[00:05:10] (03CR) 10CSteipp: [C: 04-1] "(1 comment)" [extensions/BayesianFilter] - 10https://gerrit.wikimedia.org/r/83673 (owner: 10Anubhavagarwal) [00:19:45] (03CR) 10Springle: [C: 031] "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/85787 (owner: 10EBernhardson (WMF)) [00:26:53] (03PS1) 10Bsitu: Fix some notification badge related issue [extensions/Echo] - 10https://gerrit.wikimedia.org/r/85942 [00:31:39] (03PS2) 10Bsitu: Fix some notification badge related issue [extensions/Echo] - 10https://gerrit.wikimedia.org/r/85942 [00:34:32] (03CR) 10Amire80: [C: 04-1] "The timestamp that appears in Ebrahim's link is "21 hours ago"." [core] - 10https://gerrit.wikimedia.org/r/85711 (owner: 10Ebrahim) [00:36:15] (03PS19) 10Physikerwelt: Backup [extensions/Math] - 10https://gerrit.wikimedia.org/r/85801 [00:36:29] (03CR) 10jenkins-bot: [V: 04-1] Backup [extensions/Math] - 10https://gerrit.wikimedia.org/r/85801 (owner: 10Physikerwelt) [00:42:18] (03CR) 10Amire80: [C: 031] "I didn't check the code deeply, but I tested it, and bug 54575 appears to be resolved." [extensions/Echo] - 10https://gerrit.wikimedia.org/r/85942 (owner: 10Bsitu) [00:44:06] ori-l: every time i use .background-image mixin it returns none [00:44:34] *returns nothing [00:47:10] (03CR) 10Amire80: [C: 04-2] "I reported it as Bug 54581." [core] - 10https://gerrit.wikimedia.org/r/85711 (owner: 10Ebrahim) [00:55:20] jdlrobson: are you @import "mediawiki.mixins.less"; -ing? [00:56:05] that's where that mixin is defined; you need to import that. [00:56:16] * jdlrobson face palm [01:08:04] (03CR) 10Daniel Friesen: [C: 04-1] "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/85926 (owner: 10MarkTraceur) [01:09:22] (03PS1) 10Legoktm: Add MassMessage extension [tools/release] - 10https://gerrit.wikimedia.org/r/85945 [01:11:14] (03CR) 10MarkTraceur: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/85926 (owner: 10MarkTraceur) [01:16:33] (03PS12) 10Mwalker: Further cleanup to Special:Version [core] - 10https://gerrit.wikimedia.org/r/65868 [01:17:02] (03CR) 10jenkins-bot: [V: 04-1] Further cleanup to Special:Version [core] - 10https://gerrit.wikimedia.org/r/65868 (owner: 10Mwalker) [01:17:34] (03PS16) 10Mwalker: Update Special:Version licensing for extensions [core] - 10https://gerrit.wikimedia.org/r/65299 [01:20:59] legoktm: why does MassMessage post as a reserved account instead of as the user sending the message? [01:21:56] mainly because of how the bot flag works [01:22:15] also because in cross-wiki messaging, the sender may not have an account on all wikis [01:23:16] marktraceur: ping [01:23:22] there's also the issue of your contributions being flooded [01:24:24] marktraceur: i'm wondering if core itself is able to register a GetBetaFeaturePreferences hook [01:31:26] who's Special:Contributions/75.90.225.247 a sock of? [01:31:41] wrong channel [01:38:25] (03PS2) 10Jdlrobson: Turn screen.css into screen.less [core] - 10https://gerrit.wikimedia.org/r/85920 [01:38:26] (03PS12) 10Jdlrobson: Vector: Add code for navigation collapsing [core] - 10https://gerrit.wikimedia.org/r/83591 [01:38:27] (03PS18) 10Jdlrobson: WIP Experimental: Apply mobile typography lessons to Vector on desktop [core] - 10https://gerrit.wikimedia.org/r/79948 [02:10:44] legoktm: pycharm is now open source [02:10:50] whaaaa [02:10:51] thats [02:10:52] awesome [02:11:19] ooh [02:11:21] pycharm3 [02:11:40] YuviPanda: wait, where does it say that? [02:11:49] legoktm: http://www.jetbrains.com/pycharm/ [02:11:54] > And much more, all available under the Apache 2 license [02:12:05] oh [02:12:09] thats the free community edition. [02:12:20] yeah, IIRC that wasn't OSS before [02:12:24] it was [02:12:32] what? [02:12:35] no! [02:12:39] only the intellij one was [02:12:40] intelliJ has always been open source [02:12:44] thats like the same thing [02:12:52] they're just rebranding [02:13:03] really? I looked at pycharm a while ago and it wasn't.. [02:13:18] Reedy: see if we can get pycharm3 professional edition licenses? :D [02:14:55] brrr, no MaxSem [02:16:09] hrm [02:16:19] YuviPanda: the terminal thingy looks really cool [02:16:29] legoktm: which terminal thingy [02:16:34] in pycharm [02:16:37] oh [02:16:41] im watching the whats new in 3.0 video [02:16:42] i just moved to emacs+evil [02:16:43] grr [02:17:22] ohhhh [02:17:25] this is so cool [02:17:28] "invert boolean" [02:17:55] switches the value, and fixes all the logic [02:27:34] legoktm: nice [02:27:38] i should give it a shot [04:02:13] (03PS2) 10Fantasticfears: Role: Add MLEB role [vagrant] - 10https://gerrit.wikimedia.org/r/85876 [04:10:16] (03CR) 10Ori.livneh: [C: 04-1] "(5 comments)" [vagrant] - 10https://gerrit.wikimedia.org/r/85876 (owner: 10Fantasticfears) [04:12:40] * YuviPanda pokes ori-l [04:12:49] Haven't forgotten [04:13:12] ori-l: it got merged, btw. [04:13:26] need to figure out how to do enable-role, disable-role and list-roles [04:16:03] (03CR) 10Ori.livneh: [C: 04-1] "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/85920 (owner: 10Jdlrobson) [04:33:55] (03CR) 10Jdlrobson: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/85920 (owner: 10Jdlrobson) [04:41:55] is there a read-only anon interface for mingle? i thought i had heard that mentioned before but dont see anything [04:43:20] oh hm, i just wasn't using https [04:58:27] ori-l: my vagrant stopped working :/ [04:58:33] i just git pull'd [04:59:00] What are you trying to do? [04:59:05] What error message do you get (if any)? [04:59:11] > The connection to the server was reset while the page was loading. [04:59:34] in firefox [04:59:38] lemme try a different browser [05:00:03] What URL? [05:01:58] http://127.0.0.1:8080/wiki/Special:MassMessage [05:02:16] legoktm: vagrant halt && vagrant up? [05:04:08] YuviPanda: did that twice now [05:04:09] same thing [05:04:10] chromium says "No data received" [05:04:11] i can just reset it [05:04:13] lets see if that works. [05:04:59] oh this is new [05:05:01] YuviPanda: https://dpaste.de/LwCmQ/raw/ [05:07:03] (03CR) 10GWicke: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/85926 (owner: 10MarkTraceur) [05:07:20] im just gonna destroy it and make a new one [05:08:10] legoktm: sorry, I'll try and help in a moment [05:10:17] Thanks, gwicke [05:10:43] no rush, i'm going to see how well destroying it and recreating it works [05:25:24] ori-l, YuviPanda: so I'm an idiot. I forgot I had something else running on 8080 >.> [05:25:31] awww [05:26:21] hm [05:26:36] when it creates a new instance, it doesn't auto-run update.php? [05:34:02] (03PS5) 10Legoktm: Use our own warning system instead of mw.notify [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/85174 [05:35:24] (03CR) 10Legoktm: [C: 032] Use our own warning system instead of mw.notify [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/85174 (owner: 10Legoktm) [05:35:34] (03Merged) 10jenkins-bot: Use our own warning system instead of mw.notify [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/85174 (owner: 10Legoktm) [06:02:50] (03CR) 10Siebrand: [C: 032] Added rank column [extensions/ContributionScores] - 10https://gerrit.wikimedia.org/r/85678 (owner: 10Mattsmith321) [06:02:53] (03Merged) 10jenkins-bot: Added rank column [extensions/ContributionScores] - 10https://gerrit.wikimedia.org/r/85678 (owner: 10Mattsmith321) [06:08:01] (03CR) 10Daniel Friesen: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/85926 (owner: 10MarkTraceur) [06:41:31] (03PS2) 10KartikMistry: WIP: Added integration tests boiler plate for TwnMainPage. [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/85688 [06:45:06] (03PS2) 10Tpt: (bug 45470) Adds support of JSON encoding and decoding [extensions/Scribunto] - 10https://gerrit.wikimedia.org/r/85426 [06:45:41] (03PS1) 10Legoktm: Throw errors if user input for the parser function is not valid [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/85956 [06:49:48] (03PS3) 10Tpt: (bug 45470) Adds support of JSON encoding and decoding [extensions/Scribunto] - 10https://gerrit.wikimedia.org/r/85426 [06:55:37] (03PS1) 10Legoktm: Use a delimiter in normalizeTargets to prevent conflicts [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/85958 [06:57:37] (03CR) 10Fantasticfears: "(1 comment)" [vagrant] - 10https://gerrit.wikimedia.org/r/85876 (owner: 10Fantasticfears) [06:57:48] (03PS1) 10PleaseStand: Installer: properly override default $wgLogo value [core] - 10https://gerrit.wikimedia.org/r/85959 [07:19:19] (03CR) 10Santhosh: "Reported https://bugzilla.wikimedia.org/show_bug.cgi?id=54592" [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/85366 (owner: 10Santhosh) [07:21:15] (03CR) 10Tpt: "PS2: Add asked changes (move to mw.text and add of some other tests)" [extensions/Scribunto] - 10https://gerrit.wikimedia.org/r/85426 (owner: 10Tpt) [08:01:21] (03CR) 10PleaseStand: "> A solution for the LuaStandalone engine is maybe to edit Scribunto_LuaStandaloneInterpreter::encodeLuaVar to return a non associative Lu" [extensions/Scribunto] - 10https://gerrit.wikimedia.org/r/85426 (owner: 10Tpt) [08:17:57] (03CR) 10Aklapper: [C: 031] "Tested locally, works. +1." [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/85883 (owner: 10Legoktm) [08:37:39] zeljkof: good morning :-] [08:39:58] hashar: morning :) [08:40:38] was looking at the tests again https://integration.wikimedia.org/ci/job/qa-browsertests-run/20/consoleFull [08:40:46] there is a bunch of failure with a 5 seconds timeout :/ [08:41:12] I am not sure what happens. Ideally we would get a bit more information from phantomJS but I have no clue how to enable traces/debugging informations [08:41:12] hashar: that should be easy to fix [08:41:26] looks like pages do not look the same when you use phantomjs [08:41:35] oh that is unfortunate [08:41:39] so the elements that we need are not there [08:41:51] is there a way to dump the rendered dom somehow ? [08:41:53] I know how to fix that, I just need some time [08:42:02] excellent! [08:42:04] yes, we can grab html and screenshots [08:43:24] should I reassign bug ""browsertests under phantomjs have failures"" https://bugzilla.wikimedia.org/show_bug.cgi?id=54489 to you? [08:43:39] feel free to do so [08:43:43] done :) [08:43:56] I will probably not have the time to fix it this week, I have a conference on Friday [08:44:01] preparing for that [08:44:05] I guess [08:44:07] great, thanks [08:44:25] how do you get the dom page dumped ? would it make sense to dump them on error and archive them in the build? [08:45:03] no :) [08:45:20] I can pretty much tell from the error message what the problem is [08:45:27] and I have to reproduce it on my machine anyway [08:45:34] so html dump does not really help [08:45:53] ok was just wondering whether it would be useful :-D [08:46:30] it could be useful, but we have other ways to fix problems :) [08:46:56] you are the expert! [08:56:35] (03CR) 10Dzahn: [C: 032] Optional / at the end of the url, allow for - to be in bugtype [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/85883 (owner: 10Legoktm) [08:57:32] (03CR) 10Dzahn: [V: 032] Optional / at the end of the url, allow for - to be in bugtype [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/85883 (owner: 10Legoktm) [08:57:43] few guys alone in #mediawiki-dev btw :) [09:23:20] we have so many channels hehe [09:24:12] hehehe, full ack [09:24:26] i joined mediawiki-dev by accident :p [09:55:46] mutante: but you'll stay by non-accident, right? ;) [10:07:35] (03PS1) 10Raimond Spekking: Consistency tweaks in preparation for adding extension to translatewiki.net [extensions/BayesianFilter] - 10https://gerrit.wikimedia.org/r/85966 [10:54:53] (03CR) 10Dzahn: [C: 032] Add bmo extension InlineHistory [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/76105 (owner: 10Aklapper) [11:00:59] (03CR) 10Dzahn: [V: 032] "since this is an opt-in thing in user settings (IF user.id && user.settings.inline_history.value == "on" ) it should be good. and we didn'" [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/76105 (owner: 10Aklapper) [11:07:16] (03CR) 10Dzahn: "deployed and ran checksetup.pl." [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/76105 (owner: 10Aklapper) [12:13:49] (03PS2) 10Krinkle: Exception: Clean up html document for error pages [core] - 10https://gerrit.wikimedia.org/r/84991 [12:14:11] (03CR) 10Krinkle: "(re-triggering to assert that the newly deployed version of jsduck works)" [core] - 10https://gerrit.wikimedia.org/r/84991 (owner: 10Krinkle) [12:14:26] (03CR) 10Krinkle: ".. which doesn't run on non-js commits." [core] - 10https://gerrit.wikimedia.org/r/84991 (owner: 10Krinkle) [12:14:50] (03PS8) 10Krinkle: Title: Add byte class to unicode class conversion for js [core] - 10https://gerrit.wikimedia.org/r/82040 [12:14:58] (03PS11) 10Krinkle: mw.Title: Rewrite from scratch (porting logic from Title.php) [core] - 10https://gerrit.wikimedia.org/r/83047 [12:21:43] (03CR) 10PleaseStand: [C: 04-1] "Comments on PS1 apply." [core] - 10https://gerrit.wikimedia.org/r/84991 (owner: 10Krinkle) [12:26:14] (03CR) 10Bartosz Dziewoński: [C: 031] Add wgRevisionId variable to ResourceLoader [core] - 10https://gerrit.wikimedia.org/r/81032 (owner: 10Parent5446) [12:27:16] (03CR) 10Krinkle: [C: 04-1] "Conflicts with a change made in SVN a while ago. Would prefer some further discussion. Can wing either way." [core] - 10https://gerrit.wikimedia.org/r/85637 (owner: 10Ori.livneh) [12:31:36] (03PS1) 10Hashar: point operations-puppet-doc to git:// url [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/85972 [12:32:17] (03CR) 10Hashar: [C: 032] point operations-puppet-doc to git:// url [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/85972 (owner: 10Hashar) [12:32:47] (03Merged) 10jenkins-bot: point operations-puppet-doc to git:// url [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/85972 (owner: 10Hashar) [12:37:01] (03CR) 10Krinkle: "(4 comments)" [core] - 10https://gerrit.wikimedia.org/r/85544 (owner: 10Ori.livneh) [12:38:59] (03CR) 10Krinkle: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/85153 (owner: 10Mattflaschen) [12:43:10] (03CR) 10Krinkle: "The difference however is that that is at keypress, not keydown. And it doesn't duplicate the handling of the keys (e.g. move cursor down " [core] - 10https://gerrit.wikimedia.org/r/82814 (owner: 10Bartosz Dziewoński) [12:43:45] (03Abandoned) 10Krinkle: When --wiki is provided, use MW_DB and MW_PREFIX constants to override $wgDBname and $wgDBprefix as expected. [core] - 10https://gerrit.wikimedia.org/r/9804 (owner: 10Platonides) [12:44:25] ori-l: i'm writing a script vagrant-labs that just wraps around helper.rb to provde role management and stuff to labsvagrant [12:45:19] (03CR) 10Krinkle: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/77686 (owner: 10Siebrand) [12:46:57] (03CR) 10Krinkle: "Maybe have those users load it from Common.js based on the presence of certain elements and/or mw.config pagename?" [core] - 10https://gerrit.wikimedia.org/r/70740 (owner: 10Ori.livneh) [12:48:52] (03CR) 10Krinkle: [C: 04-1] "Skin::getDefaultModules may be a better place. Note that when using modulestyles, the dependency will *not* be satisfied. So any modules d" [core] - 10https://gerrit.wikimedia.org/r/70740 (owner: 10Ori.livneh) [12:49:13] dang it, isn't there anything I can merge to test jsduck? I don't want to trigger it manually or with a dummy commit. [12:50:18] (03CR) 10Bartosz Dziewoński: "Admittedly I tested it rather briefly, but I didn't encounter any unexpected behavior. It would be easier to tell what this was for if it " [core] - 10https://gerrit.wikimedia.org/r/82814 (owner: 10Bartosz Dziewoński) [12:50:45] (03CR) 10Krinkle: "+1 on removing them, though agree with the test issues and the need for documenting the messages are no longer supposed to be used (probab" [core] - 10https://gerrit.wikimedia.org/r/80061 (owner: 10Chad) [13:02:43] (03CR) 10Krinkle: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/63364 (owner: 10Daniel Friesen) [13:07:26] Krinkle: [13:07:33] Krinkle: you could rebuild the last job [13:07:51] ... I knew that, I knew that! [13:07:56] https://integration.wikimedia.org/ci/job/mediawiki-core-jsduck/ [13:07:57] Thank you so much [13:07:59] press Rebuild Last :D [13:08:34] jsduck-publish not jsduck [13:08:43] that will publish the doc [13:08:53] Yes [13:09:06] and also run the test first [13:09:24] The test I already know still works, I'm more worried about the html format [13:09:37] like javascript, there's a lot of cases where it might pass lint, but doesn't do that you think. [13:09:40] it's quite loose [13:09:45] it'll make something of it [13:10:16] so mediawiki-core-jsduck only lint ? [13:11:03] Krinkle: https://integration.wikimedia.org/ci/job/mediawiki-core-jsduck/ws/docs/js/ [13:11:13] Krinkle: the mediawiki-core-jsduck does generate the doc to docs/js/ [13:11:23] so you can fetch all the generated files from the build and look at it locally [13:11:26] without publishing :-] [13:11:36] (03PS6) 10Krinkle: Drop support for old skins that don't use head element. [core] - 10https://gerrit.wikimedia.org/r/63364 (owner: 10Daniel Friesen) [13:11:43] hashar: is it possible to have different lint checks not only per gerrit project but also depending on what files/path is touched? [13:12:06] (03CR) 10Krinkle: "Rebased to resolve conflict in SkinTemplate (methods have been split since this patch)." [core] - 10https://gerrit.wikimedia.org/r/63364 (owner: 10Daniel Friesen) [13:12:11] mutante: what is your use case ? [13:12:20] mutante: yes (but I'll let hashar handle it further) [13:12:42] hashar: example, it would run apache-lint if i touch /apache/sites/* in puppet repo, not just in apache-config repo [13:12:46] gotcha [13:13:05] mutante: good point, I guess we need to add a new job to operations/puppet :-] [13:13:10] hmm, oh well, but they are .erb [13:13:14] but also Apache :p [13:13:29] Krinkle: https://integration.wikimedia.org/ci/job/mediawiki-core-jsduck/ws/docs/js/index.html :-D [13:13:47] Generated for branch master on Wed 25 Sep 2013 13:08:22 by JSDuck 4.10.4. [13:13:49] \O/ [13:13:51] that is funny [13:15:07] (03PS1) 10Aklapper: Indent data added by InlineHistory so it is still easy to skim through normal comments. Followup to Change 76105. [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/85978 [13:16:03] (03CR) 10Aklapper: [C: 031] "Tested on wmflabs and in-browser." [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/85978 (owner: 10Aklapper) [13:18:10] (03CR) 10Dzahn: [C: 032 V: 032] "simple CSS change" [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/85978 (owner: 10Aklapper) [13:24:51] (03CR) 10Krinkle: [C: 031] Drop support for old skins that don't use head element. [core] - 10https://gerrit.wikimedia.org/r/63364 (owner: 10Daniel Friesen) [13:25:59] (03Abandoned) 10Siebrand: Fix CodeSniffer errors and warnings [core] - 10https://gerrit.wikimedia.org/r/77686 (owner: 10Siebrand) [13:29:03] (03CR) 10Siebrand: [C: 032 V: 032] Consistency tweaks in preparation for adding extension to translatewiki.net [extensions/BayesianFilter] - 10https://gerrit.wikimedia.org/r/85966 (owner: 10Raimond Spekking) [13:30:22] (03CR) 10Siebrand: [C: 031] "i18n/L10n reviewed." [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/85956 (owner: 10Legoktm) [14:14:44] (03CR) 10Anomie: [C: 04-1] "Issues remain from PS1." [extensions/Scribunto] - 10https://gerrit.wikimedia.org/r/85426 (owner: 10Tpt) [14:22:48] (03CR) 10Anomie: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/83219 (owner: 10Legoktm) [14:24:52] anomie: hey - just wanted to ping and ask whether that Scribunto dev usability thing I fwded to you is reasonable, off-base, or something else [14:25:40] sumanah: Scribunto dev usability thing? [14:26:16] I ... did not forward it to you! [14:26:28] that would explain why you do not know what I mean. [14:26:31] That explains why I couldn't find it! [14:26:46] http://lists.wikimedia.org/pipermail/wikitech-l/2013-August/071189.html [14:27:01] and http://lists.wikimedia.org/pipermail/wikitech-l/2013-August/071210.html as the followup [14:27:09] I was wondering whether it would be good to follow up on those in some way [14:30:44] sumanah: I can't find a sample of the UI on the site he mentions, so I can't really say whether it's good or bad or would fit in with MediaWiki. But if someone wanted to work on UI, I see no reason to stop them in general. [14:31:03] got it [14:43:07] Reedy: around? [14:43:12] Indeed [14:43:19] yay! [14:43:20] https://bugzilla.wikimedia.org/show_bug.cgi?id=54497 [14:43:22] errr [14:43:22] I'm boggling at how cheap some things are in the US [14:43:26] https://gerrit.wikimedia.org/r/#/c/85837/ [14:43:49] to fix that issue [14:46:36] (03CR) 10Anomie: [C: 031] "Looks ok. Haven't tested." [core] - 10https://gerrit.wikimedia.org/r/82085 (owner: 10Brian Wolff) [14:53:05] (03CR) 10Anomie: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/78926 (owner: 10Brian Wolff) [14:56:15] (03CR) 10Fantasticfears: [C: 04-1] "The CLDR will be removed when disable mleb or education" [vagrant] - 10https://gerrit.wikimedia.org/r/85876 (owner: 10Fantasticfears) [14:58:59] (03CR) 10Anomie: "(1 comment)" [extensions/CommonsMetadata] - 10https://gerrit.wikimedia.org/r/80403 (owner: 10Brian Wolff) [15:02:04] (03PS1) 10Fantasticfears: Add Extension::Thanks to role::echo [vagrant] - 10https://gerrit.wikimedia.org/r/85983 [15:02:58] Reedy: OTOH, unless you live in the boondocks, cost of rent, power, etc is annoyingly high. [15:03:24] It's not exactly cheap over here either ;) [15:04:15] (03PS1) 10Pginer: CSS fixes for language selector on project tiles [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/85985 [15:08:20] (03CR) 10Kipcool: [C: 032 V: 032] owAddSyntrans typo; Downloads minor corrections [extensions/WikiLexicalData] - 10https://gerrit.wikimedia.org/r/85831 (owner: 10Hiong3-eng5) [15:13:26] (03PS1) 10Tobias Gritschacher: Add msg for wm-commons sitename shown in Wikidata sitelinks-table [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/85986 [15:34:36] YuviPanda: If you haven't seen the irc logs from url shortener RfC discussion yesterday you should take a look [15:34:38] https://www.mediawiki.org/wiki/Talk:Requests_for_comment/URL_shortener#IRC_meeting_2013-09-24 [15:36:13] There was some discussion about if Extension:ShortUrl could be modified/merged with whatever comes of the RfC [15:40:41] (03CR) 10Aude: [C: 04-1] "(2 comments)" [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/85986 (owner: 10Tobias Gritschacher) [15:42:50] (03PS2) 10Tobias Gritschacher: Add msg for wm-commons sitename shown in Wikidata sitelinks-table [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/85986 [15:44:25] (03CR) 10Krinkle: [C: 032] Remove legacy scripts from .jshintignore [core] - 10https://gerrit.wikimedia.org/r/85543 (owner: 10Ori.livneh) [15:48:05] (03CR) 10Aaron Schulz: [C: 032] Add option to chose what language to fetch file description in. [core] - 10https://gerrit.wikimedia.org/r/82085 (owner: 10Brian Wolff) [15:51:15] (03Merged) 10jenkins-bot: Add option to chose what language to fetch file description in. [core] - 10https://gerrit.wikimedia.org/r/82085 (owner: 10Brian Wolff) [15:52:01] nice going facebook: http://cl.ly/image/0A423R441i2T [15:54:05] facebook's language selector: http://cl.ly/image/2t0S3S2u1n04 [15:59:14] (03PS4) 10Andrew Bogott: Modify ldap schema for hosts. [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/84546 [15:59:15] (03PS1) 10Andrew Bogott: Rename 'hostname' to 'instanceid' in a bunch of places. [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/85991 [15:59:16] (03PS1) 10Andrew Bogott: Clear the host array for each IP. [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/85992 [15:59:31] (03CR) 10CSteipp: [C: 032] "Thanks!" [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/85958 (owner: 10Legoktm) [15:59:43] (03Merged) 10jenkins-bot: Use a delimiter in normalizeTargets to prevent conflicts [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/85958 (owner: 10Legoktm) [16:01:28] (03CR) 10Andrew Bogott: "The bug I was chasing turns out to be unrelated to this patch; it's fixed by https://gerrit.wikimedia.org/r/85992" [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/84546 (owner: 10Andrew Bogott) [16:02:17] (03PS3) 10Aklapper: Configure Bugzilla's Guided Bug Entry Form [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/74360 [16:03:58] (03CR) 10CSteipp: [C: 031] "Looks good. Thanks!" [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/85956 (owner: 10Legoktm) [16:15:01] (03PS1) 10Brion VIBBER: bug 23839: "Refreshing links table" -> "Refreshing links tables" [core] - 10https://gerrit.wikimedia.org/r/85997 [16:15:23] (03CR) 10Aklapper: [C: 031] "Some CSS cleanup, can be tested on http://boogs.wmflabs.org/enter_bug.cgi?product=Wikimedia&format=guided" [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/74360 (owner: 10Aklapper) [16:17:07] James_F: "Ellison also owns at least two military jets: a SIAI-Marchetti S.211, a training aircraft designed in Italy, and a decommissioned MiG-29, for which the US Government has refused him permission to import." [16:17:11] * AaronSchulz chuckles [16:17:25] lol [16:17:26] Lovely. [16:18:25] wot, is Oracle launching a preemptive strike on Google? [16:19:40] Probably needed to defend his island in Hawaii from foreign invasion [16:21:05] Google would probably try to see him coming in Street View. [16:22:37] (03CR) 10Aaron Schulz: [C: 032] bug 23839: "Refreshing links table" -> "Refreshing links tables" [core] - 10https://gerrit.wikimedia.org/r/85997 (owner: 10Brion VIBBER) [16:25:21] (03Merged) 10jenkins-bot: bug 23839: "Refreshing links table" -> "Refreshing links tables" [core] - 10https://gerrit.wikimedia.org/r/85997 (owner: 10Brion VIBBER) [16:25:41] \o/ little merges [16:30:08] (03PS1) 10Ori.livneh: ResourceLoader: expose registry; add mw.loader.getSize [core] - 10https://gerrit.wikimedia.org/r/85999 [16:32:00] RoanKattouw: ^ flagging this patch as requiring your input :) [16:33:02] (03CR) 10jenkins-bot: [V: 04-1] ResourceLoader: expose registry; add mw.loader.getSize [core] - 10https://gerrit.wikimedia.org/r/85999 (owner: 10Ori.livneh) [16:33:21] ori-l: Principle is fine by me. VE, for instance, has no real private members anywhere, just properties that happen to not have getters [16:34:11] Mostly our reason for that is that you can't do protected (as opposed to private, as opposed to public) members in JS, and we inherit the crap out of our classes [16:34:58] RoanKattouw: cool. So this actually means that getSize could live outside mediawiki.js. What's your preference there? [16:35:12] It's fine in mw.js I think [16:35:14] csteipp: should https://bugzilla.wikimedia.org/show_bug.cgi?id=53956 be closed? [16:35:38] awesome [16:36:02] Though, I guess arguably, the only reason you'd ever need is is instrumentation, and you can write that on the outside by inspecting .registry [16:36:05] So I think I'd prefer that [16:36:24] Mostly because someone might decide they want a different metric later, and then they're gonna have to do that on the outside anyway [16:37:04] I could go either way. I think it fits rather well with getState & co., but you're right that the use-case is rather specific [16:37:18] out of curiosity, what sort of different metric do you envision? [16:37:30] i guess you could take dependencies into account [16:38:03] Or messages, or whatever [16:38:10] oh yeah, good point [16:38:15] Actually I guess you can't track those from registry state directly [16:39:23] (03CR) 10Jdlrobson: "Bump.. please someone review / +2 - judging from the above review process this should be ready to merge?" [core] - 10https://gerrit.wikimedia.org/r/83591 (owner: 10Jdlrobson) [16:40:03] (03CR) 10Catrope: [C: 04-1] "(2 comments)" [core] - 10https://gerrit.wikimedia.org/r/85999 (owner: 10Ori.livneh) [16:41:50] (03PS1) 10Legoktm: Ensure memcache key has no control characters by taking hash of it [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/86000 [16:41:52] so does the message consideration change things? i.e., does it mean we should have in mediawiki.js because it could be expanded to cover messages as well? [16:42:42] I don't know how module messages are encoded, tbh [16:43:11] csteipp: just uploaded the md5 patchset. was there a specific reason you didnt +2 on https://gerrit.wikimedia.org/r/#/c/85956/ ? [16:43:40] duh: Nope, just didn't know if you wanted it merged or not [16:43:47] yes please :) [16:43:49] i think i'll at least split this into two patches [16:44:55] (03CR) 10Siebrand: "(1 comment)" [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/85986 (owner: 10Tobias Gritschacher) [16:48:52] (03PS2) 10Ori.livneh: ResourceLoader: expose registry as mediaWiki.loader.registry [core] - 10https://gerrit.wikimedia.org/r/85999 [16:51:41] (03CR) 10jenkins-bot: [V: 04-1] ResourceLoader: expose registry as mediaWiki.loader.registry [core] - 10https://gerrit.wikimedia.org/r/85999 (owner: 10Ori.livneh) [16:52:25] god damn it, jsduck. [16:54:00] (03PS3) 10Ori.livneh: ResourceLoader: expose registry as mediaWiki.loader.registry [core] - 10https://gerrit.wikimedia.org/r/85999 [16:55:48] (03PS2) 10Aaron Schulz: Factored out a small filebackend getContentType() function [core] - 10https://gerrit.wikimedia.org/r/85765 [16:56:24] (03CR) 10Ori.livneh: "Sorry, I missed Roan's prompt for Timo to weigh in, and I went ahead and removed getSize for now. I figured that I can start building the " [core] - 10https://gerrit.wikimedia.org/r/85999 (owner: 10Ori.livneh) [16:57:37] (03CR) 10jenkins-bot: [V: 04-1] ResourceLoader: expose registry as mediaWiki.loader.registry [core] - 10https://gerrit.wikimedia.org/r/85999 (owner: 10Ori.livneh) [16:57:57] OK, what the fuck. [16:58:53] Duplicate property name registry [16:58:56] and that's all i know :o [17:00:35] csteipp: yes if you could merge it that would be appreciated [17:00:59] (03CR) 10CSteipp: [C: 032] Throw errors if user input for the parser function is not valid [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/85956 (owner: 10Legoktm) [17:01:36] (03CR) 10CSteipp: [C: 032] Ensure memcache key has no control characters by taking hash of it [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/86000 (owner: 10Legoktm) [17:01:39] :) [17:01:40] csteipp: https://gerrit.wikimedia.org/r/86000 is the md5( $host ) patch [17:01:54] (03Merged) 10jenkins-bot: Throw errors if user input for the parser function is not valid [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/85956 (owner: 10Legoktm) [17:01:59] (03Merged) 10jenkins-bot: Ensure memcache key has no control characters by taking hash of it [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/86000 (owner: 10Legoktm) [17:02:58] (03PS4) 10Ori.livneh: ResourceLoader: expose registry as mediaWiki.loader.registry [core] - 10https://gerrit.wikimedia.org/r/85999 [17:03:16] (03PS7) 10Bartosz Dziewoński: Fix double-parsing of account creation messages. [core] - 10https://gerrit.wikimedia.org/r/47821 (owner: 10Parent5446) [17:03:35] (03CR) 10Bartosz Dziewoński: "I assume this fixes bug 52191 as well." [core] - 10https://gerrit.wikimedia.org/r/47821 (owner: 10Parent5446) [17:06:36] (03CR) 10jenkins-bot: [V: 04-1] ResourceLoader: expose registry as mediaWiki.loader.registry [core] - 10https://gerrit.wikimedia.org/r/85999 (owner: 10Ori.livneh) [17:07:23] i hope you're all enjoying the show [17:07:34] Nope. [17:08:17] csteipp: thanks! last request, can you comment on the bug so we can go ahead with the initial deployment on thurs? [17:08:40] duh: Sure, what was the number? [17:09:18] csteipp: https://bugzilla.wikimedia.org/show_bug.cgi?id=52723 [17:10:07] Hm.. Pushing to gerrit gives me: [17:10:09] hash mismatch [17:10:10] key_verify failed for server_host_key [17:10:10] fatal: Could not read from remote repository. [17:10:18] Trying again worked though [17:10:19] suspicious [17:10:33] (03PS5) 10Ori.livneh: ResourceLoader: expose registry as mediaWiki.loader.moduleRegistry [core] - 10https://gerrit.wikimedia.org/r/85999 [17:10:40] Krinkle: this happened multiple times to multiple people [17:10:43] we should report it [17:11:17] ^ Krenair [17:11:42] yeah [17:11:45] ori-l: you can poke at the registry in any proper debugger that supports inspecting private variables [17:11:50] happened to me as well [17:12:09] and getState and getModuleNames exposes it for anything that needs it automated without allowing it to do anything else. [17:12:46] 10:17 Krenair: [16:57:17] has gerrit's key changed? I'm getting key_verify failed for server_host_key [17:12:46] 10:17 Krenair: [16:58:08] .. and it's now working. weird. [17:12:51] 9/21 [17:13:39] Krinkle: I don't see what the issue is [17:13:44] (03CR) 10jenkins-bot: [V: 04-1] ResourceLoader: expose registry as mediaWiki.loader.moduleRegistry [core] - 10https://gerrit.wikimedia.org/r/85999 (owner: 10Ori.livneh) [17:14:29] I can also derive the content of the object using wireshark, that doesn't mean there isn't value to be had in making it accessible [17:14:46] (03PS2) 10Anubhavagarwal: added a hook on articleDelete [extensions/BayesianFilter] - 10https://gerrit.wikimedia.org/r/83673 [17:16:31] (03PS6) 10Ori.livneh: ResourceLoader: expose registry as mediaWiki.loader.moduleRegistry [core] - 10https://gerrit.wikimedia.org/r/85999 [17:25:50] hey bd808 [17:25:57] yeah, saw the logs. commented a bit on the shorturl one [17:26:09] YuviPanda|train: sweet [17:26:14] :) [17:33:53] brion: does https://gerrit.wikimedia.org/r/#/c/71966/4 look sane? [17:34:16] moment [17:34:27] (03CR) 10Nikerabbit: [C: 032] CSS fixes for language selector on project tiles [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/85985 (owner: 10Pginer) [17:34:33] (03Merged) 10jenkins-bot: CSS fixes for language selector on project tiles [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/85985 (owner: 10Pginer) [17:35:53] (03CR) 10Krinkle: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/85999 (owner: 10Ori.livneh) [17:37:39] (03PS7) 10Ori.livneh: ResourceLoader: expose registry as mediaWiki.loader.moduleRegistry [core] - 10https://gerrit.wikimedia.org/r/85999 [17:38:27] (03CR) 10Bartosz Dziewoński: [C: 032] "Okay, this works and looks straightforward. Let's do it." [core] - 10https://gerrit.wikimedia.org/r/81909 (owner: 10TTO) [17:40:57] Krinkle: thanks. To say the same thing, but in a less cranky fashion: I know that you can use the debugger to punch through the closure and get at the registry object, but I'd like to build something that makes this information accessible and informative to average users, and I think that for that to be successful it's vital not to require familiarity with the JS debugger [17:42:17] ori-l: Still, does that go beyond the console? e.g. do you expect to genuinely use this in a scripted fashion (even if its only a gadget or plugin for analytical/stat purposes?) [17:42:26] (03Merged) 10jenkins-bot: Add Special:CreateAccount to Special:Specialpages [core] - 10https://gerrit.wikimedia.org/r/81909 (owner: 10TTO) [17:42:40] (03CR) 10Brion VIBBER: "It looks like this changes the database-backed job queue size check from a quick row-count estimation (wildly inaccurate in details, inten" [core] - 10https://gerrit.wikimedia.org/r/71966 (owner: 10Aaron Schulz) [17:42:53] Krinkle: yeah, I think so. I'd like to have a gadget that actually gives you a visual breakdown of asset size [17:42:58] ori-l: What kind of data do you expect to extract with it, just want to document some kind of usecase (the commit message is a bit vague and makes it hard to touch this in the future with no specifics) [17:43:46] Krinkle: for an example of what I have in mind, see http://kaaes.github.io/timing/ (click 'Loadtime Breakdown') [17:44:00] ori-l: that data isn't and probably shouldn't reside inside mw.loader though, right? You'd only need the keys. The code that extracts and tracks that data could expose that on its own. [17:44:25] at which point you don't even need the keys [17:44:25] what do you mean, only need the keys? [17:44:34] module names [17:44:47] there are a lot of ways to slice the data and extract useful things from it [17:44:59] I had a getter for module size in an earlier version of that patch, dunno if you saw it [17:45:24] what, besides the names and state, does the registry contain that you expect to want to extract from it in a scripted fashion? [17:45:35] * Krinkle checks [17:45:37] module js/css size [17:45:47] see PS1 [17:46:45] but: you might also want to consider dependencies [17:47:25] That's all available on the server side and the same for all clients. That way you can include the real size (e.g. including messages as well), and have it for all modules, instead of just the ones loaded on the current page. [17:48:16] Krinkle, seriously, at some point you just need to trust that I'm not a complete idiot and that what I have in mind may need some fleshing out, but it's worth trusting it rather than treating it with skepticism [17:48:37] I don't see what the cost is of exposing it that you need to guard is so vigilantly [17:49:34] This is about getSize, not the registry. I'm not treating it with skepticism. I'm both trying to understand it (as I've done very little with event logging), and two minds do know more than one. "No" is acceptable answer, too. [17:51:42] I think the extent to which average users can be relied on to actually care for and scrutinize profiling data has a lot to do with how easy it is to acquire and understand, and I think the easiest and most compelling way is to have the ability to load some on-page display that contains this information, visualized in such a way that makes bottlenecks easy to spot [17:52:04] (03CR) 10Ryan Lane: [C: 032] Clear the host array for each IP. [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/85992 (owner: 10Andrew Bogott) [17:53:15] I also think that the way the registry object is structured is essential to grokking ResourceLoader, and the fact that it can't easily be manipulated in the debug console is frustrating, since there's no hard reason for it as far as I can tell. [17:53:15] (03PS8) 10Krinkle: ResourceLoader: expose registry as mediaWiki.loader.moduleRegistry [core] - 10https://gerrit.wikimedia.org/r/85999 (owner: 10Ori.livneh) [17:53:25] (03CR) 10Krinkle: [C: 031] ResourceLoader: expose registry as mediaWiki.loader.moduleRegistry [core] - 10https://gerrit.wikimedia.org/r/85999 (owner: 10Ori.livneh) [17:53:55] thanks, I appreciate your confidence and will try to gratify it by building things that don't totally suck :) [17:54:08] ori-l: What also plays a role in my guarding of mw.loader is what people have done with it in the wild, which can sometimes spread very quickly and cause a maintenance nightmare. [17:54:59] I don't worry much about code written in a git repository. [17:55:51] ori-l: Hmm.. on that note, maybe expose it as mw.loader.registry and move the docs down instead of having two pointers? [17:56:19] It's fine as-is, but that would make it even more transparent I think. [17:56:49] Krinkle: I thought about that, but there's a lot of code throughout the file that references the registry object, and I think the inline reading experience is much better if registry is documented at the top [17:57:00] k [17:57:08] (03PS1) 10Kaldari: Moving dependnecies from ext.echo.base to more specific modules [extensions/Echo] - 10https://gerrit.wikimedia.org/r/86006 [17:57:11] and jsduck doesn't allow private and public to have the same name [17:57:16] (03PS1) 10Aaron Schulz: Support rate-limiting thumbnail generation [core] - 10https://gerrit.wikimedia.org/r/86007 [17:57:35] I realized that after staring at addStyleTag: newStyleTag, for a bit [17:57:51] ori-l: what I do in VE for those cases is document the private one as the public one, and have the real public object key not have a /** comment. [17:58:05] so no duplicate indexing [17:58:12] (03PS2) 10Kaldari: Moving dependnecies from ext.echo.base to more specific modules [extensions/Echo] - 10https://gerrit.wikimedia.org/r/86006 [17:58:17] ori-l: oh that reminds me -- can we devise an instrumentation for ResourceLoader that measures the relative code size of each module loaded? [17:58:38] we could do that as offline analysis more easily maybe though [17:58:54] brion: welcome to the conversation :P that's roughly what we're talking about [17:59:02] i just want to make a module size blame-map for that 'commons has a meg of javascript' issue :D [17:59:04] awesome [17:59:27] ori-l: but out of curiousity (this is what I was trying to ask earlier) - why not get the size from the server-side? I'm not saying it makes more sense, just trying to see what I missed. [17:59:44] (03CR) 10Aaron Schulz: "I think by the time that gets slow, you wouldn't want to be using that class anyway for other reasons...we don't." [core] - 10https://gerrit.wikimedia.org/r/71966 (owner: 10Aaron Schulz) [17:59:48] Krinkle: is it sent down to clients? [18:00:22] I mean, you can determine from a one-off maintenance which modules are loaded on commonswiki, and determine the size of the module size (scripts, styles, messages, it's all available) [18:00:42] by default that is, not per-user of course (the might have a ton of gadgets enabled) [18:00:51] if one calculates the size server-side, one could drop an extra integer field into the output along with the closure (or css string, or message array) [18:00:56] then it could be tallied up client-side [18:01:03] brion: it's not necessary, though [18:01:13] oh can we get at the source of each closure? that'd do it [18:01:18] right [18:01:21] \o/ [18:01:31] i'll just leave you guys to it then, you seem to have the question well in hand :D [18:01:36] see https://gerrit.wikimedia.org/r/#/c/85999/1/resources/mediawiki/mediawiki.js [18:01:57] (removed from subsequent patchsets but to be introduced elsewhere) [18:02:20] Krinkle: but gadgets are important [18:02:40] sure, they'd be included. They are no different. [18:02:41] and it'd be good to have users serve as eyes and ears for JS perf [18:02:55] rather than rely on us being explicitly aware of a problem [18:03:25] i.e., if admins are considering enabling a gadget by default [18:03:37] e.g., not i.e. [18:04:22] it just seems like emitting by the server by default is much more invasive and controversial than just deriving it on the client-side, which is simple to do [18:04:37] gotta run for a bit [18:05:09] ori-l: Having it per users is nice indeed, they can look it up themselves without needing us and report it accordingly. [18:05:21] Great. I'm in. [18:05:50] :) thank you [18:06:34] (03CR) 10Krinkle: [C: 032] ResourceLoader: expose registry as mediaWiki.loader.moduleRegistry [core] - 10https://gerrit.wikimedia.org/r/85999 (owner: 10Ori.livneh) [18:09:30] (03Merged) 10jenkins-bot: ResourceLoader: expose registry as mediaWiki.loader.moduleRegistry [core] - 10https://gerrit.wikimedia.org/r/85999 (owner: 10Ori.livneh) [18:14:07] hi marktraceur just checking whether micru and you are aware about the deadline for the GSoC mentor report in 48h [18:14:35] Yuuup [18:14:40] I'm just about to hit the button [18:14:41] marktraceur, (micru is attending mentors summit so mho he could do the big chunk of the homework) [18:14:47] ajh ok :)\ [18:14:50] Oh, he is? [18:15:10] yesss [18:17:13] (03CR) 10Ryan Lane: [C: 04-1] "(2 comments)" [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/84546 (owner: 10Andrew Bogott) [18:17:13] qgil: Done [18:17:32] wow, is that fast? (I can see the evaluation forms myself - I think) [18:17:42] The report isn't that long [18:17:52] It's like "How does that make you feeeeel" and then "pass/fail" [18:18:26] :D [18:19:05] (03PS1) 10Umherirrender: Use always context for messages on ImagePage [core] - 10https://gerrit.wikimedia.org/r/86008 [18:21:09] (03CR) 10MarkTraceur: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/85926 (owner: 10MarkTraceur) [18:21:20] Man, that patch. Not easy. [18:21:56] (03CR) 10Umherirrender: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/86008 (owner: 10Umherirrender) [18:21:59] * YuviPanda|train guesses what patch that is [18:22:03] gwicke: Daniel brings up a good point, and absolute URLs would "solve" it - is there room in the spec for both? [18:22:07] * YuviPanda|train guesses "agora style changes" [18:22:12] Heh, guess again [18:22:25] YuviPanda|train: I'm breaking _every_ part of core now. Diversification. Important stuff. [18:22:32] right [18:22:34] * marktraceur silences YuviPanda|train [18:22:54] ah, nice [18:23:10] marktraceur: in return for you attempting to break stuff on core, perhaps I should break stuff in UW :P [18:23:23] I'm pretty sure you'd do that anyway [18:23:55] i haven't done that yet at all, y'know [18:24:06] You, um [18:24:09] You kinda have [18:24:09] (03PS2) 10Andrew Bogott: Rename 'hostname' to 'instanceid' in a bunch of places. [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/85991 [18:24:10] (03PS5) 10Andrew Bogott: Modify ldap schema for hosts. [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/84546 [18:24:11] (03PS2) 10Andrew Bogott: Clear the host array for each IP. [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/85992 [18:24:17] merging your EL patch doesn't count? :P [18:24:22] https://bugzilla.wikimedia.org/show_bug.cgi?id=52611 [18:25:05] marktraceur: well, if we break what people may think is an important user preference, one that we think a lot of people will make a fuss about, and then nobody seems to give a shit, did we really break anything? :P [18:25:09] i kid, i kid [18:25:28] YuviPanda: Actually, if you wanted to use your stat1 access for good, you could run a query on how many people use that preference, and how many of them have uploaded recently [18:25:44] If you can't do that then we can wait for proper EL data [18:26:00] I'm cooking up a way to do that automatically [18:26:12] marktraceur: also, I'll note that ironholds *already* has that exact data for every preference ever [18:26:22] * marktraceur imagines YuviPanda in a kitchen with an apron and chef's hat, coding [18:26:31] Ir damnit [18:26:34] (03PS3) 10Andrew Bogott: Rename 'hostname' to 'instanceid' in a bunch of places. [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/85991 [18:26:35] (03PS3) 10Andrew Bogott: Clear the host array for each IP. [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/85992 [18:28:28] marktraceur: sorry, got dc. missed anything? [18:28:46] Naw [18:29:12] ok [18:37:19] cscott: thoughts on setting up a vagrant puppet role for togetherjs? [18:38:04] does the visual editor vagrant-puppet role work on labs? [18:38:21] cscott: yup. [18:38:29] was merged y'day [18:38:51] cscott: want me to setup an instance on the VE project with that role, so you can then experiment? [18:38:52] it shouldn't be hard to add the togetherjs extension to that, but i was kinda hoping to do "real work" (ie, on features not ops) on it today. [18:39:04] cscott: oh, that's fine. [18:39:12] cscott: I'll poke you later sometime :) [18:39:13] YuviPanda: but ping me, oh say tomorrow [18:39:17] yeah [18:39:19] sounds like a good idea [18:43:47] YuviPanda: do you still need me to review anything? [18:43:51] I'm available now [18:44:02] ori-l: the patch was merged, but you should still look at it [18:44:11] let me find it [18:44:28] ori-l: primarily, https://gerrit.wikimedia.org/r/#/c/85814/ [18:44:56] ori-l: i need to do one more patch, which is adding a 'labsvagrant' script that has 'provison', 'list-roles', 'enable-role', 'disable-role' commands [18:45:04] which will re-use code from helpers.rb and friends [18:45:17] and then I'll have to figure out if I should put that script in the vagrant repo or the puppet repo [18:45:28] vagrant repo makes lesser sense since this is used only in labs [18:45:35] looks good at a glance, especially if it works [18:45:48] ori-l: it works on my testing :D [18:45:54] per style guide, your arrows should be aligned, and octal file permissions should be quoted [18:46:04] which style guide? :P [18:46:24] I... should probably read the upstream puppet one [18:46:27] http://docs.puppetlabs.com/guides/style_guide.html & https://wikitech.wikimedia.org/wiki/Puppet_coding [18:46:28] [18:46:35] ori-l: thoughts on the labsvagrant script? [18:46:45] YuviPanda: oh, and documenting this whole process would also be nice. i'd like to actually make the ve-vagrant-puppet instance myself, if it is well-documented how to do so. [18:46:46] it looks great [18:46:57] cscott: indeed, I will do that! [18:47:08] thanks for working on this! [18:47:11] ori-l: where do you think I should put it? in the ops repo? [18:47:30] but that depends on helpers.rb... [18:47:37] by 'the labsvagrant script' do you mean https://gerrit.wikimedia.org/r/#/c/85814/13/modules/labsvagrant/files/vagrant-puppet-runner.bash ? [18:47:44] I think we may be talking about different things [18:48:14] ori-l: an extension of that to do enable-role, disable-role and list-roles [18:48:21] ori-l: but in ruby, so I can reuse helpers.rb code from vagrant [18:49:18] where would i be running 'enable-role'? on my machine or on the VM itself? [18:49:29] on the VM itself [18:49:35] nothing's done from your machine [18:49:58] makes sense to have a standalone ruby tool [18:50:17] hmmm [18:50:19] it'd be nice to have it for local VMs too, i accidentally run 'vagrant' inside he VM 50 times a day [18:50:24] hehe :P [18:50:37] right, so I can put it in mediawiki/vagrant.git [18:50:45] yeah, sure [18:51:00] but, before you get started on that [18:51:20] i think it's clear that the notion of 'role' is somewhat confused [18:51:40] so i was going to propose some additional means of managing individual extensions [18:51:56] and see what people on wikitech-l think [18:51:59] sure! [18:52:13] cool. so give me a day or three :) [18:52:38] ori-l: I guess I should just replicate teh equivalent for now, at least so cscott can have his puppetized labs instance? [18:53:35] (03CR) 10Ori.livneh: "There are FIXMEs for image embedding -- what's the status of those?" [core] - 10https://gerrit.wikimedia.org/r/83591 (owner: 10Jdlrobson) [18:53:53] YuviPanda: I mean, sure, if you promise not to be annoyed if it has to be modified somewhat soon [18:54:05] ori-l: sure, as long as you tell me when you modify it :P [18:54:08] for now, at least. [18:54:40] (03CR) 10Ori.livneh: [C: 032] Add Extension::Thanks to role::echo [vagrant] - 10https://gerrit.wikimedia.org/r/85983 (owner: 10Fantasticfears) [18:54:44] (03Merged) 10jenkins-bot: Add Extension::Thanks to role::echo [vagrant] - 10https://gerrit.wikimedia.org/r/85983 (owner: 10Fantasticfears) [18:54:47] well, i'm not itching to puppetize yet. pinklake.wmflabs.org seems to be working well at the moment, i just need to get more features implemented. ;) [18:55:02] shushhhh, cscott [18:55:36] (03CR) 10Brion VIBBER: "Note that in a default installation, this won't protect against DoS attackers triggering thumbnailing via wikitext parsing or API imageinf" [core] - 10https://gerrit.wikimedia.org/r/86007 (owner: 10Aaron Schulz) [18:55:39] feeeaaatures. [18:57:05] Amgine: hrm? [18:57:37] like "braaaaainnnssss", only more frankensteinish. [18:57:48] * cscott is a "senior features engineer". i write features for seniors. [18:58:14] Can I blame you for cialis? [18:58:25] I'm Batmobile Developer (International) [18:59:16] my employment history: olpc = computers for young people. litl = computers for old people. back to olpc = young people again. wikimedia = senior features. it's a pattern. [18:59:24] Amgine: No, but you can blame him for Rascals. [18:59:32] [19:00:32] omg. A car just drove by thumping M.C. Hammer. [19:01:16] Amgine: Well don't just stand there, they'll kill him! [19:01:19] Call the cops! [19:01:20] (03CR) 10Jdlrobson: "To my knowledge the existing embed mixin doesn't work with the following:" [core] - 10https://gerrit.wikimedia.org/r/83591 (owner: 10Jdlrobson) [19:02:17] Geneva is soo weird. [19:03:59] (03CR) 10Yuvipanda: [C: 032] "Seems to work!" [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/85154 (owner: 10DMaggot) [19:04:07] (03Merged) 10jenkins-bot: Uncomment the options Schema Definition [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/85154 (owner: 10DMaggot) [19:04:28] ori-l: heading to bed now, will hopefully wake up to your email :P [19:04:39] day or three I said! [19:05:12] ori-l: 50% chance! [19:05:21] (that's how it works, right!?) [19:06:23] ori-l: less makes this https://gerrit.wikimedia.org/r/#/c/79948/ so much easier [19:06:46] https://gerrit.wikimedia.org/r/#/c/79948/18/skins/vector/variablesBeta.less :D < StevenW superm401 spagewmf [19:10:57] (03CR) 10Ori.livneh: "(4 comments)" [core] - 10https://gerrit.wikimedia.org/r/85544 (owner: 10Ori.livneh) [19:11:28] (03PS2) 10Ori.livneh: Clean-up of skins/common/IEFixes.js [core] - 10https://gerrit.wikimedia.org/r/85544 [19:11:59] jdlrobson: yay :P [19:13:59] (03CR) 10Ori.livneh: "> Personally I think this is a tiny trade off to make but if we want to invest time in fixing the mixin for those cases as part of this pa" [core] - 10https://gerrit.wikimedia.org/r/83591 (owner: 10Jdlrobson) [19:16:21] (03CR) 10Brion VIBBER: "I think that case should be done with another mixin (which would also be semantically clearer, in that it could be labeled as 'background-" [core] - 10https://gerrit.wikimedia.org/r/83591 (owner: 10Jdlrobson) [19:18:04] (03CR) 10Ori.livneh: "Krinkle, you mean this commit, right?" [core] - 10https://gerrit.wikimedia.org/r/85637 (owner: 10Ori.livneh) [19:20:55] brion: what is 'linear-gradient(transparent, transparent)' doing exactly? [19:21:00] other than giving me a headache [19:21:15] oh, never mind, I get it [19:21:32] what, the headache? [19:22:13] no, not the headache :P [19:22:35] :P [19:23:31] looks like an svg background hack [19:23:39] Yuvipanda: go to sleep. Wikimedia prefers to suck the souls of well-rested devs. [19:24:04] ori-l: yeah it's to keep the svg from being loaded broken by bowsers that don't grok svg [19:24:06] Amgine: I ate rice, but I'm pretty sure it has spoilt. peeling a mosambi so I can wash the weird taste off my mouth with fruit [19:24:12] Amgine: you're a bit macabre tonight, hm? [19:24:21] it's a hack based on knowing that browsers that support gradients also support svg in css backgrounds :P [19:24:29] brion: yep, figured that out [19:24:39] Perhaps, ori-l. but 'tis true nonetheless. [19:24:50] i guess it'd be OK to proliferate various mixins for embed [19:25:00] they aren't included in the output unless they're actually used [19:25:17] yeah, it just costs a little time and memory in the parse, but very little i would think [19:25:19] so they don't cause any bloating in the generated CSS at all [19:25:29] right [19:25:57] I think PHP has an internal cache for the file functions that embed / embeddable call, too [19:35:19] (03PS1) 10Jeroen De Dauw: Add HookRunnerTest [extensions/ParserHooks] - 10https://gerrit.wikimedia.org/r/86015 [19:35:20] (03PS1) 10Jeroen De Dauw: Add to HookRegistrantTest [extensions/ParserHooks] - 10https://gerrit.wikimedia.org/r/86016 [19:37:43] (03CR) 10Jeroen De Dauw: [C: 032] Add HookRunnerTest [extensions/ParserHooks] - 10https://gerrit.wikimedia.org/r/86015 (owner: 10Jeroen De Dauw) [19:37:51] (03CR) 10Jeroen De Dauw: [C: 032] Add to HookRegistrantTest [extensions/ParserHooks] - 10https://gerrit.wikimedia.org/r/86016 (owner: 10Jeroen De Dauw) [19:43:49] (03PS1) 10Ebrahim: [WIP] Sharing native digits CSS with print output [core] - 10https://gerrit.wikimedia.org/r/86018 [19:51:42] i want to develop a bot to create a page similar as https://en.wikipedia.org/wiki/User:Emijrp/List_of_Wikipedians_by_number_of_edits. need some assistance regarding this [19:59:01] (03PS1) 10Jeroen De Dauw: 1.1 release [extensions/ParserHooks] - 10https://gerrit.wikimedia.org/r/86022 [19:59:17] (03CR) 10Jeroen De Dauw: [C: 032] 1.1 release [extensions/ParserHooks] - 10https://gerrit.wikimedia.org/r/86022 (owner: 10Jeroen De Dauw) [20:01:41] (03PS1) 10MarkTraceur: Fixes from security review [extensions/BetaFeatures] - 10https://gerrit.wikimedia.org/r/86023 [20:04:45] csteipp: ^^ there are the suggested fixes, if you care to review :) [20:04:54] Or, you know, I can bug bd808 like always. [20:05:04] marktraceur: I will after I finish deploying [20:05:16] (03CR) 10Aaron Schulz: "Yeah, I thought of that. I could maybe use a comment." [core] - 10https://gerrit.wikimedia.org/r/86007 (owner: 10Aaron Schulz) [20:05:45] * bd808 will let csteipp have some of the marktraceur merge fun [20:07:43] It may be he's too busy [20:08:04] (03PS2) 10Aaron Schulz: Support rate-limiting thumbnail generation [core] - 10https://gerrit.wikimedia.org/r/86007 [20:09:04] marktraceur: is your review "urgent"? Riding a deploy train soon? [20:09:08] Naw [20:09:25] bd808: We still need to make sure we have features, and we need to make sure the deploy train makes sense... -.- [20:09:43] I wanted to get it to the state where we *could* deploy it, and then we wouldn't need to wait later on [20:10:12] Duh! [20:10:18] o/ [20:10:22] good plan. Also that will make it easier to push back against major feature change requests :) [20:10:30] Yup [20:10:42] "Do you *want* me to bother Chris and Sam again? Really?" [20:11:12] Though the next iteration's backlog looks like it's roughly all frontend features [20:11:18] * bd808 goes back to code review on candidate project [20:11:31] * Isarra hugs ^d and runs away. [20:11:35] Except for unit tests and "Smooth out the process for moving from a beta feature to a non-beta feature" [20:11:48] Which I'm still not sure about. [20:11:59] brion: anything else with 71966 ? [20:12:09] * marktraceur lassoos Isarra [20:12:19] she's quick that one [20:12:22] you may miss [20:12:40] or need a stronger rope [20:12:43] mwalker: Compounded by the fact that I can't lassoo [20:12:46] * Isarra runs into mwalker. [20:12:52] mwalker: Good block [20:13:09] * mwalker falls over [20:13:10] Hey guys. 'Sup? [20:13:20] * Isarra pulls mwalker up and tries to look innocent. [20:13:33] Isarra: You know. Breaking wikis. The us'. [20:13:41] well; apparently I'm being run over by crazed wikipedians [20:13:45] Ah, yes. I know it well. [20:13:54] not much [20:13:57] (03PS1) 10CSteipp: Update CentralAuth to master [core] (wmf/1.22wmf18) - 10https://gerrit.wikimedia.org/r/86024 [20:14:05] Sub. [20:14:12] sandwhich? [20:14:13] AaronSchulz to the not-muchth power. [20:14:20] (03CR) 10CSteipp: [C: 032] Update CentralAuth to master [core] (wmf/1.22wmf18) - 10https://gerrit.wikimedia.org/r/86024 (owner: 10CSteipp) [20:14:27] (03CR) 10Brion VIBBER: [C: 031] "True enough. :)" [core] - 10https://gerrit.wikimedia.org/r/71966 (owner: 10Aaron Schulz) [20:14:28] With bacon. [20:14:40] mwalker: Sandthatone [20:15:28] * mwalker has parse error at line 1 character 5; expecting or missing context [20:17:00] (03PS1) 10CSteipp: Update CentralAuth to master [core] (wmf/1.22wmf17) - 10https://gerrit.wikimedia.org/r/86025 [20:17:25] qgil: do you know of a good first person to have a volunteer contact if they're interested in developing tools? [20:18:46] (03Merged) 10jenkins-bot: Update CentralAuth to master [core] (wmf/1.22wmf18) - 10https://gerrit.wikimedia.org/r/86024 (owner: 10CSteipp) [20:19:03] Isarra: ordinarily I would say good choice -- but right now, having filled myself with curry, bacon just doesnt sound good [20:20:16] There should have been bacon in the curry. [20:20:36] it was a pork curry [20:20:43] some bacon would not have gone amiss [20:24:10] (03CR) 10CSteipp: [C: 032] Update CentralAuth to master [core] (wmf/1.22wmf17) - 10https://gerrit.wikimedia.org/r/86025 (owner: 10CSteipp) [20:26:53] (03Merged) 10jenkins-bot: Update CentralAuth to master [core] (wmf/1.22wmf17) - 10https://gerrit.wikimedia.org/r/86025 (owner: 10CSteipp) [20:44:58] (03CR) 10Brion VIBBER: [C: 031] "Any JS issues blocking further review? Seems to work for me -- can confirm that shift+enter in the search box in Chrome opens a new window" [core] - 10https://gerrit.wikimedia.org/r/48252 (owner: 10Bartosz Dziewoński) [20:47:33] (03CR) 10Bartosz Dziewoński: "The most troublesome thing about this patch is the dependency at https://gerrit.wikimedia.org/r/#/c/82814/1 , which worked for me when I t" [core] - 10https://gerrit.wikimedia.org/r/48252 (owner: 10Bartosz Dziewoński) [20:48:53] (03PS1) 10Jeroen De Dauw: Update links in readme [extensions/ParserHooks] - 10https://gerrit.wikimedia.org/r/86026 [20:49:45] (03CR) 10Jeroen De Dauw: [C: 032] Update links in readme [extensions/ParserHooks] - 10https://gerrit.wikimedia.org/r/86026 (owner: 10Jeroen De Dauw) [20:52:41] (03PS1) 10Hoo man: Clear legacy restriction field on page restriction update [core] - 10https://gerrit.wikimedia.org/r/86028 [20:53:15] (03CR) 10Hoo man: "Untested hotfix for itwiki problems." [core] - 10https://gerrit.wikimedia.org/r/86028 (owner: 10Hoo man) [21:02:46] mwalker, hi, you can tell him/her to contact me [21:03:07] awesome; that's pretty much what I did [21:03:34] and point them to the tool-labs project list [21:10:57] (03PS5) 10EBernhardson (WMF): Generate recentchanges rows from Flow [extensions/Flow] - 10https://gerrit.wikimedia.org/r/81017 [21:17:06] (03CR) 10Brion VIBBER: "I think it's there just to be consistent in which events get propagated out through keydown, keypress, and keyup." [core] - 10https://gerrit.wikimedia.org/r/82814 (owner: 10Bartosz Dziewoński) [21:24:20] (03PS5) 10Dan-nl: addressing brian wolff’s comments [extensions/GWToolset] - 10https://gerrit.wikimedia.org/r/83657 [21:29:32] (03CR) 10CSteipp: [C: 031] "thanks" [extensions/BetaFeatures] - 10https://gerrit.wikimedia.org/r/86023 (owner: 10MarkTraceur) [21:32:30] (03PS1) 10Bsitu: Bundle message should show diff of all included revisions [extensions/Echo] - 10https://gerrit.wikimedia.org/r/86031 [21:33:31] (03CR) 10Bsitu: [C: 04-2] Bundle message should show diff of all included revisions [extensions/Echo] - 10https://gerrit.wikimedia.org/r/86031 (owner: 10Bsitu) [21:58:07] (03CR) 10Bsitu: [C: 032] Moving dependencies from ext.echo.base to more specific modules [extensions/Echo] - 10https://gerrit.wikimedia.org/r/86006 (owner: 10Kaldari) [21:58:36] (03Merged) 10jenkins-bot: Moving dependencies from ext.echo.base to more specific modules [extensions/Echo] - 10https://gerrit.wikimedia.org/r/86006 (owner: 10Kaldari) [22:01:09] bd808: csteipp only +1'd, you're up [22:07:47] marktraceur: looking now [22:09:49] (03CR) 10BryanDavis: [C: 032] "good changes" [extensions/BetaFeatures] - 10https://gerrit.wikimedia.org/r/86023 (owner: 10MarkTraceur) [22:09:59] (03Merged) 10jenkins-bot: Fixes from security review [extensions/BetaFeatures] - 10https://gerrit.wikimedia.org/r/86023 (owner: 10MarkTraceur) [22:14:52] (03PS1) 10Amire80: Replace Pakaran with Punjabi [core] - 10https://gerrit.wikimedia.org/r/86040 [22:15:41] Thankee [22:17:54] (03CR) 10Nikerabbit: [C: 032] Replace Pakaran with Punjabi [core] - 10https://gerrit.wikimedia.org/r/86040 (owner: 10Amire80) [22:24:26] (03Merged) 10jenkins-bot: Replace Pakaran with Punjabi [core] - 10https://gerrit.wikimedia.org/r/86040 (owner: 10Amire80) [22:29:29] (03PS1) 10Nikerabbit: Plural support for stat tiles [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86043 [22:31:17] (03CR) 10Amire80: [C: 032] Plural support for stat tiles [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86043 (owner: 10Nikerabbit) [22:31:43] (03CR) 10Ryan Lane: [C: 032] Use $wgRedactedFunctionArguments for sensitive data [extensions/LdapAuthentication] - 10https://gerrit.wikimedia.org/r/85395 (owner: 10Reedy) [22:31:47] (03CR) 10Ryan Lane: [V: 032] Use $wgRedactedFunctionArguments for sensitive data [extensions/LdapAuthentication] - 10https://gerrit.wikimedia.org/r/85395 (owner: 10Reedy) [22:31:53] (03Merged) 10jenkins-bot: Plural support for stat tiles [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86043 (owner: 10Nikerabbit) [22:32:08] (03CR) 10Ryan Lane: [C: 032 V: 032] Use $wgRedactedFunctionArguments for sensitive data [extensions/OATHAuth] - 10https://gerrit.wikimedia.org/r/85396 (owner: 10Reedy) [22:39:08] (03PS4) 10Nikerabbit: Move statsbar construction to frontend [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/85366 (owner: 10Santhosh) [22:43:43] (03PS1) 10CSteipp: Vary on forceHTTPS cookie [core] - 10https://gerrit.wikimedia.org/r/86044 [22:52:10] do we have a bug for killing that ungodly piece of crap that tidy is? [22:52:41] Yes. [22:52:46] Sort of. [22:52:49] Search for "tidy sucks"? [22:53:12] I guess it got renamed. [22:53:29] "tidy" yields 79 bugs [22:53:33] Dereckson... [22:53:37] https://bugzilla.wikimedia.org/show_bug.cgi?id=2542 ? [22:53:50] Yes. [22:53:50] i'm renaming it back. [22:53:51] https://bugzilla.wikimedia.org/show_bug.cgi?id=2542 [22:54:07] Can I bug-conflict with you? [22:54:08] tidy fucking sucks and there's nothing we can do about it [22:54:11] other than killing it [22:54:14] be my guest [22:54:35] Actually, you can do it. [22:54:38] there's gotta be a non-shitty php library that will just close unclosed tags and not fuck around with their contents [22:54:43] You want to re-purpose that bug? [22:54:55] What's wrong with Tidy? [22:54:58] We need an RFC! [22:55:11] to be brief, it fucks shit up. :D [22:55:26] !bug tidy [22:55:27] https://bugzilla.wikimedia.org/buglist.cgi?quicksearch=tidy [22:55:33] here's a list of things that are wrong ^ [22:55:42] https://bugzilla.wikimedia.org/show_bug.cgi?id=54616 from today [22:56:15] That bug is... [22:56:18] Who's using
? [22:56:24] my favourite offender is https://bugzilla.wikimedia.org/show_bug.cgi?id=260 [22:56:29] wikipedians, probably [22:56:45] ==
...
== seems like a pretty retarded construct to me. [22:56:59] https://bugzilla.wikimedia.org/show_bug.cgi?id=38800 , actually [22:56:59] Though Tidy obviously shouldn't mangle it. [22:57:07] that one is not retarded [22:57:39] anyway. [22:57:43] http://www.bioinformatics.org/phplabware/internal_utilities/htmLawed/index.php looks sane. [22:58:08] and not hundreds of files like http://htmlpurifier.org/ [22:58:20] I'd recommend filing a new bug. [22:58:20] let's file a bug [22:58:31] "Replace HTMLTidy" or some such. [22:59:29] (03PS5) 10Nikerabbit: Move statsbar construction to frontend [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/85366 (owner: 10Santhosh) [22:59:43] MatmaRex: I'm not sure the answer is to replace HTMLTidy, though... [22:59:54] I'm not sure what problems are trying to be addressed. [22:59:56] Unclosed tags? [23:00:01] (03PS1) 10Krinkle: Update VisualEditor and TemplateData submodule [core] (wmf/1.22wmf18) - 10https://gerrit.wikimedia.org/r/86048 [23:00:06] easiest way to kill htmltidy? replace the parser with parsoid, which emits good html5 ;)) [23:00:32] Let's just wrap everything in . [23:00:34] Elsie: yes. creating a page with "
" will mess up the layout in interesting ways [23:00:49] Will it? [23:00:53] I thought that was Monobook-specific. [23:01:07] https://bugzilla.wikimedia.org/show_bug.cgi?id=54617 [23:01:08] it will [23:01:18] results may vary across skins [23:01:23] anyway, tidy sucks, that's a fact [23:01:30] https://wikimediafoundation.org/wiki/Wikimedia:Sandbox [23:01:43] you can try
, too. [23:01:54] Hmmmmm. [23:01:57]
gets closed. [23:02:02] What's that about... [23:02:27] tidy is probably enabled there [23:02:33] i think it's enabled on all wmf wikis [23:02:36] https://noc.wikimedia.org/conf/InitialiseSettings.php.txt [23:02:39] !wg UseTidy [23:02:40] https://www.mediawiki.org/wiki/Manual:$wgUseTidy [23:02:44] It's not enabled there. [23:02:48] it sucks so bad we have a variable to disable it [23:02:58] (03PS1) 10Krinkle: Update VisualEditor submodule [core] (wmf/1.22wmf17) - 10https://gerrit.wikimedia.org/r/86049 [23:03:21] (03CR) 10Krinkle: [C: 032] Update VisualEditor and TemplateData submodule [core] (wmf/1.22wmf18) - 10https://gerrit.wikimedia.org/r/86048 (owner: 10Krinkle) [23:03:37] MatmaRex: I wonder if core also will close tags. [23:03:50] Something is adding
to that page's output and Tidy is allegedly disabled. [23:03:56] (03CR) 10Krinkle: [C: 032] Update VisualEditor submodule [core] (wmf/1.22wmf17) - 10https://gerrit.wikimedia.org/r/86049 (owner: 10Krinkle) [23:06:07] (03PS6) 10Nikerabbit: Move statsbar construction to frontend [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/85366 (owner: 10Santhosh) [23:07:25] (03Abandoned) 10Brion VIBBER: (Bug 45141) Move the timestamp in the enhanced-rc lists into the toggle object for collapsible items [core] - 10https://gerrit.wikimedia.org/r/49841 (owner: 10Isarra) [23:08:20] (03Merged) 10jenkins-bot: Update VisualEditor and TemplateData submodule [core] (wmf/1.22wmf18) - 10https://gerrit.wikimedia.org/r/86048 (owner: 10Krinkle) [23:09:57] (03Merged) 10jenkins-bot: Update VisualEditor submodule [core] (wmf/1.22wmf17) - 10https://gerrit.wikimedia.org/r/86049 (owner: 10Krinkle) [23:10:53] (03PS5) 10Brion VIBBER: Handle strings evaluating false correctly [core] - 10https://gerrit.wikimedia.org/r/66375 (owner: 10Liangent) [23:11:05] (03CR) 10Amire80: [C: 032] Move statsbar construction to frontend [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/85366 (owner: 10Santhosh) [23:12:06] (03Merged) 10jenkins-bot: Move statsbar construction to frontend [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/85366 (owner: 10Santhosh) [23:16:04] (03CR) 10jenkins-bot: [V: 04-1] Handle strings evaluating false correctly [core] - 10https://gerrit.wikimedia.org/r/66375 (owner: 10Liangent) [23:18:52] (03CR) 10Aaron Schulz: [C: 032] Vary on forceHTTPS cookie [core] - 10https://gerrit.wikimedia.org/r/86044 (owner: 10CSteipp) [23:21:56] (03CR) 10Brion VIBBER: "So... the patch still merges but the parser test came up with the wrong value. Wtf?" [core] - 10https://gerrit.wikimedia.org/r/66375 (owner: 10Liangent) [23:22:06] (03Merged) 10jenkins-bot: Vary on forceHTTPS cookie [core] - 10https://gerrit.wikimedia.org/r/86044 (owner: 10CSteipp) [23:24:53] (03CR) 10Aaron Schulz: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/66366 (owner: 10Parent5446) [23:34:30] (03PS3) 10Nikerabbit: WIP: Added integration tests boiler plate for TwnMainPage. [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/85688 (owner: 10KartikMistry) [23:35:09] (03PS4) 10Nikerabbit: WIP: Added integration tests boiler plate for TwnMainPage. [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/85688 (owner: 10KartikMistry) [23:35:34] (03CR) 10Amire80: [C: 032] Added integration tests framework for TwnMainPage. [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/85688 (owner: 10KartikMistry) [23:35:38] (03Merged) 10jenkins-bot: Added integration tests framework for TwnMainPage. [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/85688 (owner: 10KartikMistry) [23:46:32] (03PS9) 10Brion VIBBER: Mouseover explanations for interlanguage links in native language [core] - 10https://gerrit.wikimedia.org/r/35871 (owner: 10Wizardist) [23:53:38] (03PS10) 10Brion VIBBER: Mouseover explanations for interlanguage links in native language [core] - 10https://gerrit.wikimedia.org/r/35871 (owner: 10Wizardist) [23:56:10] (03CR) 10Brion VIBBER: [C: 031] "Rebased and fixed patch to work." [core] - 10https://gerrit.wikimedia.org/r/35871 (owner: 10Wizardist)