[00:00:43] Krinkle: Any idea what's up with https://integration.wikimedia.org/ci/job/jsduck/395/consoleFull ? [00:00:50] Krinkle: (https://gerrit.wikimedia.org/r/#/c/196481/) [00:01:27] James_F: Needs submodules [00:01:32] legoktm broke that earlier [00:01:34] :D [00:01:35] Krinkle: Oh. [00:01:41] legoktm: Can you please not? :-) [00:01:53] After I reassured him it wouldn't break anything [00:01:54] Being unable to merge is a bit sucky. [00:02:16] Well then, can you please revert? [00:07:11] (03PS1) 10Jforrester: Update VE core submodule to master (792b450) [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196490 [00:08:18] James_F: in 6 + 8 + 3 minutes it'll be live. [00:09:17] (03CR) 10jenkins-bot: [V: 04-1] Update VE core submodule to master (792b450) [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196490 (owner: 10Jforrester) [00:17:43] Krinkle: That's… worryingly precise. [00:18:07] (03CR) 10Krinkle: "recheck" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196490 (owner: 10Jforrester) [00:18:58] (03CR) 10Krinkle: "recheck" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196481 (https://phabricator.wikimedia.org/T92427) (owner: 10Trevor Parscal) [00:24:41] James_F, Krinkle it's better than 42. [00:27:01] (03CR) 10Jforrester: [C: 031] Stringify and parse rendered references [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196481 (https://phabricator.wikimedia.org/T92427) (owner: 10Trevor Parscal) [00:36:21] 10VisualEditor, 10WikiEditor, 6Collaboration-Team, 10Flow, 7Design: Editing toolbar(s) for Flow (VE and/or wikitext toolbar) - https://phabricator.wikimedia.org/T78346#1115463 (10Mattflaschen) >>! In T78346#1115277, @DannyH wrote: > I added new screenshots from the prototype to T90764. > > I have a ques... [00:37:45] 10OOjs, 10Continuous-Integration, 6Labs, 10Wikimedia-Labs-Infrastructure, 6operations: Jenkins failing with "Error: GET https://saucelabs.com: Couldn't resolve host name." - https://phabricator.wikimedia.org/T92351#1115471 (10scfc) I don't think so because that was merged earlier. But on March 6th https... [00:46:26] editing complex tables with VE like it's nothing. we've come a long way ... [00:54:00] (03PS1) 10Alex Monk: Don't try to access x-parsoid-performance on restbase cache hits [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196503 [00:54:49] (03CR) 10Jforrester: [C: 032] Don't try to access x-parsoid-performance on restbase cache hits [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196503 (owner: 10Alex Monk) [00:55:05] Eloquence: :-) [00:55:10] Eloquence: More to come! [00:55:15] (03CR) 10GWicke: [C: 032] Don't try to access x-parsoid-performance on restbase cache hits [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196503 (owner: 10Alex Monk) [00:58:03] (03Merged) 10jenkins-bot: Don't try to access x-parsoid-performance on restbase cache hits [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196503 (owner: 10Alex Monk) [01:03:30] (03PS1) 10Legoktm: jenkins job validation, do not submit [extensions/Citoid] - 10https://gerrit.wikimedia.org/r/196506 [01:08:48] (03Abandoned) 10Legoktm: jenkins job validation, do not submit [extensions/Citoid] - 10https://gerrit.wikimedia.org/r/196506 (owner: 10Legoktm) [01:26:34] edsanders|away: Has italic-cyrl-ka.svg gone walkies or did it never exist? [01:27:05] Ooooh. [01:27:08] It was a fun toy. [01:29:16] (03CR) 10Jforrester: "recheck" [extensions/Citoid] - 10https://gerrit.wikimedia.org/r/195802 (https://phabricator.wikimedia.org/T62768) (owner: 10Mooeypoo) [01:30:26] Krinkle: Can I grab a quick +2 on https://gerrit.wikimedia.org/r/#/c/196490/ to unbreak the VE link icon for Beta Cluster? [01:33:05] James_F: Checking [01:33:21] 10VisualEditor, 10VisualEditor-Performance, 3VisualEditor 2014/15 Q3 blockers: Retrieve page HTML directly from RESTbase - https://phabricator.wikimedia.org/T90374#1115575 (10GWicke) [01:33:23] 10VisualEditor, 10MediaWiki-General-or-Unknown, 10RESTBase, 6Services, and 4 others: Parsoid performance: Use RESTBase from the MediaWiki Virtual Rest Service - https://phabricator.wikimedia.org/T89066#1115574 (10GWicke) [01:33:25] 10VisualEditor, 10VisualEditor-Performance, 10MediaWiki-General-or-Unknown, 10MediaWiki-Interface, and 8 others: Use Parsoid HTML for read-only views too - https://phabricator.wikimedia.org/T55784#1115572 (10GWicke) [01:33:26] 10VisualEditor, 10VisualEditor-Performance, 10Parsoid, 10RESTBase, 7Epic: Store & load data-mw separately - https://phabricator.wikimedia.org/T78676#1115576 (10GWicke) [01:33:39] Krinkle: Thanks. [01:34:56] (03PS1) 10Jforrester: [WIP] Move all our icons to OOjs UI's icon packs [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/196511 [01:35:14] 10VisualEditor, 10MediaWiki-General-or-Unknown, 10RESTBase, 6Services, and 4 others: Parsoid performance: Use RESTBase from the MediaWiki Virtual Rest Service - https://phabricator.wikimedia.org/T89066#1026793 (10GWicke) VE is now using RESTBase on all phase0 wikis including https://www.mediawiki.org. Furt... [01:35:41] (03PS1) 10Jforrester: [WIP] Bring in remaining VisualEditor icons for Apex and MediaWiki themes [oojs/ui] - 10https://gerrit.wikimedia.org/r/196512 [01:36:25] (03PS1) 10Alex Monk: Don't try to access x-parsoid-performance on restbase cache hits [extensions/VisualEditor] (wmf/1.25wmf21) - 10https://gerrit.wikimedia.org/r/196514 [01:36:25] 10VisualEditor, 10MediaWiki-General-or-Unknown, 10RESTBase, 6Services, and 4 others: Parsoid performance: Use RESTBase from the MediaWiki Virtual Rest Service: phase0 done, others to follow - https://phabricator.wikimedia.org/T89066#1115581 (10GWicke) [01:36:27] (03CR) 10jenkins-bot: [V: 04-1] [WIP] Bring in remaining VisualEditor icons for Apex and MediaWiki themes [oojs/ui] - 10https://gerrit.wikimedia.org/r/196512 (owner: 10Jforrester) [01:37:30] (03CR) 10Krinkle: [C: 032] Update VE core submodule to master (792b450) [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196490 (owner: 10Jforrester) [01:39:07] James_F http://imgur.com/NxAbGMX [01:39:19] Krenair: Fancy. :-) [01:39:38] that's using wikieditor's i18n messages for the different pages [01:39:41] Krenair: How do we i18n the names of the groups currently? Is it preserved in the JSON lookup? [01:39:42] I guess those should be in MW core too :/ [01:39:43] (03Merged) 10jenkins-bot: Update VE core submodule to master (792b450) [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196490 (owner: 10Jforrester) [01:39:43] Aha, right. [01:39:52] Yeah. That'll be a fun transport. [01:45:23] James_F: https://doc.wikimedia.org/_krinkle/coverage/oojs/master/PhantomJS%201.9.8%20(Linux)/ [01:45:40] Krinkle: Nice. [01:46:13] Krinkle: Look to be some simple things to add test coverage for, too. [01:46:42] Unfortunately, there is no path to integration.wikimedia.org right now [01:46:46] oo.setProp called without an Object [01:46:49] where the coverage directory is [01:47:08] Different servers? [01:47:15] No, just different docroots [01:47:26] It's possible, but just requires more refactoring [01:47:29] Sure. [01:48:16] (03CR) 10GWicke: [C: 032] Add bash script for updating submodule [services/citoid/deploy] - 10https://gerrit.wikimedia.org/r/195911 (owner: 10Mvolz) [01:48:52] (03CR) 10GWicke: [C: 032] Remove the zotero translation server and translators [services/citoid/deploy] - 10https://gerrit.wikimedia.org/r/194548 (https://phabricator.wikimedia.org/T89872) (owner: 10Alexandros Kosiaris) [01:48:59] James_F: It doesn't merge coverage from Chrome/Firefox/Node.js though [01:49:14] So it'll never reach 100% of all branches [01:49:31] 10OOjs-UI: oojs ui `npm build` fails in Rasterizing SVG to PNG with "Warning: Bad argument - https://phabricator.wikimedia.org/T92479#1115587 (10Spage) >>! In T92479#1113001, @matmarex wrote: > If it doesn't help, also wipe the NPM cache with `npm cache clean`, then remove `node_modules` and do `npm install`.... [01:49:44] Short of mocking a lot, which is possible, but not worth it right now [01:49:48] Krinkle: Is that even something that'd be wise? [01:49:50] * James_F nods. [01:50:03] It is covered in tests. So it's not like we're not covering it. [01:50:05] Just not in the same report [01:50:45] 10OOjs-UI, 7Upstream: oojs ui `npm build` fails in Rasterizing SVG to PNG with "Warning: Bad argument - https://phabricator.wikimedia.org/T92479#1115588 (10Spage) [01:52:32] I mean, is it wise trying to show that code is test-covered in at least one of three environments? [01:52:56] If oo.clone is only tested in Firefox (say) that's a risk, not a bonus. [01:52:57] Etc. [01:53:12] Yeah [01:53:43] I guess we could decorate functions and even lines with comments explaining which environments they're for… [01:53:45] But ick. [01:53:56] but I mean, we are running tests in all three (Chrome, Firefox, Node.js) and some of the branches indicated as not coverage are in fact covered in those other runs and would result in build failure if that code regressed. But the publsihed report is only of the coverage of one of those runs. [01:54:06] Yeah. [01:54:44] (03PS2) 10Jforrester: [WIP] Bring in remaining VisualEditor icons for Apex and MediaWiki themes [oojs/ui] - 10https://gerrit.wikimedia.org/r/196512 [01:54:46] OK, anyway. Home time. [01:54:49] * James_F waves. [01:55:00] Krinkle: Please make the 09:00 tomorrow. [01:55:17] They should already be indicating browser-specific branches. jQuery more or less standardised on a format of // Support: \n // Explanation. [01:55:29] I did the same in MediaWiki core [01:55:30] (03CR) 10jenkins-bot: [V: 04-1] [WIP] Bring in remaining VisualEditor icons for Apex and MediaWiki themes [oojs/ui] - 10https://gerrit.wikimedia.org/r/196512 (owner: 10Jforrester) [01:55:40] James_F: Health check? [01:55:44] Yup. [01:56:02] (03CR) 10Jforrester: "Whee, now with added npm." [extensions/Citoid] - 10https://gerrit.wikimedia.org/r/195802 (https://phabricator.wikimedia.org/T62768) (owner: 10Mooeypoo) [01:56:56] 10VisualEditor, 10VisualEditor-Performance, 3VisualEditor 2014/15 Q3 blockers, 5WMF-deploy-2015-03-18_(1.25wmf22): Don't disable native object editing in IE and WebKit browsers - https://phabricator.wikimedia.org/T89928#1115596 (10Jdforrester-WMF) [01:58:52] 10OOjs, 10Continuous-Integration: Publish QUnit coverage on integration.wikimedia.org - https://phabricator.wikimedia.org/T87490#1115600 (10Krinkle) In oojs, oojs-ui and VisualEditor, test coverage can be generated locally by running `npm install && npm test` in their directory and opening the coverage directo... [02:10:34] 10OOjs, 10Continuous-Integration, 6Labs, 10Wikimedia-Labs-Infrastructure, 6operations: Jenkins failing with "Error: GET https://saucelabs.com: Couldn't resolve host name." - https://phabricator.wikimedia.org/T92351#1115608 (10coren) The only net effect the change can make is that //iff// the fqdn has exa... [02:25:19] James_F, I still think we need to fix the scrollbar disappearance issue with modal dialogs ... very visually apparent in a lot of contexts [02:25:26] triggers reflows etc. [02:25:44] Eloquence: Yeah, it's on Bartosz's/Ed's list to fix/ [02:25:49] nice [02:30:47] 10VisualEditor, 10VisualEditor-Performance, 10MediaWiki-General-or-Unknown, 10MediaWiki-Interface, and 8 others: Use Parsoid HTML for read-only views too - https://phabricator.wikimedia.org/T55784#1115631 (10GWicke) [02:41:14] 10VisualEditor, 10VisualEditor-Performance, 10Flow, 10Parsoid, and 5 others: Move data-parsoid and possibly data-mw out of the DOM, add uids - https://phabricator.wikimedia.org/T54936#1115641 (10GWicke) [02:41:16] 10VisualEditor, 10VisualEditor-MediaWiki, 10Flow, 10MediaWiki-General-or-Unknown, and 5 others: Use PHP Virtual REST Service to access Parsoid - https://phabricator.wikimedia.org/T1218#1115642 (10GWicke) [03:01:51] 10VisualEditor, 10VisualEditor-Performance, 10Parsoid, 10RESTBase, 7Epic: Store & load data-mw separately - https://phabricator.wikimedia.org/T78676#1115648 (10GWicke) [03:19:25] 10VisualEditor: Schema:Edit seems to incorrectly set users as anonymous. - https://phabricator.wikimedia.org/T92596#1115650 (10Milimetric) 3NEW a:3Krenair [04:15:08] (03CR) 10GWicke: [V: 032] Add bash script for updating submodule [services/citoid/deploy] - 10https://gerrit.wikimedia.org/r/195911 (owner: 10Mvolz) [04:15:29] (03CR) 10GWicke: [C: 032 V: 032] Update citoid submodule to master (2cd246b) [services/citoid/deploy] - 10https://gerrit.wikimedia.org/r/195912 (owner: 10Mvolz) [04:15:48] (03CR) 10GWicke: [V: 032] Remove the zotero translation server and translators [services/citoid/deploy] - 10https://gerrit.wikimedia.org/r/194548 (https://phabricator.wikimedia.org/T89872) (owner: 10Alexandros Kosiaris) [08:37:24] 10VisualEditor, 10VisualEditor-EditingTools, 10VisualEditor-Tables: Table with some template-generated rows appears with additional jumbled empty table cells - https://phabricator.wikimedia.org/T70306#1115952 (10Tbayer) Another example with [[en:Template:Singlechart]]: https://en.wikipedia.org/w/index.php?ti... [09:29:17] (03PS2) 10Gerrit Patch Uploader: Highlight target references [extensions/Cite] - 10https://gerrit.wikimedia.org/r/196172 (https://phabricator.wikimedia.org/T10303) [09:29:20] (03CR) 10Gerrit Patch Uploader: "This commit was uploaded using the Gerrit Patch Uploader [1]." [extensions/Cite] - 10https://gerrit.wikimedia.org/r/196172 (https://phabricator.wikimedia.org/T10303) (owner: 10Gerrit Patch Uploader) [09:33:12] (03CR) 10Bartosz Dziewoński: "Works for me. Can you file a bug, providing information about your browser?" [oojs/ui] - 10https://gerrit.wikimedia.org/r/188289 (owner: 10Jforrester) [09:44:44] (03CR) 10Schnark: "According to https://phabricator.wikimedia.org/T86782#1000248 there are plans to change the class="references" to class="mw-references". H" [extensions/Cite] - 10https://gerrit.wikimedia.org/r/196172 (https://phabricator.wikimedia.org/T10303) (owner: 10Gerrit Patch Uploader) [09:55:52] (03CR) 10Schnark: Highlight target references (031 comment) [extensions/Cite] - 10https://gerrit.wikimedia.org/r/196172 (https://phabricator.wikimedia.org/T10303) (owner: 10Gerrit Patch Uploader) [10:09:02] hello [10:10:34] i did some changes and after creating new branch it doesn't show any changes when i do git diff [10:15:31] hi [10:17:30] ff is recently full of problems [10:19:53] ff? [10:20:10] fast forward [10:35:54] (03CR) 10Thiemo Mättig (WMDE): [C: 031] "Restoring my +1." [extensions/Cite] - 10https://gerrit.wikimedia.org/r/196172 (https://phabricator.wikimedia.org/T10303) (owner: 10Gerrit Patch Uploader) [10:43:29] 10VisualEditor: Make the visual template editor drag-able & re-sizable to allow editing easy - https://phabricator.wikimedia.org/T92396#1116243 (1001tonythomas) [10:43:30] 10OOjs-UI: OOjs UI: Dialogs should be repositionable/draggable - https://phabricator.wikimedia.org/T51969#1116244 (1001tonythomas) [10:47:04] 10VisualEditor: Fix styling inconsistencies between context item and inspector - https://phabricator.wikimedia.org/T90896#1116253 (10Aklapper) I agree. These needs steps (or links to documentation) how to see the "inspector" and what "context item" refers to here [[ https://www.mediawiki.org/wiki/Phabricator/Pro... [11:02:09] Juandev: Firefox problems? Could you elaborate? [11:02:15] siddhism: Are you still fighting git? [11:02:28] yes [11:02:55] Same problem as an hour ago? [11:02:59] RoanKattouw: well, edit toolbar doesn not travel with you scrolling down [11:03:03] yes [11:03:19] Juandev: On which wiki? [11:03:24] git status doesn't list my changes too [11:03:35] That sounds like a bug that we caused yesterday and should have been fixed yesterday after I went to sleep [11:03:39] RoanKattouw: after saving FF still think you are in edit mode (everything is gray) and when you leave it warns you that there is unsaved window [11:03:44] RoanKattouw: cs.wv [11:03:48] siddhism: Could you pastebin the output of git status and of git log -5 --oneline ? [11:03:55] Juandev: Checking [11:04:06] RoanKattouw: ah, nice:-) it works properly in Google Chrome [11:04:21] Thanks Juandev , I was literally about to ask that question :) [11:05:12] and it kind of mismach the design of the page, but I dont know in which state, weather in edit mode or after saving [11:05:25] eg. category bar is placed into the middle of the page [11:05:29] Hmm, WFM https://cs.wikiversity.org/w/index.php?title=U%C5%BEivatel:Roan_Kattouw_%28WMF%29/Sandbox&action=history [11:05:35] Juandev: Can you reproduce these problems when logged out? [11:06:56] RoanKattouw: well, not. because when I am logged out I cannot use VE at all [11:07:30] Juandev: You can if you manually add ?veaction=edit to the URL [11:07:33] e.g. https://cs.wikiversity.org/wiki/U%C5%BEivatel:Roan_Kattouw_%28WMF%29/Sandbox?veaction=edit [11:07:39] oh key [11:09:24] it works properly [11:09:24] http://pastebin.com/HcXAf1av [11:09:42] Juandev: OK so I suspect this is a problem with a Gadget or user JS of some kind [11:09:58] Juandev: Could you log back in and reproduce it, but this time keep the console open and see what errors you get? [11:10:05] that would be easy, there are about 3 gadgets available [11:11:06] siddhism: Thanks. Could you also pastebin the output of git status and git branch ? [11:11:52] muh, now it works properly [11:12:44] What you described does fit the description of a bug we had [11:12:52] So you may have been running cached code from before we fixed it or something [11:13:11] Especially if you encountered this yesterday; we did a deploy around midnight CET last night [11:15:32] RoanKattouw:http://pastebin.com/2sLyuUt7 its the git status, i am siddhism [11:17:05] http://pastebin.com/8chkPjzv its the git branch output [11:18:28] RoanKattouw: great [11:18:33] thx [11:21:02] rits: OK so git diff --staged will show you the diff to MessagesEn.php (I think that was the file you meant to modify?), as well as a lot of maintenance/vendor stuff you probably didn't mean to do [11:21:25] rits: git diff --staged tells you what's going to happen when you run git commit, basically [11:21:58] If you're extra paranoid you can also run git commit -v and it'll tell you what it's going to commit in the editor where you write the commit message, and you can use 'git show' after committing to check what was just committed [11:22:34] In your situation, I'm guessing you probably didn't intend to add all the maintenance/vendor/monolog stuff; if you need to remove that from what will be committed, then run git reset maintenance/vendor/monolog [11:22:46] (If you did intend to commit that, then you're on the right track) [11:22:56] Then once git diff --staged matches what you want to do, run git commit and you should be all set [11:23:23] RoanKattouw: no i actually made the changes to a css file of visualeditor, MessagesEn.php change was a different, i didnt want to commit it [11:23:43] Oh OK I see [11:23:54] You appear to have been running git status etc in the MediaWiki core directory [11:23:59] VisualEditor is a separate repositoiry [11:24:01] *repository [11:24:11] all other changes are been tracked but that css files doesnt shows up, [11:24:15] So you need to cd extensions/VisualEditor or wherever it is, then run git diff / git status [11:24:59] oh ok i will try it, just a minute [11:25:13] * RoanKattouw is on babysitting duty + lunch, back in 20-30 mins [11:46:52] (03CR) 10Catrope: [C: 032] Stringify and parse rendered references [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196481 (https://phabricator.wikimedia.org/T92427) (owner: 10Trevor Parscal) [11:49:03] (03Merged) 10jenkins-bot: Stringify and parse rendered references [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196481 (https://phabricator.wikimedia.org/T92427) (owner: 10Trevor Parscal) [11:49:14] 10VisualEditor: "Cannot read property 'getSurface' of null" error when clicking in 'Insert paragraph' in Basic citation inspector - https://phabricator.wikimedia.org/T92572#1116354 (10Catrope) This looks like it's probably related to T92427 and T92583 and maybe fixed by https://gerrit.wikimedia.org/r/#/c/19648... [11:49:15] RoanKattouw:i did git diff and git status from the visualeditor repository its doesnt show the changes then also [11:49:30] rits: Does git diff --staged show them? [11:50:39] no it just displays the MessagesEn.php change [11:51:33] css file is not tracked anywhere, [11:51:56] rits: You need to cd into the VisualEditor directory [11:52:02] Then run git diff and friends [11:52:23] If that doesn't work try cd'ing to the directory where that CSS file is [11:52:45] yes i did i am in the VisualEditor directory, there also git diff --staged shows the changes to messagesEn.php only [11:53:17] wtf that shouldn't happen [11:53:28] You shouldn't be able to see MessagesEn.php changes from the VisualEditor directory :S [11:53:44] oh ok i will do it hopefully moving to the css file directory could solve the issue [11:55:46] RoanKattouw: see here is the output of git diff --staged http://pastebin.com/ptVP3eNX, i am also surprised how could it shows the MessagesEn.php changes [11:56:18] 10VisualEditor: "Cannot read property 'getSurface' of null" error when clicking in 'Insert paragraph' in Basic citation inspector - https://phabricator.wikimedia.org/T92572#1116375 (10Catrope) Yup, this is "fixed" now in that clicking the slug doesn't cause an error or do anything, but we probably shouldn't be... [11:58:18] 10VisualEditor: Nonfunctional slugs appear in reference context item - https://phabricator.wikimedia.org/T92572#1116379 (10Catrope) [11:59:28] 10VisualEditor, 10VisualEditor-ContentEditable, 7Technical-Debt, 3VisualEditor 2014/15 Q3 blockers: Audit and fix events bound to CE nodes but not unbound when destroy() is called - https://phabricator.wikimedia.org/T92583#1116382 (10Catrope) Hmm, maybe we could use the DM rendering rather than the CE rend... [12:00:33] RoanKattouw: moved to the css file directory the same output no change, and more surprisingly the MessagesEn.php change is displayed there also [12:01:17] (03PS1) 10Esanders: Compensate for loss of margin when opening modals [oojs/ui] - 10https://gerrit.wikimedia.org/r/196571 (https://phabricator.wikimedia.org/T89612) [12:01:24] 10OOjs-UI, 10VisualEditor, 10VisualEditor-MediaWiki: The OOUI hack to prevent scrolling while dialog is open causes pages to scroll when dialog opens - https://phabricator.wikimedia.org/T89612#1116393 (10Esanders) [12:03:09] rits: What is the CSS file directory exactly? [12:05:44] RoanKattouw: this is the path /var/www/html/core/extensions/VisualEditor/lib/ve/src/ui/styles/widgets [12:06:06] Weird, that should totally be working [12:06:07] Hmm [12:06:23] Does /var/www/html/core/extensions/VisualEditor/lib/ve/.git exist? Does /var/www/html/core/extensions/VisualEditor/.git exist? [12:09:48] no it is not there i just saw [12:10:55] oh i just remember that i downloaded the tarball as for the mediawiki alpha [12:11:01] aaaaah [12:11:02] Aha [12:11:10] OK, so git was never initialized for VE [12:11:16] I can help you solve that [12:11:25] rits: Did you change anything else except for that CSS file? [12:11:34] no only the css file [12:11:39] OK [12:11:55] Copy the CSS file to somewhere safe (i.e. somewhere outside of /var/www , maybe your home directory or something) [12:12:13] ok [12:13:42] Then once you've done that do: cd /var/www/html/core/extensions ; rm -rf VisualEditor ; git clone ssh://yourgerritusername@gerrit.wikimedia.org:29418/mediawiki/extensions/VisualEditor.git ; cd VisualEditor ; git submodule update --init [12:15:43] ok thank you sooooo much, i was so stuck here, thank you so much :) [12:44:51] (03PS13) 10Catrope: Load RL modules in one load.php request, rather than in two stages [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/193026 (https://phabricator.wikimedia.org/T53569) [13:02:59] RoanKattouw_away, insert/replace operations use annotation store indices, whereas annotate operations use the annotation objects themselves [13:30:53] (03CR) 10Esanders: [C: 04-1] "Error doesn't fit: http://i.imgur.com/q7KqbZc.png" [extensions/Citoid] - 10https://gerrit.wikimedia.org/r/195802 (https://phabricator.wikimedia.org/T62768) (owner: 10Mooeypoo) [13:35:16] edsanders: Hmm, right [13:35:28] edsanders: Seems worth fixing [13:37:49] (03PS3) 10Esanders: Use OOjs UI progress bar while VE is loading [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/175488 [13:37:55] 10VisualEditor: Cite context should update after changes made to the cite - https://phabricator.wikimedia.org/T92624#1116551 (10Florian) 3NEW [13:43:27] 10VisualEditor: Citation context should update after changes made to the citation - https://phabricator.wikimedia.org/T92624#1116563 (10Florian) [14:33:50] (03PS14) 10Catrope: Load RL modules in one load.php request, rather than in two stages [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/193026 (https://phabricator.wikimedia.org/T53569) [14:46:28] Morning, lovely people. [14:47:06] (03CR) 10Jforrester: "> Error doesn't fit: http://i.imgur.com/q7KqbZc.png" [extensions/Citoid] - 10https://gerrit.wikimedia.org/r/195802 (https://phabricator.wikimedia.org/T62768) (owner: 10Mooeypoo) [14:47:55] (03CR) 10Esanders: [C: 04-1] "still depends on styles - may be abandoned in favour of other progress bar commits." [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/175488 (owner: 10Esanders) [14:48:46] James_F: What do you want me to review most right now? [14:48:50] (03CR) 10Jforrester: "Add oojs-ui.styles to the initinit dependency and it should work?" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/175488 (owner: 10Esanders) [14:48:52] "Load RL modules in one load.php request," ? [14:48:59] Krinkle: Yes. [14:49:02] Krinkle: Yes [14:49:11] Krinkle: Unless RoanKattouw wants your assistance in part II of that. [14:49:15] Ori reviewed it last night and I responded to that [14:49:19] No I think I'll be good for now [14:49:23] Kk. [14:49:34] In response to Ori's review, the plugin code is now much smaller [14:49:34] RoanKattouw: On what are you working? Need anything? [14:49:38] Functional programming FTW [14:49:38] Wanna make sure I"m not reviewing while you refactor it underneath me [14:49:43] No go for it [14:49:47] I'm working on part 2 [14:50:11] However I resisted the temptation to replace function ( callback ) { callback(); } with Function.prototype.call.bind( Function.prototype.call ); :P [14:50:37] edsanders: How'd you feel about a quick trip to Linz with me in mid-April? [14:50:50] RoanKattouw, do we average stats for how loading the various stages of loading take [14:51:10] edsanders: I don't think so, but I should probably talk to ori about doing that, it sounds like a good idea [14:51:40] James_F, what for? [14:51:47] RoanKattouw, what stats do we have? [14:51:59] I thought that's what mw.track did [14:52:18] edsanders: Talking to the substance.io/PLOS Tahi editor people. [14:52:23] (03PS1) 10Mvolz: Add ability to set request language [extensions/Citoid] - 10https://gerrit.wikimedia.org/r/196581 (https://phabricator.wikimedia.org/T90603) [14:52:42] (03CR) 10jenkins-bot: [V: 04-1] Add ability to set request language [extensions/Citoid] - 10https://gerrit.wikimedia.org/r/196581 (https://phabricator.wikimedia.org/T90603) (owner: 10Mvolz) [14:53:46] (03PS2) 10Mvolz: Add ability to set request language [extensions/Citoid] - 10https://gerrit.wikimedia.org/r/196581 (https://phabricator.wikimedia.org/T90603) [14:53:57] (03PS3) 10Mvolz: Add ability to set request language [extensions/Citoid] - 10https://gerrit.wikimedia.org/r/196581 (https://phabricator.wikimedia.org/T90603) [14:55:44] edsanders: So our ve.track() stuff does allow us to capture how long each loading phase takes in vbench [14:55:50] But we don't currently send that data to EventLogging [14:56:01] We do measure how long the entire loading process takes I believe [14:56:23] RoanKattouw: It seems a few modules are no longer referenced: ext.ve's mwreference.core, mwtransclusion.core, and language [14:56:30] I guess they're dependencies? [14:56:35] Or otherwise redundant? [14:56:59] The *.core ones are probably dependencies of things and not currently used, although they were probably used by mobile in the past [14:57:03] language sounds like a preference module [14:57:15] $wgVisualEditorPreferenceModules or whatever it's called ends up referencing modules indirectly [14:57:20] It used to be hard coded in the same array [14:58:25] Language is a plugin module. [14:58:31] It used to be a preference module. [14:59:24] mwref.core and mwtrans.core are dependencies. So they're redundant and will be auto-added by mw.loader to the queue. [14:59:40] ext.visualEditor.language doesn't seem to be included in any obvious manner however [15:00:19] Krinkle: ve.init.mw.Target constructor. [15:00:39] Oh, did it move there? [15:00:45] I must have rebased across thatw rong [15:00:57] That's where it is now, yes. [15:01:04] (In master.) [15:01:18] (03CR) 10Krinkle: [C: 04-1] "This removes ext.visualEditor.language, ext.visualEditor.mwreference.core, and ext.visualEditor.mwtransclusion.core from the module list. " [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/193026 (https://phabricator.wikimedia.org/T53569) (owner: 10Catrope) [15:01:30] (03PS15) 10Catrope: Load RL modules in one load.php request, rather than in two stages [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/193026 (https://phabricator.wikimedia.org/T53569) [15:02:29] Krinkle: PS15 adds .language, the mwfoo.core ones are still loaded because the mwfoo ones are listed and they depend on their .core modules [15:04:19] RoanKattouw: init.init and targetLoader only depend on ve.track, which does not depend on oojs or es5-shim in anyway [15:05:21] After the third promise callback they will be available inside initinit.js [15:05:28] But not in the second where it uses forEach, I think. [15:05:35] Ahm [15:05:40] ES5 is required for VE [15:05:53] yes [15:06:03] You should not be able to cause the code that interacts with TargetLoader to run at all if you're in a non-ES5 environment [15:07:07] Right. [15:07:16] It's inside initinit,js though, of which most does run before that check [15:08:10] Hmm although now I'm looking at it, this doesn't seem to be watertight [15:08:22] What prevents the onclick handlers from being bound if !init.isAvailable? [15:08:42] RoanKattouw: Hm.. I'm curious if we can move that code to TargetLoader perhaps and make it pull out the two bits from mw.libs.ve, instead of pushing [15:08:53] (plugins list and setupSectionLinks method) [15:09:08] Oh I see, $caVeEdit is removed [15:09:28] Yeah but even then targetLoader() would not resolve [15:09:34] So it's okay as it is [15:09:35] TargetLoader itself also contains a partial plugin list [15:09:53] The problem is that VPT init init is desktop-specific while TargetLoader is meant to be more generic [15:09:58] Mobile should also be able to use TargetLoader [15:09:58] RIght [15:10:03] OK [15:10:09] Although eventually we should come up with a more generic version of init init as well [15:10:25] Or move some of that code around, or whatever [15:10:40] RoanKattouw: And maybe modules should decide desktop vs. mobile status rather than a central init method? [15:11:17] Perhaps [15:11:38] Although I generally favor a loader deliberately loading a set of modules over each module knowing when it's supposed to be loaded [15:12:41] * James_F shrugs. [15:12:48] Why not make RL do it? [15:12:54] target = mobile -> load in mobile. [15:12:55] Done. [15:12:59] Krinkle: Am I missing something or do we insert and bind handlers to VE section edit links even in non-ES5? [15:12:59] (03CR) 10Krinkle: Load RL modules in one load.php request, rather than in two stages (031 comment) [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/193026 (https://phabricator.wikimedia.org/T53569) (owner: 10Catrope) [15:14:04] James_F: The module registry is static and global across targets. Targets decide their visibility. They don't have their own registry. [15:14:15] Krinkle: I'm aware. [15:14:40] RoanKattouw: You're missing someting :) [15:14:43] Krinkle: I'm explicitly saying that targets picking which modules to load, gates on which modules RL lets them load, seems silly. [15:15:10] RoanKattouw: init.setupSkin(); runs on domready if userPrefEnabled which includes isAvailabe which includes support [15:15:54] @media screen { [15:15:54] div#bodyContent { [15:15:54] font-family: "Calibri", "Aboriginal Sans", "DejaVu Sans", "Arial Unicode MS", sans-serif !important; [15:15:56] font-size: 0.8em !important; [15:15:58] } [15:16:00] } [15:16:05] Well, that explains things. [15:16:09] https://nv.wikipedia.org/wiki/MediaWiki:Vector.css [15:16:46] RoanKattouw: Hm.. no userPrefEnabled does not include isAvailable, but shouldn't since we should set up the skin either way. How it is set up should depend on availability and support. [15:16:47] Hm [15:17:05] James_F: Of all things that's what stands out :P [15:17:16] The pony rainbow tabs [15:17:36] or scattered blue art background [15:17:46] Decoration is decoration. [15:17:56] The entire content being 20% larger than normal is what stands out. [15:18:03] HEhe, yeah [15:18:22] And VE still worked fine. [15:18:39] James_F: The editnotice is now so helpful though [15:18:54] "To save your changes click "bee lą́ ashłeeh" below" [15:19:04] >< [15:19:08] * James_F grins. [15:19:11] Yeah. [15:21:22] Krinkle: Sigh, the section edit links are hidden with CSS [15:21:38] But they're still in the DOM and they still have click handlers bound to them [15:21:56] Meanwhile the VE edit *tab* is actually removed by init init in that case [15:22:04] Consistency, we've heard of it :P [15:22:42] RoanKattouw: BTW, have you noticed that if you're browsing ?debug=true in master the VE edit tab never shows up/ [15:22:53] RoanKattouw: I assume it's some init init issue. [15:23:09] No I didn't notice that [15:23:21] http://en.wikipedia.beta.wmflabs.org/wiki/User:Jdforrester_(WMF)/Sandbox?debug=true [15:23:40] I just used your maxim of "debug is broken, don't use it". :-) [15:27:53] RoanKattouw: Clearly I shouldn't have said anything, sorry. [15:28:23] (03PS16) 10Catrope: Load RL modules in one load.php request, rather than in two stages [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/193026 (https://phabricator.wikimedia.org/T53569) [15:28:31] James_F: debug in beta is incredibly broken [15:28:44] I would be more inclined to believe you if you could reproduce that locally [15:28:56] RoanKattouw: I can reproduce in production. [15:29:00] :O [15:29:01] RoanKattouw: https://en.wikipedia.org/wiki/User:Jdforrester_(WMF)/Sandbox?debug=true [15:29:08] OK that probably means it's a real problem [15:29:12] I just used Beta so you didn't suspect it was a non-master issue. [15:29:18] However, it's debug. No-one cares. [15:29:27] Don't get distracted from value-adding activities. :-) [15:29:42] (03PS1) 10Esanders: Make loading progress bar determinate [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196583 [15:30:06] (03PS9) 10Mvolz: Add IDs for tests for Citoid extension [services/citoid] - 10https://gerrit.wikimedia.org/r/185837 (owner: 10Anishvisa) [15:30:22] edsanders: Wait, really? [15:30:23] Yeah I need to wrap up this thing first anyway [15:30:40] what? [15:31:05] edsanders: I believe that that's a 40-pointer. [15:31:28] bug? [15:31:59] https://phabricator.wikimedia.org/T90622 [15:32:01] I did start it a while ago [15:32:02] No, only 8. [15:32:24] "It's tagged as taking days. To be fair [15:32:27] RoanKattouw: Surely I can close https://phabricator.wikimedia.org/T86732 as done, right/ [15:32:31] * James_F nods. [15:32:32] I did start working on it a few hours ago." :P [15:32:39] RoanKattouw: :-) [15:32:41] Yeah probably [15:32:43] That's our edsanders. [15:32:53] 10VisualEditor, 10VisualEditor-Performance, 3VisualEditor 2014/15 Q3 blockers: Attempt to optimize / profile VisualEditor's existing load code flow to see what is actually slow? - https://phabricator.wikimedia.org/T86732#1116720 (10Jdforrester-WMF) 5Open>3Resolved [15:33:03] (03PS2) 10Esanders: Make loading progress bar determinate [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196583 (https://phabricator.wikimedia.org/T90622) [15:33:42] 10VisualEditor, 10VisualEditor-Initialisation, 10VisualEditor-MediaWiki, 10VisualEditor-Performance, and 2 others: Come up with a way of estimating load time and progress along that scale - https://phabricator.wikimedia.org/T90622#1116725 (10Jdforrester-WMF) 5stalled>3Open a:3Esanders [15:35:12] (03PS10) 10Mvolz: [WIP] Add IDs for tests for Citoid extension [services/citoid] - 10https://gerrit.wikimedia.org/r/185837 (owner: 10Anishvisa) [15:39:47] (03PS17) 10Catrope: Load RL modules in one load.php request, rather than in two stages [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/193026 (https://phabricator.wikimedia.org/T53569) [15:44:25] 10VisualEditor, 10MediaWiki-Categories: Sorting key once set for a specific category cannot be emptied and one has to delete the category. - https://phabricator.wikimedia.org/T92632#1116758 (10Swainr) 3NEW [15:52:15] (03PS1) 10Catrope: Send Parsoid API request immediately when user clicks edit [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196584 (https://phabricator.wikimedia.org/T90372) [15:52:23] (03PS18) 10Catrope: Load RL modules in one load.php request, rather than in two stages [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/193026 (https://phabricator.wikimedia.org/T53569) [15:52:30] (03PS2) 10Catrope: Send Parsoid API request immediately when user clicks edit [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/196584 (https://phabricator.wikimedia.org/T90372) [15:52:58] good morning VE folks [15:53:20] so far things are looking good with VE using RESTBase on mediawiki.org [15:54:10] we could consider adding some smaller wikipedias soon [16:00:59] RoanKattouw, edsanders, Krenair, MatmaRex, ryasmeen: Meeting. :-) [16:01:20] yes. [16:01:50] i'm trying to figure out how to join it. [16:02:05] as the hangout looks empty? [16:02:11] James_F: [16:02:22] MatmaRex: Will invite you. [16:03:03] RoanKattouw: We lost you. [16:30:01] 10VisualEditor, 10MediaWiki-Categories: Sorting key once set for a specific category cannot be emptied and one has to delete the category. - https://phabricator.wikimedia.org/T92632#1116845 (10Swainr) screencast- https://www.youtube.com/watch?v=Ik3zM5eC0VA [16:47:06] 10VisualEditor, 10MediaWiki-General-or-Unknown, 10RESTBase, 6Services, and 4 others: Parsoid performance: Use RESTBase from the MediaWiki Virtual Rest Service on phase1/phase2 wikis - https://phabricator.wikimedia.org/T89066#1116873 (10Eloquence) [16:57:59] 10VisualEditor, 10Flow, 3Collaboration-Team-Sprint-S-2015-03-25, 3Roadmap: S5. Flow: Implement VE toolbar layout - https://phabricator.wikimedia.org/T90764#1116890 (10Mattflaschen) a:3Mattflaschen I'm going to do at least part of this as part of the mention patch. [17:08:39] 10OOjs-UI, 6Mobile-Web, 10UI-Standardization, 6WMF-Design, 7Design: Hover styles differ between MediaWiki UI buttons and OOjs UI buttons - https://phabricator.wikimedia.org/T92566#1116920 (10bmansurov) I vote for MediaWiki UI! [17:12:56] (03PS1) 10Kmenger: Window: Clarify description of setDimensions method [oojs/ui] - 10https://gerrit.wikimedia.org/r/196598 [17:22:08] (03PS1) 10Kmenger: Dialog: Fix links to static properties [oojs/ui] - 10https://gerrit.wikimedia.org/r/196600 [17:36:11] I'll probably be off IRC for the rest of the day as I'll be traveling, but I'll be reachable on Hangouts [17:36:19] Krinkle|detached: Review my TargetLoader changes :P [17:57:01] 10VisualEditor, 10MediaWiki-General-or-Unknown, 10RESTBase, 6Services, and 4 others: Parsoid performance: Use RESTBase from the MediaWiki Virtual Rest Service on group1/group2 wikis - https://phabricator.wikimedia.org/T89066#1117170 (10Krenair) [18:02:18] (03CR) 10Jforrester: [C: 032] Dialog: Fix links to static properties [oojs/ui] - 10https://gerrit.wikimedia.org/r/196600 (owner: 10Kmenger) [18:05:12] (03Merged) 10jenkins-bot: Dialog: Fix links to static properties [oojs/ui] - 10https://gerrit.wikimedia.org/r/196600 (owner: 10Kmenger) [18:10:52] (03CR) 10Jforrester: [C: 032] Window: Clarify description of setDimensions method [oojs/ui] - 10https://gerrit.wikimedia.org/r/196598 (owner: 10Kmenger) [18:13:19] (03Merged) 10jenkins-bot: Window: Clarify description of setDimensions method [oojs/ui] - 10https://gerrit.wikimedia.org/r/196598 (owner: 10Kmenger) [18:16:53]