458 (10tstarling) Well, strangely enough, the patched version is 1.7 times faster on my laptop when streaming a 1MB file with readfile(). I thought that lower system bus...
[06:39:56] <wikibugs>	 6MediaWiki-Core-Team, 10SUL-Finalization: Run MassMessage to contact user talk pages of all affected accounts - https://phabricator.wikimedia.org/T90820#1127509 (10Ricordisamoa) Sending one notice for each of the two clashing accounts: https://it.wikipedia.org/w/index.php?title=Discussioni_utente%3ABarbaking&d...
[06:43:04] <wikibugs>	 6MediaWiki-Core-Team, 10SUL-Finalization: Run MassMessage to contact user talk pages of all affected accounts - https://phabricator.wikimedia.org/T90820#1127523 (10Legoktm) >>! In T90820#1127509, @Ricordisamoa wrote: > Sending one notice for each of the two clashing accounts: > https://it.wikipedia.org/w/index...
[07:55:30] <wikibugs>	 6MediaWiki-Core-Team, 10hardware-requests, 6operations, 5Patch-For-Review: Fluorine needs bigger disks - https://phabricator.wikimedia.org/T92417#1127597 (10yuvipanda) p:5High>3Normal Back to normal, since there don't seem to have been any alerts of late.
[07:55:52] <wikibugs>	 6MediaWiki-Core-Team, 10hardware-requests, 6operations, 5Patch-For-Review: Fluorine needs bigger disks - https://phabricator.wikimedia.org/T92417#1127599 (10yuvipanda) a:3Andrew (Assigning to @Andrew as the last person to have touched it :) )
[08:40:58] <wikibugs>	 6MediaWiki-Core-Team, 7HHVM, 7Upstream: HHVM with FastCGI does not support streaming output - https://phabricator.wikimedia.org/T91468#1127825 (10Joe) @tstarling did you test the patch against the development HEAD of HHVM? I'll try seeing if it applies to our version (3.3.x) as the package needed refreshing...
[09:05:13] <wikibugs>	 6MediaWiki-Core-Team, 10SUL-Finalization: Run MassMessage to contact user talk pages of all affected accounts - https://phabricator.wikimedia.org/T90820#1127909 (10Nemo_bis) p:5Triage>3High
[09:08:17] <wikibugs>	 6MediaWiki-Core-Team, 10MassMessage, 10SUL-Finalization: MassMessage sent twice to local user talk - https://phabricator.wikimedia.org/T93049#1127919 (10Nemo_bis) 3NEW a:3Legoktm
[09:10:44] <wikibugs>	 6MediaWiki-Core-Team, 10SUL-Finalization: Run MassMessage to contact user talk pages of all affected accounts - https://phabricator.wikimedia.org/T90820#1127929 (10Nemo_bis) Over a million messages delivered by now, if I remember the previous edit count correctly (it's now at 1.6 M).
[10:04:02] <wikibugs>	 6MediaWiki-Core-Team, 7HHVM, 7Upstream: HHVM with FastCGI does not support streaming output - https://phabricator.wikimedia.org/T91468#1128024 (10tstarling) I think it would take a lot of work to backport it, since the relevant code was [[https://github.com/facebook/hhvm/commit/87f853d16db9fb0b443f856de1c5f2...
[10:25:54] <Nikerabbit>	 New issues related to user preferences has come up, and I need help to debug: https://phabricator.wikimedia.org/T92232
[12:11:51] <wikibugs>	 6MediaWiki-Core-Team, 10MassMessage, 10SUL-Finalization: MassMessage sent twice to local user talk - https://phabricator.wikimedia.org/T93049#1128250 (10Ricordisamoa)
[12:35:21] <wikibugs>	 6MediaWiki-Core-Team, 6Commons, 6Multimedia, 6operations: Deleted image pages do not show up as deleted until a manual action=purge - https://phabricator.wikimedia.org/T93052#1128169 (10yuvipanda)
[12:44:49] <wikibugs>	 6MediaWiki-Core-Team, 6Commons, 6Multimedia, 6operations: Deleted image pages do not show up as deleted until a manual action=purge - https://phabricator.wikimedia.org/T93052#1128301 (10yuvipanda) (am fairly unfamiliar with our LocalFile setup, am trying to trace the actions caused by action=purge, to see...
[13:21:39] <wikibugs>	 6MediaWiki-Core-Team, 6Commons, 6Multimedia, 6operations: Deleted image pages do not show up as deleted until a manual action=purge - https://phabricator.wikimedia.org/T93052#1128387 (10Gilles)
[14:12:09] <wikibugs>	 6MediaWiki-Core-Team, 10CirrusSearch: Wikipedia search engine not updating for moved article - https://phabricator.wikimedia.org/T93040#1128565 (10Manybubbles)
[14:16:03] <wikibugs>	 6MediaWiki-Core-Team, 10SUL-Finalization: Run MassMessage to contact user talk pages of all affected accounts - https://phabricator.wikimedia.org/T90820#1128584 (10Nemo_bis) Re https://meta.wikimedia.org/w/index.php?diff=11604303&oldid=11598114&rcid=6087341 and others, in hindsight it would be better if the te...
[14:52:11] <Nikerabbit>	 bd808: the bug you reported about LU... that's a blocker for making it work, right?
[14:53:17] <bd808>	 Nikerabbit: Only in the default "pull data from github" config. I don't think it will cause a problem on the WMF cluster
[14:53:45] <bd808>	 I haven't checked the clones on tin to make sure that the VE submodule is cloned properly though
[14:53:49] <Nikerabbit>	 bd808: oh right
[14:54:06] <Nikerabbit>	 bd808: will someone cherry-pick the fixes or will we wait for next branch?
[14:54:54] <bd808>	 There will be a branch today and I *think* that the newest branch is used for running scripts on tin
[14:55:26] <bd808>	 but we can cherry-pick and do a submodule update tomorrow in swat if that's a wrong assumption
[15:03:28] <wikibugs>	 6MediaWiki-Core-Team, 10SUL-Finalization: Run MassMessage to contact user talk pages of all affected accounts - https://phabricator.wikimedia.org/T90820#1128747 (10Asahiko) @Nemo_bis I copyedited P398 (a.k.a. "finalized text" according to T90820) after it was posted, and observed it being reflected on the actu...
[15:04:05] <anomie>	 Anyone: https://gerrit.wikimedia.org/r/#/c/197571/ (and its parent) need review before 1.25wmf22 gets cut.
[15:04:23] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-extensions-Scribunto, 5Patch-For-Review: Debug console not appearing; web console shows "TypeError: mw.scribunto is undefined" - https://phabricator.wikimedia.org/T93051#1128748 (10Anomie)
[15:04:39] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-extensions-Scribunto, 5Patch-For-Review: Debug console not appearing; web console shows "TypeError: mw.scribunto is undefined" - https://phabricator.wikimedia.org/T93051#1128750 (10Anomie) a:3Anomie
[15:08:11] <bd808>	 anomie: {{done}}
[15:08:33] <anomie>	 bd808: Thanks!
[15:09:32] <bd808>	 the parser test thing was collateral damage from an omg-fix-now thing that was rushed pretty hard yesterday
[15:09:42] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-extensions-Scribunto, 5Patch-For-Review: Debug console not appearing; web console shows "TypeError: mw.scribunto is undefined" - https://phabricator.wikimedia.org/T93051#1128784 (10Anomie) 5Open>3Resolved Fixed in time for 1.25wmf22 (unless Jenkins suddenly fails the mer...
[15:10:43] <anomie>	 bd808: Yeah. But someone posted on the previous two Scribunto patches "oh, by the way, we're breaking this again" rather than submitting and merging the fix.
[15:10:51] <bd808>	 blerg
[15:11:25] <bd808>	 parser tests are a train wreck of interdependencies :/
[15:30:26] <wikibugs>	 6MediaWiki-Core-Team, 10SUL-Finalization: Run MassMessage to contact user talk pages of all affected accounts - https://phabricator.wikimedia.org/T90820#1128867 (10Legoktm) @Nemo_bis: I can do that for english, and if people want to provide translations I can adjust the other languages as well.  @Asahiko, the...
[15:31:46] <legoktm>	 3 extensions are being loaded via extension registration in production :D
[15:37:52] <wikibugs>	 6MediaWiki-Core-Team, 10CirrusSearch: Wikipedia search engine not updating for moved article - https://phabricator.wikimedia.org/T93040#1128880 (10Manybubbles) >>! In T93040#1128564, @Manybubbles wrote: > Anyway, I'm going to get the script running soon to handle this.  It ought to work - it was built to detec...
[15:50:07] <wikibugs>	 6MediaWiki-Core-Team, 10Deployment-Systems, 5Patch-For-Review: Can't update l10n cache - https://phabricator.wikimedia.org/T92900#1128921 (10greg) I presume this is done/fixed now?
[15:51:06] <wikibugs>	 6MediaWiki-Core-Team, 10SUL-Finalization: Run MassMessage to contact user talk pages of all affected accounts - https://phabricator.wikimedia.org/T90820#1128928 (10Legoktm) >>! In T90820#1127929, @Nemo_bis wrote: > Over a million messages delivered by now, if I remember the previous edit count correctly (it's...
[15:54:55] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-Page-editing, 7I18n, 5Patch-For-Review: Long edit comments get entirely removed instead of truncated (error in cutting multibyte chars?) - https://phabricator.wikimedia.org/T85700#1128938 (10Umherirrender) With version 1.25wmf22 new long comments for revision will be trunc...
[15:55:16] <^d>	 manybubbles: I think the last of Matthias' changes will be fine, I just want unit tests.
[15:55:30] <manybubbles>	 ^d: cool!
[15:55:33] <^d>	 (and actually, it'll improve our index size by properly filtering more defunct fields)
[15:56:21] <manybubbles>	 yay
[16:05:31] <wikibugs>	 6MediaWiki-Core-Team, 10CirrusSearch: inefficient work of CirrusSearch in Russian Wikipedia - https://phabricator.wikimedia.org/T88724#1129024 (10Jdouglas)
[16:06:35] <^d>	 Search backend error during full_text search for '　' after 95.  Parse error on '　': Encountered "<EOF>" at line 1, column 1. [Called from CirrusSearch\ElasticsearchIntermediary::failure in /srv/
[16:06:35] <^d>	 mediawiki/php-1.25wmf20/extensions/CirrusSearch/includes/ElasticsearchIntermediary.php at line 98] in /srv/mediawiki/php-1.25wmf20/includes/debug/MWDebug.php on line 300
[16:06:43] <^d>	 manybubbles: We're still getting a lot of those ^
[16:06:52] <^d>	 My extra trim()ing didn't help
[16:06:59] <manybubbles>	 ah - hmm
[16:10:24] <wikibugs>	 6MediaWiki-Core-Team, 10CirrusSearch: inefficient work of CirrusSearch in Russian Wikipedia - https://phabricator.wikimedia.org/T88724#1129048 (10Jdouglas)
[16:11:30] <wikibugs>	 6MediaWiki-Core-Team, 10CirrusSearch: inefficient work of CirrusSearch in Russian Wikipedia - https://phabricator.wikimedia.org/T88724#1018869 (10Jdouglas)
[16:12:23] <wikibugs>	 6MediaWiki-Core-Team, 10CirrusSearch: inefficient work of CirrusSearch in Russian Wikipedia - https://phabricator.wikimedia.org/T88724#1018869 (10Jdouglas)
[16:14:51] <wikibugs>	 6MediaWiki-Core-Team, 6Multimedia, 3Multimedia-Sprint-2015-03-18, 5Patch-For-Review: SHA-1 file name support - https://phabricator.wikimedia.org/T1210#1129087 (10Gilles)
[16:23:11] <anomie>	 Ah, crazy bugs (re T93030). mb_strtolower doesn't turn ᾼ to ᾳ, because mb_strtolower only processes uppercase characters and ᾼ is apparently one of the very few "titlecase" characters in Unicode.
[16:31:58] <wikibugs>	 7Blocked-on-MediaWiki-Core, 10Continuous-Integration, 6Scrum-of-Scrums: MediaWiki installs in Jenkins frequently fail to access their sqlite database due to locks - https://phabricator.wikimedia.org/T89180#1129169 (10Krinkle)
[16:32:36] <wikibugs>	 7Blocked-on-MediaWiki-Core, 10Continuous-Integration, 6Scrum-of-Scrums: Jenkins jobs using MediaWiki frequently fail due to database locks  - https://phabricator.wikimedia.org/T89180#1129171 (10Krinkle)
[16:34:20] <wikibugs>	 6MediaWiki-Core-Team, 10Wikidata-Query-Service: Investigate Apache Accumulo for WDQ - https://phabricator.wikimedia.org/T88714#1129176 (10JanZerebecki) 5Open>3declined
[16:34:32] <wikibugs>	 6MediaWiki-Core-Team, 10Wikidata-Query-Service: Investigate InfoGrid for WDQ - https://phabricator.wikimedia.org/T88721#1129177 (10JanZerebecki) 5Open>3declined
[16:56:57] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-extensions-CentralAuth, 10SUL-Finalization: Add logging to RenameuserSQL - https://phabricator.wikimedia.org/T89681#1042502 (10Legoktm)
[17:06:34] <wikibugs>	 6MediaWiki-Core-Team, 7HHVM, 7Upstream: HHVM with FastCGI does not support streaming output - https://phabricator.wikimedia.org/T91468#1129320 (10Joe) I think we should upgrade HHVM to 3.6 as soon as possible when it comes out. It will make 95% of our patches redundant, a fact that will make me very happy :)...
[17:27:15] <wikibugs>	 6MediaWiki-Core-Team, 6Commons, 6Multimedia, 6operations: Deleted image pages do not show up as deleted until a manual action=purge - https://phabricator.wikimedia.org/T93052#1129391 (10Tgr) Side note: the less obvious but much easier way to upload screenshots is to just drag&drop the file into the comment...
[17:47:55] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-Page-editing, 7I18n, 5Patch-For-Review: Long edit comments get entirely removed instead of truncated (error in cutting multibyte chars?) - https://phabricator.wikimedia.org/T85700#1129462 (10Aklapper) p:5High>3Lowest Thanks! Lowering priority accordingly.
[17:52:41] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-Page-editing, 7I18n, 5Patch-For-Review: Long edit comments get entirely removed instead of truncated (error in cutting multibyte chars?) - https://phabricator.wikimedia.org/T85700#1129495 (10Aklapper) p:5Lowest>3Normal Uh, 2nd patch still waiting (sorry didn't see that...
[18:03:05] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-extensions-CentralAuth, 5Patch-For-Review: Global renames should leave local log entries on affected wikis - https://phabricator.wikimedia.org/T72851#1129544 (10Legoktm) a:3Legoktm
[18:03:34] <legoktm>	 Needs Review column is full at 25
[18:05:44] <bd808>	 I guess we should figure out how to get some of anomie and AaronSchulz 's things reviewed then
[18:11:37] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-extensions-SecurePoll, 5Patch-For-Review: Board election SecurePoll type fails with exception on creation - https://phabricator.wikimedia.org/T89425#1129613 (10Legoktm) 5Open>3Resolved
[18:12:41] <legoktm>	 one down :P
[18:14:44] <wikibugs>	 6MediaWiki-Core-Team, 10Deployment-Systems, 10Incident-20150312-whitespace, 5Patch-For-Review: scap's check_php_syntax() should check for text before '<?php' in PHP files - https://phabricator.wikimedia.org/T92534#1129627 (10Legoktm) 5Open>3Resolved
[18:18:35] <wikibugs>	 6MediaWiki-Core-Team, 10Incident-20150205-SiteOutage, 10MediaWiki-Debug-Logging, 10Wikimedia-Logstash, and 2 others: Decouple logging infrastructure failures from MediaWiki logging - https://phabricator.wikimedia.org/T88732#1129636 (10Legoktm)
[18:18:36] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-Debug-Logging, 5Patch-For-Review: Convert JobRunner.php to PSR-3 logging and add levels - https://phabricator.wikimedia.org/T87521#1129635 (10Legoktm)
[18:26:22] <wikibugs>	 6MediaWiki-Core-Team, 10SUL-Finalization, 5Patch-For-Review: Don't list users who are now attached to a global account on Special:UsersWhoWillBeRenamed - https://phabricator.wikimedia.org/T93044#1129665 (10Legoktm) 5Open>3Resolved
[18:26:27] <wikibugs>	 6MediaWiki-Core-Team, 10Deployment-Systems, 5Patch-For-Review: Can't update l10n cache - https://phabricator.wikimedia.org/T92900#1129666 (10bd808) 5Open>3Resolved >>! In T92900#1128921, @greg wrote: > I presume this is done/fixed now?  Yup. The underlying problem was fixed with https://gerrit.wikimedia....
[18:42:41] <wikibugs>	 6MediaWiki-Core-Team, 10SUL-Finalization: Run MassMessage to contact user talk pages of all affected accounts - https://phabricator.wikimedia.org/T90820#1129754 (10Ricordisamoa) It is astonishing that several long-time users are receiving the message: * https://commons.wikimedia.org/w/index.php?title=User_talk...
[19:10:51] <wikibugs>	 6MediaWiki-Core-Team, 10MassMessage, 5Patch-For-Review, 7Wikimedia-log-errors: Invalid parameter for message "logentry-massmessage-failure" - https://phabricator.wikimedia.org/T93110#1129863 (10Legoktm)
[19:13:04] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-General-or-Unknown, 5Patch-For-Review: HttpError should not be logged as an unhandled exception. - https://phabricator.wikimedia.org/T85795#1129872 (10hoo) 5Open>3Resolved a:3hoo
[19:15:12] <legoktm>	 could someone review https://gerrit.wikimedia.org/r/#/c/197372/ and https://gerrit.wikimedia.org/r/#/c/197479/ ? I've already been running it in production for a while
[19:29:24] <legoktm>	 bd808: thanks :)
[19:29:32] <bd808>	 yw
[19:46:19] <legoktm>	 bd808: also https://gerrit.wikimedia.org/r/197719 ? :) (haven't run that in production yet since I'll have to get that deployed...)
[19:49:59] <bd808>	 legoktm: Does LinkBatch pre-cache the data needed for Title::newFromText ?
[19:50:09] <legoktm>	 yup
[19:50:14] <bd808>	 neat
[19:50:31] <legoktm>	 it fills up LinkCache::singleton(), and then Title::isRedirect() checks with LinkCache
[19:50:59] <bd808>	 *nod* I was reading the class but not exactly sure that was the magic
[20:09:08] <wikibugs>	 6MediaWiki-Core-Team: [draft] MediaWiki Core Roadmap April - June 2015 (Q4 2014/2015) - https://phabricator.wikimedia.org/T93027#1130055 (10bd808)
[20:09:09] <wikibugs>	 6MediaWiki-Core-Team, 6Multimedia, 6Parsoid-Team, 6Release-Engineering, and 3 others: Prepare Platform April 2015 quarterly review presentation - https://phabricator.wikimedia.org/T91803#1130056 (10bd808)
[20:10:37] <wikibugs>	 6MediaWiki-Core-Team, 6Multimedia, 6Parsoid-Team, 6Release-Engineering, and 3 others: Prepare Platform April 2015 quarterly review presentation - https://phabricator.wikimedia.org/T91803#1096558 (10bd808)
[20:17:20] <wikibugs>	 6MediaWiki-Core-Team, 10Wikidata-Query-Service: Add bestRank to RDF export - https://phabricator.wikimedia.org/T92990#1130098 (10Manybubbles) Another way to do this that would be pretty simple is to teach blazegraph that ranks or ordered and just let you write queries like FILTER (?r >= "normal"^^wdo:rank)  Th...
[20:21:56] <wikibugs>	 6MediaWiki-Core-Team, 6Multimedia, 6Parsoid-Team, 6Release-Engineering, and 3 others: Prepare Platform April 2015 quarterly review presentation - https://phabricator.wikimedia.org/T91803#1130115 (10bd808)
[20:25:44] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-RfCs: RFC: Master/slave datacenter strategy for MediaWiki - https://phabricator.wikimedia.org/T88666#1130121 (10daniel) Related patch set, implementing a cache that is replicated across data centers: https://gerrit.wikimedia.org/r/#/c/187074/
[20:33:21] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-RfCs: RfC: AuthManager - https://phabricator.wikimedia.org/T91105#1130148 (10tstarling) IRC meeting 2015-03-04: http://tools.wmflabs.org/meetbot/wikimedia-office/2015/wikimedia-office.2015-03-04-21.00.html
[20:35:35] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-RfCs: RfC: AuthManager - https://phabricator.wikimedia.org/T91105#1130156 (10daniel) Action items from the IRC meeting:  *    SMalyshev is concerned about lack of model/controller split (TimStarling, 21:32:16) *    several people would like to see interface definitions (TimSta...
[20:36:05] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-RfCs: RfC: AuthManager - https://phabricator.wikimedia.org/T91105#1130166 (10Spage) @bd808 can you respond to the issues from the meeting? E.g. - SMalyshev is concerned about lack of model/controller split (TimStarling, 21:32:16) - several people would like to see interface de...
[20:41:01] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-RfCs: RfC: AuthManager - https://phabricator.wikimedia.org/T91105#1130187 (10bd808) @anomie has a work in progress patch at <https://gerrit.wikimedia.org/r/#/c/195297/> which defines interfaces and is starting to flesh out additional details of the implementation.  @csteipp ha...
[20:43:46] <csteipp>	 Looks like we forgot to address Stas'es concerns about model / controller..
[20:45:02] <anomie>	 Which means what, exactly?
[20:47:05] <AaronSchulz>	 csteipp: I just saw T93137 at https://phabricator.wikimedia.org/
[20:47:09] * AaronSchulz sighs
[20:47:41] * AaronSchulz discovers https://phabricator.wikimedia.org/tag/spam/
[20:47:51] <AaronSchulz>	 didn't know we had a tag for that, huh
[20:54:09] <legoktm>	 AaronSchulz: thanks. There's also an update for CentralAuth to go with that one: https://gerrit.wikimedia.org/r/#/c/197683/
[20:59:49] <wikibugs>	 6MediaWiki-Core-Team, 10SUL-Finalization: Create script to remove attached accounts from users_to_rename - https://phabricator.wikimedia.org/T93139#1130251 (10Legoktm) 3NEW a:3Legoktm
[21:02:00] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-extensions-CentralAuth, 5Patch-For-Review: Global renames should leave local log entries on affected wikis - https://phabricator.wikimedia.org/T72851#1130261 (10Legoktm) 5Open>3Resolved
[21:02:09] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-extensions-CentralAuth, 10SUL-Finalization, 5Patch-For-Review: Add logging to RenameuserSQL - https://phabricator.wikimedia.org/T89681#1130262 (10Legoktm) 5Open>3Resolved
[21:08:57] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-RfCs: RFC: Master/slave datacenter strategy for MediaWiki - https://phabricator.wikimedia.org/T88666#1130266 (10daniel)
[21:32:35] <wikibugs>	 6MediaWiki-Core-Team, 7Epic: Convert various core/extension cache users to the WAN cache as needed - https://phabricator.wikimedia.org/T93141#1130304 (10aaron) 3NEW a:3aaron
[21:40:01] <wikibugs>	 6MediaWiki-Core-Team: Configure 'enqueue' job queue type to be DC-local and in a dedicated runner loop - https://phabricator.wikimedia.org/T92863#1130335 (10aaron)
[21:40:02] <wikibugs>	 6MediaWiki-Core-Team, 7Epic: MediaWiki multi-datacenter investigation and work - https://phabricator.wikimedia.org/T88445#1130336 (10aaron)
[21:40:58] <wikibugs>	 6MediaWiki-Core-Team: Configure 'enqueue' job queue type to be DC-local and in a dedicated runner loop - https://phabricator.wikimedia.org/T92863#1122149 (10aaron)
[21:40:59] <wikibugs>	 6MediaWiki-Core-Team, 7Epic: MediaWiki multi-datacenter investigation and work - https://phabricator.wikimedia.org/T88445#1011867 (10aaron)
[21:41:54] <wikibugs>	 6MediaWiki-Core-Team, 7Epic: MediaWiki multi-datacenter investigation and work - https://phabricator.wikimedia.org/T88445#1130347 (10aaron)
[21:41:55] <wikibugs>	 6MediaWiki-Core-Team: Configure 'enqueue' job queue type to be DC-local and in a dedicated runner loop - https://phabricator.wikimedia.org/T92863#1122149 (10aaron)
[21:42:51] <wikibugs>	 6MediaWiki-Core-Team, 6operations, 7Varnish: Create HTTP verb and sticky cookie DC routing in VCL - https://phabricator.wikimedia.org/T91820#1130349 (10mark)
[21:50:35] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-extensions-Scribunto, 5Patch-For-Review: Debug console not appearing; web console shows "TypeError: mw.scribunto is undefined" - https://phabricator.wikimedia.org/T93051#1130382 (10MrStradivarius) That was quick! Thank you. :)
[21:52:24] <wikibugs>	 6MediaWiki-Core-Team, 10CirrusSearch: Wikipedia search engine not updating for moved article - https://phabricator.wikimedia.org/T93040#1130386 (10Manybubbles) I've run the script and it cleaned up that page.  The script is finding more errors than I'd like - I imagine we should be running this as a cron job t...
[21:53:22] <wikibugs>	 6MediaWiki-Core-Team: Memcached class with relayed delete() and daemon - https://phabricator.wikimedia.org/T88340#1130387 (10aaron) 5Open>3declined Folding into T88492
[21:54:42] <wikibugs>	 6MediaWiki-Core-Team, 5Patch-For-Review: Devise caching (memcached) strategy for multi-DC mediawiki - https://phabricator.wikimedia.org/T88492#1013160 (10aaron) See https://gerrit.wikimedia.org/r/#/c/187074/ and https://github.com/AaronSchulz/python-memcached-relay/blob/master/mcrelayd.py for the prototype dae...
[22:04:30] <wikibugs>	 6MediaWiki-Core-Team, 10SUL-Finalization: Run MassMessage to contact user talk pages of all affected accounts - https://phabricator.wikimedia.org/T90820#1130413 (10Keegan) Finalized text P398  This will start running shortly.  >>! In T90820#1129754, @Ricordisamoa wrote: > It is astonishing that several long-ti...
[22:15:49] <wikibugs>	 6MediaWiki-Core-Team, 6Multimedia, 6Parsoid-Team, 6Release-Engineering, and 3 others: Prepare Platform April 2015 quarterly review presentation - https://phabricator.wikimedia.org/T91803#1130446 (10ssastry)
[22:17:32] <wikibugs>	 6MediaWiki-Core-Team, 10MediaWiki-RfCs: RFC: Master/slave datacenter strategy for MediaWiki - https://phabricator.wikimedia.org/T88666#1130454 (10Spage) RFC meeting was held 2015-03-18 ([[ https://tools.wmflabs.org/meetbot/wikimedia-office/2015/wikimedia-office.2015-03-18-21.01.html | Meetbot log ]]) , conclus...
[22:33:00] <wikibugs>	 6MediaWiki-Core-Team, 10CirrusSearch: Wikipedia search engine not updating for moved article - https://phabricator.wikimedia.org/T93040#1130480 (10Krenair) A few weeks/months ago I noticed that "WYSIWYG Web builder" still appears in enwiki's search suggestions dropdown, but not the results (because it was dele...
[22:55:36] <AaronSchulz>	 Deskana: looks like you are chugging scotch over there :)
[23:03:59] <AaronSchulz>	 ^d: where you you get 4 from? I see the LCStore thing and an SMW test
[23:05:14] <^d>	 LCStore, SMW, WhosOnline, Cargo?
[23:05:22] <^d>	 Last one is commented out, so probably nvm
[23:05:57] <^d>	 Ah WhosOnline was out of date
[23:05:57] <^d>	 nvm
[23:07:10] <AaronSchulz>	 LCStore is interesting though, since it looking at what it was doing suggests a bug in core
[23:07:43] <^d>	 Holy crap that Cargo code is weird
[23:08:25] <^d>	 I'm just gonna pretend I didn't see it
[23:19:46] <^d>	 AaronSchulz: Should we be rolling back in core you mean?
[23:21:21] <AaronSchulz>	 I mean lots of stuff does try {} catch () for single statement errors or everything (calling rollback already)
[23:21:26] <AaronSchulz>	 you don't always want rollback
[23:21:34] <AaronSchulz>	 my problem is ignoreErrors() being left on
[23:22:14] <AaronSchulz>	 that just booby-traps the next callers
[23:24:39] <AaronSchulz>	 ^d: does MWExceptionHandler already handle MWException loops?
[23:24:59] <^d>	 Yeah you're right about it bleeding through
[23:25:38] <^d>	 We already catch "Exception caught inside exception handler"
[23:25:53] <AaronSchulz>	 *doesn't
[23:26:04] <AaronSchulz>	 so yeah, I need to find why that was there to begin with
[23:26:29] <^d>	 if ( $e instanceof MWException ) {
[23:26:29] <^d>	 			try {
[23:26:29] <^d>	 				// Try and show the exception prettily, with the normal skin infrastructure
[23:26:31] <^d>	 				$e->report();
[23:26:33] <^d>	 			} catch ( Exception $e2 ) {
[23:26:38] <AaronSchulz>	 yep, I was looking at that
[23:28:38] <MaxSem>	 btw TimStarling, instead of messing with FileRepo for WikiHiero I decided to stick to RESTBase, so no refactoring of common interface with Math and Score
[23:29:10] <AaronSchulz>	 why would hiero use FileRepo?
[23:29:23] <TimStarling>	 for storing PNGs
[23:29:29] <MaxSem>	 and SVGs
[23:29:30] <AaronSchulz>	 ugh
[23:29:47] <TimStarling>	 the new system generates SVGs on demand and renders them to PNGs
[23:30:52] <AaronSchulz>	 FileRepo is really only useful for standard uploaded content with File: pages
[23:31:13] <MaxSem>	 FileBackend, or whatever:P
[23:31:30] <AaronSchulz>	 I'd imagine the pngs are small, so restbase would handle those fine afaik
[23:31:45] <MaxSem>	 yup
[23:32:23] <AaronSchulz>	 it's also a good thing to test with
[23:54:15] <wikibugs>	 6MediaWiki-Core-Team, 10SUL-Finalization: Run MassMessage to contact user talk pages of all affected accounts - https://phabricator.wikimedia.org/T90820#1130742 (10Ricordisamoa) >>! In T90820#1130413, @Keegan wrote: > There was a bug that hit commons briefly, since fixed: [[https://commons.wikimedia.org/wiki/C...