[00:00:03] Helder: You want to talk to designers and you want to use IRC… :-) [00:00:22] hmm... didn't know it was contradictory =/ [00:00:30] :-( [00:00:58] Helder: Some of the designers are on IRC, but not all. I don't know which are the experts for which bits of MediaWiki UI. [00:01:58] We noticed a weird formatting in the font of the edit box after the dependency "mediawiki.ui.input" was added to a gadget... [00:02:05] (on ptwiki) [00:03:40] 3MediaWiki extensions / 3WikiEditor: WikiEditor: Add SVG versions of editing toolbar icons - 10https://bugzilla.wikimedia.org/35342#c64 (10Jared Zimmerman (WMF)) part of me is still wondering why we aren't using the icons that are in use in VisualEditor currently, any rational? [00:04:29] the effect can be seen by going to https://en.wikipedia.org/wiki/special:random?action=edit and typing mw.loader.load('mediawiki.ui.input') in the console [00:04:42] 3MediaWiki extensions / 3WikiEditor: WikiEditor: Add SVG versions of editing toolbar icons - 10https://bugzilla.wikimedia.org/35342#c65 (10James Forrester) Using the OOUI icons means using OOUI. Porting the wikitext editor to OOUI is a big piece of work. [00:05:12] 3VisualEditor: VisualEditor: line height in intro message is wrong causes collusions - 10https://bugzilla.wikimedia.org/72964#c3 (10Jared Zimmerman (WMF)) on en.wiki: go to article bonshō, click edit beta. [00:06:17] Helder: there's a #wikimedia-design where you'll find some of them [00:06:27] 3MediaWiki extensions / 3WikiEditor: WikiEditor: Add SVG versions of editing toolbar icons - 10https://bugzilla.wikimedia.org/35342#c66 (10Jared Zimmerman (WMF)) No, i just mean the same SVG assets. for bold, italic, image, link, etc. Not the code behind an OOUI control [00:07:26] 3MediaWiki extensions / 3WikiEditor: WikiEditor: Add SVG versions of editing toolbar icons - 10https://bugzilla.wikimedia.org/35342#c67 (10James Forrester) (In reply to Jared Zimmerman (WMF) from comment #66) > No, i just mean the same SVG assets. for bold, italic, image, link, etc. Not > the code behind an... [00:08:03] legoktm: Huh. Didn't know of that one. [00:08:55] James_F: it's +s for some reason.... [00:09:27] legoktm: thanks! I forgot about that channel! [00:12:49] legoktm: Even better. :-) [00:17:14] 3VisualEditor / 3ContentEditable: VisualEditor: Tables - applying headings format to table cells will split the table - 10https://bugzilla.wikimedia.org/72990 (10Rummana Yasmeen) [00:17:14] 3VisualEditor / 3Editing Tools: VisualEditor: Support tables (tracking) - 10https://bugzilla.wikimedia.org/39596 (10Rummana Yasmeen) [00:18:26] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf7] Link Suggestions are not appearing on placing the cursor inside it - 10https://bugzilla.wikimedia.org/72959#c3 (10Rummana Yasmeen) Verified the fix in Betalabs [00:21:12] 3VisualEditor / 3Editing Tools: VisualEditor: Regular context menu trying to render on TableSelections - 10https://bugzilla.wikimedia.org/72802#c7 (10Rummana Yasmeen) Verified the fix in Betalabs [00:21:13] 3VisualEditor / 3ContentEditable: VisualEditor: Drag-and-drop an inserted image slices a table - 10https://bugzilla.wikimedia.org/72995 (10etonkovidova) 3NEW p:3Unprio s:3normal a:3None Created attachment 17024 --> https://bugzilla.wikimedia.org/attachment.cgi?id=17024&action=edit Dragging images t... [00:27:11] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf7] Cannot add any Citations or Basic reference in a page , throwing Typeerror " Cannot read property 'replace' of undefined " - 10https://bugzilla.wikimedia.org/72948#c5 (10Rummana Yasmeen) 5RESO/FIX>3REOP This started happening again? Alth... [00:27:26] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf7] Cannot add any Citations or Basic reference in a page , throwing Typeerror " Cannot read property 'replace' of undefined " - 10https://bugzilla.wikimedia.org/72948 (10Rummana Yasmeen) 5REOP>3ASSI [00:37:14] 3VisualEditor / 3Mobile: VisualEditor Mobile: [Regression wmf4] Link suggestions are appearing in a broken way inside Link Inspector in mobile - 10https://bugzilla.wikimedia.org/72402#c2 (10James Forrester) 5ASSI>3RESO/FIX p:5Unprio>3Normal Appears to now be fixed having tested with an iPad against B... [00:47:43] RoanKattouw_away: https://gerrit.wikimedia.org/r/#/c/171011/ [00:50:34] RoanKattouw: https://gerrit.wikimedia.org/r/#/c/110129/ [00:53:24] (03PS1) 10Catrope: Fix indicator alignment in MWCategoryItemWidget [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171165 (https://bugzilla.wikimedia.org/72962) [00:54:43] (03CR) 10Jforrester: [C: 032] "Fine." [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171165 (https://bugzilla.wikimedia.org/72962) (owner: 10Catrope) [00:56:01] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf7] The down arrow to bring up the context menu for Categories is not center-aligned, again - 10https://bugzilla.wikimedia.org/72962 (10James Forrester) 5PATC>3RESO/FIX a:3Roan Kattouw [01:01:24] (03Merged) 10jenkins-bot: Fix indicator alignment in MWCategoryItemWidget [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171165 (https://bugzilla.wikimedia.org/72962) (owner: 10Catrope) [01:02:33] 3VisualEditor / 3ContentEditable: VisualEditor: Safari only - typing text in a table cell - the page scrolls up - 10https://bugzilla.wikimedia.org/72999 (10etonkovidova) 3NEW p:3Unprio s:3normal a:3None 1. Safari - open any page and make sure that a page has enough content to have a vertical scrollba... [01:07:56] 3MediaWiki extensions / 3WikiEditor: WikiEditor: Add SVG versions of editing toolbar icons - 10https://bugzilla.wikimedia.org/35342#c68 (10Jared Zimmerman (WMF)) Based on talking to James and Trevor, we can provide the icons that will improve consistency dramatically, we're targeting end of day tomorrow Nov... [01:10:11] 3VisualEditor / 3MediaWiki integration: VisualEditor: [Regression] Scroll into view of section on section edit link click has stopped working - 10https://bugzilla.wikimedia.org/72919#c1 (10Axel Boldt) I'm seeing the following behavior on Chrome 38, Windows 7: Go to https://en.wikipedia.org/wiki/OLED , scrol... [01:10:25] (03PS6) 10Catrope: Blur contentEditable node while destroying surface [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/144089 (https://bugzilla.wikimedia.org/71852) (owner: 10Inez) [01:11:49] (03PS7) 10Catrope: Blur contentEditable node while destroying surface [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/144089 (https://bugzilla.wikimedia.org/71852) (owner: 10Inez) [01:12:23] (03CR) 10Catrope: [C: 032] Blur contentEditable node while destroying surface [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/144089 (https://bugzilla.wikimedia.org/71852) (owner: 10Inez) [01:12:54] (03PS1) 10Catrope: Revert "Clear selection after tearing down surface to work around Firefox bug" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171177 [01:14:09] (03Merged) 10jenkins-bot: Blur contentEditable node while destroying surface [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/144089 (https://bugzilla.wikimedia.org/71852) (owner: 10Inez) [01:15:04] (03CR) 10Catrope: [C: 032] Kill MWCategoryWidget#onLookupEnter [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171033 (https://bugzilla.wikimedia.org/72906) (owner: 10Alex Monk) [01:18:16] (03Merged) 10jenkins-bot: Kill MWCategoryWidget#onLookupEnter [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171033 (https://bugzilla.wikimedia.org/72906) (owner: 10Alex Monk) [01:20:22] (03PS1) 10Catrope: Update VE core submodule to master (caaff2c) [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171180 [01:20:42] (03PS2) 10Catrope: Revert "Clear selection after tearing down surface to work around Firefox bug" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171177 [01:28:56] 3VisualEditor / 3MediaWiki integration: VisualEditor: [Regression] Don't let the user insert the same category multiple times - 10https://bugzilla.wikimedia.org/72906 (10James Forrester) 5PATC>3RESO/FIX [01:29:20] (03CR) 10Jforrester: [C: 032] Revert "Clear selection after tearing down surface to work around Firefox bug" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171177 (owner: 10Catrope) [01:29:27] (03CR) 10Jforrester: [C: 032] Update VE core submodule to master (caaff2c) [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171180 (owner: 10Catrope) [01:35:11] 3VisualEditor / 3MediaWiki integration: VisualEditor: [Regression] Scroll into view of section on section edit link click has stopped working - 10https://bugzilla.wikimedia.org/72919#c2 (10Axel Boldt) The VE version on mediawiki shows slightly different behavior, for instance on https://www.mediawiki.org/w/i... [01:39:11] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf7] Cannot add any Citations or Basic reference in a page , throwing Typeerror " Cannot read property 'replace' of undefined " - 10https://bugzilla.wikimedia.org/72948#c6 (10James Forrester) 5ASSI>3RESO/FIX Opening a distinct bug for that one. [01:39:15] (03Merged) 10jenkins-bot: Update VE core submodule to master (caaff2c) [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171180 (owner: 10Catrope) [01:39:20] (03Merged) 10jenkins-bot: Revert "Clear selection after tearing down surface to work around Firefox bug" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171177 (owner: 10Catrope) [01:50:42] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf7] Cite command not working (but cite-tranclusion does?) - 10https://bugzilla.wikimedia.org/73001 (10James Forrester) p:5Unprio>3Highes [01:50:43] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf7] Cite command not working (but cite-tranclusion does?) - 10https://bugzilla.wikimedia.org/73001 (10James Forrester) 3ASSI p:3Unprio s:3critic a:3Ed Sanders Can't work out why; supportedSelections is set for both. [01:52:11] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression wmf3] In Firefox, Cursor is appearing in Read mode at the end of the document after saving a page and after discarding the current edit in Firefox - 10https://bugzilla.wikimedia.org/71852 (10James Forrester) 5PATC>3RESO/FIX [01:56:25] 3VisualEditor / 3Technical Debt: VisualEditor: Remove ve.bind() - 10https://bugzilla.wikimedia.org/72156 (10James Forrester) [07:39:51] (03CR) 10Prtksxna: "Bartosz, thanks! Tested." [oojs/ui] - 10https://gerrit.wikimedia.org/r/170893 (owner: 10Prtksxna) [09:14:29] 3MediaWiki / 3ResourceLoader: referenceError for userdefined javascript functions - 10https://bugzilla.wikimedia.org/73011 (10Mayank Tiwari) 3UNCO p:3Unprio s:3major a:3None Created attachment 17028 --> https://bugzilla.wikimedia.org/attachment.cgi?id=17028&action=edit javascript reference error fo... [09:16:54] 3MediaWiki / 3ResourceLoader: referenceError for userdefined javascript functions - 10https://bugzilla.wikimedia.org/73011 (10Mayank Tiwari) [10:44:38] (03CR) 10Zfilipin: "And RuboCop complaining about a few minor things:" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/169251 (owner: 10Ryasmeen) [11:00:37] (03PS6) 10Esanders: Support for generic file drop handlers [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/170907 [11:20:49] (03PS1) 10Esanders: paralell -> parallel [oojs/ui] - 10https://gerrit.wikimedia.org/r/171233 [11:42:00] (03PS7) 10Esanders: Support for generic file drop handlers [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/170907 [11:51:54] 3VisualEditor / 3Editing Tools: VisualEditor: Tool to set/unset text colour - 10https://bugzilla.wikimedia.org/52645#c8 (10Elitre) Would we need to file a separate bug for background text color? [12:17:00] (03PS1) 10Prtksxna: mediawiki: Replace instances of 0.3em border-radius with variable [oojs/ui] - 10https://gerrit.wikimedia.org/r/171240 [12:18:04] (03CR) 10Prtksxna: "Need to discuss with May how I can get rid of the other border radii. Trevor, did you add the other values according to some spec?" [oojs/ui] - 10https://gerrit.wikimedia.org/r/171240 (owner: 10Prtksxna) [12:34:10] 3MediaWiki / 3ResourceLoader: referenceError for userdefined javascript functions - 10https://bugzilla.wikimedia.org/73011#c1 (10Andre Klapper) If you assume that there is a software bug in MediaWiki, could you provide a minimal self-contained testcase? [13:51:19] (03CR) 10Krinkle: [C: 04-1] Move most of ve.js to ve.utils.js (031 comment) [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/170874 (owner: 10Catrope) [14:10:41] 3MediaWiki / 3ResourceLoader: resourceloader: Function defined javascript function does not exist in global scope - 10https://bugzilla.wikimedia.org/73011#c2 (10Krinkle) 5UNCO>3RESO/INV Executing files in the global scope is considered an anti-pattern in modern web development and causes all kinds of int... [14:12:17] (03CR) 10Krinkle: [C: 032] paralell -> parallel [oojs/ui] - 10https://gerrit.wikimedia.org/r/171233 (owner: 10Esanders) [14:14:34] (03CR) 10Krinkle: [BREAKING CHANGE] Remove the ve.bind function (031 comment) [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171002 (https://bugzilla.wikimedia.org/72156) (owner: 10Jforrester) [14:14:37] (03PS2) 10Krinkle: [BREAKING CHANGE] Remove the ve.bind function [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171002 (https://bugzilla.wikimedia.org/72156) (owner: 10Jforrester) [14:14:39] (03Merged) 10jenkins-bot: paralell -> parallel [oojs/ui] - 10https://gerrit.wikimedia.org/r/171233 (owner: 10Esanders) [14:15:33] Hi everyone! I'm trying to setup the visualeditor in conjunction with parsoid on a https Server. The visualeditor loads (having fixed quite a lot of http errors), but appears stuck in the loading animation (without any further errors). Could you provide me with a hint as to how to move on from here, and debug the setup further? [14:18:11] Parsoid is accessible from outside (via http), so that part should work. The wiki itself also works (via the basic editor). [14:22:40] mezzopiano: are there any javascript errors in the console when you try to load VE? [14:22:57] No, the javascript console appears completely empty. [14:23:34] what versions of MW and VE are you running? [14:23:49] I was just comparing the console output with the offical visualeditor demo -- the only difference I see is that my installation never gets beyond the first request to api.php, while the demo uses tons of such requests. [14:24:21] burntsushi__: Please give me a second, I'll find out. Huge thanks for taking the time to respond! [14:26:15] VE is VisualEditor: REL1_23 [14:26:39] MW is also 1.23 [14:26:56] mezzopiano: i see. i've never been able to get those old releases working. [14:27:19] This exact same setup worked on a http test machine, but totally broke after switching to https. [14:28:08] mezzopiano: did you update the Parsoid config to point to the new MW url? [14:28:13] burntsushi__: But isn't 1.23 the most recent release? [14:28:45] mezzopiano: i'm not familiar with how they label their releases. i just try to track Wikipedia, which is currently running 1.25wmf5: http://en.wikipedia.org/wiki/Special:Version [14:29:16] burntsushi__: Regarding the config: Yes, I did -- I was getting error messages prior to reconfiguring the urls. Now it's just silent. [14:29:24] also, have you checked the parsoid logs? it should show something useful when it gets a request. [14:29:46] burntsushi__: (regarding the versions:) Ah, I see -- I stuck with the stable release. [14:29:54] I'll take a look at the logs now [14:30:12] mezzopiano: the way i figure it---if the `wmf` releases are stable enough for Wikipedia, then they're stable enough for me. :-) [14:30:20] :-) [14:30:59] Ah, now I see some failed API requests. [14:32:21] ... but they aren't recent: No new log entries are added when I try to use the visualeditor. [14:33:05] weird. so i guess that probably means that MediaWiki isn't able to contact Parsoid? hmm. i'm assuming you *also* fixed the parsoid url on the mediawiki side? [14:33:13] Also, I just restarted parsoid and I was getting error messages from the wiki ui while it was down. I would interpret that as there being some kind of connection. [14:33:15] (if you didn't, i imagine that would give you an error anyway...) [14:33:23] interesting. [14:33:40] 3MediaWiki / 3ResourceLoader: resourceloader: Function defined javascript function does not exist in global scope - 10https://bugzilla.wikimedia.org/73011#c3 (10Mayank Tiwari) So you mean to say I shall write it like mw.testFunction() = function(){ }; or say mw.testFunction = function(a,b){ return a+b; };... [14:35:22] Yep, it's strange. I thought first that it might be HSTS, but still no improvements after disabling it (and clearing the browser hsts caches) [14:36:11] mezzopiano: yeah, usually your browser will report an error for an HSTS violation anyway. [14:36:44] burntsushi__: Ok, but I should see that, right? [14:36:57] yeah [14:37:13] mezzopiano: definitely strange. i'm not sure where else to go. :-/ [14:37:49] burntsushi__: btw., do you by any chance know how wikipedia runs its parsoid services? Do they use http, or did they configure parsoid to run via https? [14:38:29] burntsushi__: Hm, yeah, it's a pity that it doesn't run -- it's such a cool editor. Are you involved in development? [14:39:05] i have no clue, unfortunately. [14:39:41] mezzopiano: nope. i'm just someone who is trying to use it for $work, and i went through a lot of pain getting it working. once i moved to the latest version, things seemed to work better. [14:39:53] mezzopiano: although, i haven't done anything with https yet. [14:40:25] I've read that a lot of work is underway to package things and make the installation easier. [14:40:49] oh really? that'd be nice. :-) [14:41:04] we have everything inside dockers right now. parsoid + nginx + mysql + mediawiki [14:41:26] (it was on one of the VisualEditor wiki pages) [14:41:57] burntsushi__: Nice setup! I'm more of an amateur, so no containers here :-) [14:42:46] burntsushi__: Ok, I guess I'll have to try a more recent release. Thanks a lot for taking the time and giving me hints and pointers -- I had no idea that there was another regular release stream? (I only saw the "official" stable releases on mediawiki.org) [14:42:47] mezzopiano: Docker is the new kid on the block, but it's kind of fun to use. definitely worth a try, regardless of professional status :) [14:43:09] mezzopiano: we clone everything from `git` and then checkout the appropriate branches. [14:43:18] mezzopiano: and yeah, it's not exactly well advertised. [14:43:47] burntsushi__: Wow, but you clearly seem to have spent a lot of time figuring this out! [14:43:47] mezzopiano: you can see everything here: http://git.wikimedia.org/repositories/ [14:44:03] mezzopiano: yeah, i've been in this world for a couple months now. [14:44:15] burntsushi__: Cool, I'll have a look. (at docker too, but I'll have to get round to that). Thanks a lot! [14:44:28] no problem! good luck! [14:45:29] burntsushi__: And you, all the best, and kind regards from Germany! Thanks again for helping out [14:48:10] 3MediaWiki / 3ResourceLoader: resourceloader: Function defined javascript function does not exist in global scope - 10https://bugzilla.wikimedia.org/73011#c4 (10Krinkle) (In reply to Mayank Tiwari from comment #3) > So you mean to say I shall write it like > > mw.testFunction() = function(){ }; > > or say... [14:49:25] 3MediaWiki / 3ResourceLoader: resourceloader: Function defined javascript function does not exist in global scope - 10https://bugzilla.wikimedia.org/73011#c5 (10Krinkle) (In reply to Mayank Tiwari from comment #3) > I owe you apology in case it was not appropriate but I'll try it tomorrow > and will let you... [15:10:40] (03PS1) 10Esanders: Node method cleanups [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171269 [15:10:42] (03PS1) 10Esanders: Add isAllowed(Child|Parent)NodeType to ve.Node [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171270 [15:10:44] (03PS1) 10Esanders: Replace block drop logic with isAllowedChildNodeType check [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171271 (https://bugzilla.wikimedia.org/72995) [15:28:55] (03PS2) 10Esanders: Node method cleanups [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171269 [15:28:57] (03PS2) 10Esanders: Replace block drop logic with isAllowedChildNodeType check [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171271 (https://bugzilla.wikimedia.org/72995) [15:28:59] (03PS2) 10Esanders: Add isAllowed(Child|Parent)NodeType to ve.Node [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171270 [15:30:26] 3VisualEditor / 3ContentEditable: VisualEditor: Tables - using keyboard to move cursor throws "TypeError: surfaceView.getNativeRangeBoundingClientRect is not a function" - 10https://bugzilla.wikimedia.org/72978#c3 (10Ed Sanders) 5NEW>3RESO/FIX Was fixed by https://gerrit.wikimedia.org/r/#/c/170898/ [15:37:39] 3VisualEditor / 3Editing Tools: VisualEditor: Annotations are available even when local CSS rules make them ineffective - 10https://bugzilla.wikimedia.org/72949#c2 (10Ed Sanders) We already have this issue with headings. The text is not actually annotated as bold and they would produce different wikitext (==... [15:39:10] 3VisualEditor / 3Editing Tools: VisualEditor: Annotations can conflic with local CSS rules, e.g. bold text in headings - 10https://bugzilla.wikimedia.org/72949 (10Ed Sanders) [15:45:54] (03PS3) 10Krinkle: [BREAKING CHANGE] Remove the ve.bind function [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171002 (https://bugzilla.wikimedia.org/72156) (owner: 10Jforrester) [15:46:31] (03PS4) 10Krinkle: [BREAKING CHANGE] Remove the ve.bind function [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171002 (https://bugzilla.wikimedia.org/72156) (owner: 10Jforrester) [15:46:51] (03CR) 10Krinkle: [C: 031] [BREAKING CHANGE] Remove the ve.bind function [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171002 (https://bugzilla.wikimedia.org/72156) (owner: 10Jforrester) [15:47:40] 3VisualEditor / 3Editing Tools: VisualEditor: After saving, the order of the texts gets reversed while adding some content in a slug created inside the caption field of a table - 10https://bugzilla.wikimedia.org/72833#c3 (10Ed Sanders) Still works on my local machine. The wikitext produced is: {| |+ Foo * B... [16:01:25] 3VisualEditor / 3Editing Tools: VisualEditor: After saving, the order of the texts gets reversed while adding some content in a slug created inside the caption field of a table - 10https://bugzilla.wikimedia.org/72833#c4 (10Ed Sanders) Apparently this is a bug with HTMLTidy. [16:05:12] 3VisualEditor / 3Editing Tools: VisualEditor:After adding a focusable node inside a content cell , placing cursor at the end of the table scrolls the page down - 10https://bugzilla.wikimedia.org/72804#c3 (10Ed Sanders) 5ASSI>3RESO/DUP Oh *that* outside of the table. Yeah - you're not supposed to put your... [16:05:12] 3VisualEditor / 3ContentEditable: VisualEditor: Chrome lets you place the cursor either side of a Table (eww) and VE doesn't correct this - 10https://bugzilla.wikimedia.org/71560#c2 (10Ed Sanders) *** Bug 72804 has been marked as a duplicate of this bug. *** [16:05:40] 3VisualEditor / 3ContentEditable: VisualEditor: Chrome lets you place the cursor either side of a Table (eww) and VE doesn't correct this - 10https://bugzilla.wikimedia.org/71560#c3 (10Ed Sanders) Apparently can also cause scroll issues (bug 72804) [16:30:30] Corruption alert: visualeditor-needcheck on sqwiki: https://sq.wikipedia.org/?diff=1469472 [16:30:30] Corruption alert: visualeditor-needcheck on mediawikiwiki: https://www.mediawiki.org/?diff=1252444 [16:30:30] Corruption alert: visualeditor-needcheck on frwiki: https://fr.wikipedia.org/?diff=108856192 [16:30:30] Corruption alert: visualeditor-needcheck on zhwiki: https://zh.wikipedia.org/?diff=33186882 [16:30:30] Corruption alert: visualeditor-needcheck on zhwiki: https://zh.wikipedia.org/?diff=33188962 [16:30:31] Corruption alert: visualeditor-needcheck on cawiki: https://ca.wikipedia.org/?diff=14254451 [16:30:31] Corruption alert: visualeditor-needcheck on itwiki: https://it.wikipedia.org/?diff=68994547 [16:30:39] 3VisualEditor / 3ContentEditable: VisualEditor: Tables - multiple colspan and rowspan are not displayed correctly - 10https://bugzilla.wikimedia.org/72790#c4 (10Ed Sanders) I suspect that
Foo Foo
is bad HTML... [16:36:21] (03PS1) 10Esanders: Don't focus Edit HTML on load [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171279 [16:37:36] (03PS1) 10Esanders: Create table cells with wrapper paragraphs by default [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171280 [16:45:28] 3VisualEditor / 3ContentEditable: VisualEditor: Tables - multiple colspan and rowspan are not displayed correctly - 10https://bugzilla.wikimedia.org/72790#c5 (10Ed Sanders) The first case (merging two whole rows together) exposes a bug in the PHP parser (although not Tidy, because I don't have that locally)... [16:52:41] 3VisualEditor / 3ContentEditable: VisualEditor: Tables - using keyboard to move cursor throws "TypeError: surfaceView.getNativeRangeBoundingClientRect is not a function" - 10https://bugzilla.wikimedia.org/72978 (10James Forrester) p:5Unprio>3Normal a:3Ed Sanders [17:00:28] 3VisualEditor / 3ContentEditable: VisualEditor: Tables - applying headings format to table cells will split the table - 10https://bugzilla.wikimedia.org/72990#c1 (10Ed Sanders) MediaWiki headings are only supposed to every be at document level. We either need to allow them in table cells, or disable them in... [17:01:57] (03PS1) 10Cmcmahon: QA: remove stuff now that iframes are gone [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171283 [17:04:28] (03CR) 10Cmcmahon: [C: 032] "post-iframeocalypse maintenance" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171283 (owner: 10Cmcmahon) [17:06:42] 3MediaWiki / 3ResourceLoader: Rights icon doesn't appear when $wgResourceBasePath is used - 10https://bugzilla.wikimedia.org/73031 (10Nathan Larson) 3NEW p:3Unprio s:3minor a:3None The installer's $wgRightsIcon = "{$wgResourceBasePath}/resources/assets/licenses/cc-by-sa.png"; causes the rights icon... [17:06:43] (03Merged) 10jenkins-bot: QA: remove stuff now that iframes are gone [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171283 (owner: 10Cmcmahon) [17:18:38] (03PS1) 10Esanders: Allow headings in table cells [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171288 (https://bugzilla.wikimedia.org/72990) [17:19:55] (03CR) 10jenkins-bot: [V: 04-1] Allow headings in table cells [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171288 (https://bugzilla.wikimedia.org/72990) (owner: 10Esanders) [17:40:39] 3VisualEditor / 3Editing Tools: VisualEditor: Annotations can conflict with local CSS rules, e.g. bold text in headings - 10https://bugzilla.wikimedia.org/72949 (10Ed Sanders) [17:43:39] edsanders: Nice re-phrase there. [17:45:50] (03CR) 10Jforrester: "Test failing because of https://bugzilla.wikimedia.org/show_bug.cgi?id=72398 and lack of a local over-ride?" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171288 (https://bugzilla.wikimedia.org/72990) (owner: 10Esanders) [17:45:55] 3VisualEditor: Forward delete key does crazy things - 10https://bugzilla.wikimedia.org/72690#c2 (10Ed Sanders) Can you be more specific? Delete and backspace appear to work fine for me in that document. [17:47:44] (03CR) 10Jforrester: "Are any of the scrolling issues we have related? https://bugzilla.wikimedia.org/buglist.cgi?bug_severity=blocker&bug_severity=critical&bug" [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171279 (owner: 10Esanders) [17:47:52] I think it's a wontfix in HTML, unless you want to provide some crazy [17:47:57] 3OOjs / 3EventEmitter: EventEmitter and events such as 'watch' - 10https://bugzilla.wikimedia.org/73034 (10Jon) 3NEW p:3Unprio s:3normal a:3None {}.watch exists in Firefox when using OO.EventEmitter.on with the event name 'watch' in Firefox it triggers an error `get a JS error (TypeError: this.bindin... [17:48:02] also it's skin-dependent so meh [17:48:19] edsanders: Yeah. I thought of leaving it open in case some magic way comes along in the future. [17:50:10] 3OOjs / 3EventEmitter: EventEmitter and events such as 'watch' - 10https://bugzilla.wikimedia.org/73034#c1 (10Ed Sanders) watch sounds incredibly generic and as such a bad event name :) [17:53:12] 3VisualEditor / 3ContentEditable: VisualEditor: Safari only - the page is blurred in Edit mode - 10https://bugzilla.wikimedia.org/73035 (10etonkovidova) 3NEW p:3Unprio s:3normal a:3None Created attachment 17037 --> https://bugzilla.wikimedia.org/attachment.cgi?id=17037&action=edit Safari blurs text... [17:53:57] (03PS3) 10Mvolz: Adds GET endpoint /api, takes URL & DOI [services/citoid] - 10https://gerrit.wikimedia.org/r/169690 [17:54:10] 3OOjs / 3EventEmitter: EventEmitter and events such as 'watch' - 10https://bugzilla.wikimedia.org/73034#c2 (10Ed Sanders) How exactly are you triggering this error? widget.on('watch', function () {} ); widget.emit('watch'); works for me. [17:55:18] 3VisualEditor / 3ContentEditable: VisualEditor: Safari only - the page is blurred in Edit mode - 10https://bugzilla.wikimedia.org/73035#c1 (10etonkovidova) Created attachment 17038 --> https://bugzilla.wikimedia.org/attachment.cgi?id=17038&action=edit Safari displays clear text in Read mode [17:55:27] 3OOjs / 3EventEmitter: EventEmitter and events such as 'watch' - 10https://bugzilla.wikimedia.org/73034#c3 (10Ed Sanders) I see if you emit without binding anything you get a problem (although that begs the question why you'd want to do that). If we just wrap that code in a hasOwnProperty check there won't b... [17:59:12] 3VisualEditor / 3ContentEditable: VisualEditor: Safari only - the page is blurred in Edit mode - 10https://bugzilla.wikimedia.org/73035#c2 (10etonkovidova) Created attachment 17039 --> https://bugzilla.wikimedia.org/attachment.cgi?id=17039&action=edit Safari Read mode - text is clear [17:59:55] 3VisualEditor / 3MediaWiki integration: VisualEditor: The show/hide text a collapsed navbox disappear afted have edited - 10https://bugzilla.wikimedia.org/68279#c10 (10Alex Monk) 5ASSI>3RESO/WOR Please reopen if you can still reproduce this issue. [18:03:10] (03PS1) 10Esanders: Use hasOwn check in EventEmitter#emit [oojs/core] - 10https://gerrit.wikimedia.org/r/171293 (https://bugzilla.wikimedia.org/73034) [18:08:05] (03PS2) 10Esanders: Use hasOwn check in EventEmitter#emit [oojs/core] - 10https://gerrit.wikimedia.org/r/171293 (https://bugzilla.wikimedia.org/73034) [18:14:36] (03PS1) 10Cmcmahon: QA: WIP: sort remaining page elements that belong in the iframe [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171295 [18:15:42] (03PS3) 10Krinkle: EventEmitter: Use hasOwn check in #emit [oojs/core] - 10https://gerrit.wikimedia.org/r/171293 (https://bugzilla.wikimedia.org/73034) (owner: 10Esanders) [18:17:44] (03CR) 10Krinkle: [C: 031] EventEmitter: Use hasOwn check in #emit (031 comment) [oojs/core] - 10https://gerrit.wikimedia.org/r/171293 (https://bugzilla.wikimedia.org/73034) (owner: 10Esanders) [18:32:11] 3VisualEditor / 3ContentEditable: VisualEditor: [Regression pre-wmf7] Safari only - the page is blurred in Edit mode - 10https://bugzilla.wikimedia.org/73035 (10etonkovidova) [18:33:57] (03CR) 10Catrope: [C: 032] Don't focus Edit HTML on load [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171279 (owner: 10Esanders) [18:36:02] (03CR) 10Catrope: Move most of ve.js to ve.utils.js (031 comment) [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/170874 (owner: 10Catrope) [18:36:07] (03Merged) 10jenkins-bot: Don't focus Edit HTML on load [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171279 (owner: 10Esanders) [18:39:55] 3OOjs / 3EventEmitter: EventEmitter and events such as 'watch' - 10https://bugzilla.wikimedia.org/73034 (10Bartosz Dziewoński) [18:42:49] (03CR) 10Krinkle: [C: 032] EventEmitter: Use hasOwn check in #emit [oojs/core] - 10https://gerrit.wikimedia.org/r/171293 (https://bugzilla.wikimedia.org/73034) (owner: 10Esanders) [18:43:11] 3VisualEditor / 3ContentEditable: VisualEditor: [Regression pre-wmf7] Safari only - the page is blurred in Edit mode - 10https://bugzilla.wikimedia.org/73035#c3 (10etonkovidova) Created attachment 17040 --> https://bugzilla.wikimedia.org/attachment.cgi?id=17040&action=edit Safari Read mode - text s clear [18:46:11] (03Merged) 10jenkins-bot: EventEmitter: Use hasOwn check in #emit [oojs/core] - 10https://gerrit.wikimedia.org/r/171293 (https://bugzilla.wikimedia.org/73034) (owner: 10Esanders) [18:46:29] 3VisualEditor / 3Mobile: VisualEditor Mobile: "Uncaught Error: Unknown dependency: rangefix" appears while trying to load mobile VE - 10https://bugzilla.wikimedia.org/73038 (10Rummana Yasmeen) 3NEW p:3Unprio s:3critic a:3None Created attachment 17041 --> https://bugzilla.wikimedia.org/attachment.c... [18:46:40] 3VisualEditor / 3Mobile: VisualEditor Mobile: "Uncaught Error: Unknown dependency: rangefix" appears while trying to load mobile VE - 10https://bugzilla.wikimedia.org/73038 (10Rummana Yasmeen) p:5Unprio>3Highes [18:48:10] 3OOjs / 3EventEmitter: EventEmitter and events such as 'watch' - 10https://bugzilla.wikimedia.org/73034#c6 (10Jon) 5PATC>3RESO/FIX Perfect! :) Thanks for quick response! [18:51:59] (03PS9) 10Divec: Refactor SurfaceObserver pollOnceInternal [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/166188 [18:52:03] (03CR) 10jenkins-bot: [V: 04-1] Refactor SurfaceObserver pollOnceInternal [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/166188 (owner: 10Divec) [18:53:24] edsanders: https://gerrit.wikimedia.org/r/171291 was the patch I auto-abandoned. [18:56:15] (03CR) 10Catrope: [C: 032] Create table cells with wrapper paragraphs by default [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171280 (owner: 10Esanders) [18:56:54] RoanKattouw: Told you: https://bugzilla.wikimedia.org/show_bug.cgi?id=73038 [18:57:30] (03PS26) 10Divec: Use a single unicorn-like instead of  in inline slugs [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/151046 (owner: 10Catrope) [18:57:31] Urgh [18:57:39] I need 'target' => array( 'desktop', 'mobile' ) [18:57:42] Stupid targets [18:57:50] (03PS10) 10Divec: Refactor SurfaceObserver pollOnceInternal [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/166188 [18:57:50] * James_F nods. [18:58:32] (03Merged) 10jenkins-bot: Create table cells with wrapper paragraphs by default [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171280 (owner: 10Esanders) [18:58:43] (03CR) 10Divec: "Rebase" [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/151046 (owner: 10Catrope) [18:59:07] (03CR) 10Divec: "Rebase" [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/166188 (owner: 10Divec) [18:59:12] (03PS4) 10Catrope: Move most of ve.js to ve.utils.js [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/170874 [18:59:49] (03CR) 10jenkins-bot: [V: 04-1] Refactor SurfaceObserver pollOnceInternal [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/166188 (owner: 10Divec) [19:01:11] (03PS1) 10Jforrester: Tag v1.1.2 [oojs/core] - 10https://gerrit.wikimedia.org/r/171304 [19:01:21] Krinkle: ^^^ Want to +2? [19:01:37] James_F: not yet, I'm about to submit another oojscore commit [19:01:45] Argh. OK. [19:02:01] (03CR) 10Krinkle: [C: 04-1] "Got another minor clean up I'd like to push into 1.1.2" [oojs/core] - 10https://gerrit.wikimedia.org/r/171304 (owner: 10Jforrester) [19:03:29] (03PS1) 10Catrope: Followup 07251d0: add mobile target to rangefix [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171306 (https://bugzilla.wikimedia.org/73038) [19:04:40] (03CR) 10jenkins-bot: [V: 04-1] Followup 07251d0: add mobile target to rangefix [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171306 (https://bugzilla.wikimedia.org/73038) (owner: 10Catrope) [19:05:11] (03PS2) 10Jforrester: Follow-up 07251d0: Add mobile target to RangeFix [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171306 (https://bugzilla.wikimedia.org/73038) (owner: 10Catrope) [19:05:21] (03CR) 10Jforrester: [C: 032] Follow-up 07251d0: Add mobile target to RangeFix [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171306 (https://bugzilla.wikimedia.org/73038) (owner: 10Catrope) [19:06:18] 3VisualEditor / 3Mobile: VisualEditor Mobile: Calling VisualEditor from MF spins forever - 10https://bugzilla.wikimedia.org/73039 (10Rummana Yasmeen) [19:06:18] 3VisualEditor / 3Mobile: Calling VisualEditor from MF spins forever - 10https://bugzilla.wikimedia.org/73039#c2 (10Bingle) Prioritization and scheduling of this bug is tracked on Trello card https://trello.com/c/z22jBbGj [19:06:18] 3VisualEditor / 3Mobile: VisualEditor Mobile: "Uncaught Error: Unknown dependency: rangefix" appears while trying to load mobile VE - 10https://bugzilla.wikimedia.org/73038#c1 (10Rummana Yasmeen) *** Bug 73039 has been marked as a duplicate of this bug. *** [19:06:19] (03CR) 10jenkins-bot: [V: 04-1] Follow-up 07251d0: Add mobile target to RangeFix [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171306 (https://bugzilla.wikimedia.org/73038) (owner: 10Catrope) [19:06:20] 3VisualEditor / 3Mobile: VisualEditor Mobile: [Regression wmf7] "Uncaught Error: Unknown dependency: rangefix" appears while trying to load mobile VE - 10https://bugzilla.wikimedia.org/73038 (10James Forrester) 5NEW>3ASSI a:3Roan Kattouw [19:09:16] (03PS3) 10Jforrester: Follow-up 07251d0: Add mobile target to RangeFix [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171306 (https://bugzilla.wikimedia.org/73038) (owner: 10Catrope) [19:09:25] (03CR) 10Jforrester: [C: 032] Follow-up 07251d0: Add mobile target to RangeFix [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171306 (https://bugzilla.wikimedia.org/73038) (owner: 10Catrope) [19:09:26] 3VisualEditor / 3Mobile: Calling VisualEditor from MF spins forever - 10https://bugzilla.wikimedia.org/73039#c1 (10Rummana Yasmeen) 5NEW>3RESO/DUP a:3None *** This bug has been marked as a duplicate of bug 73038 *** [19:10:39] mooeypoo: hi [19:10:41] 3VisualEditor / 3Mobile: VisualEditor Mobile: [Regression pre-wmf7] "Uncaught Error: Unknown dependency: rangefix" appears while trying to load mobile VE - 10https://bugzilla.wikimedia.org/73038 (10James Forrester) 5PATC>3RESO/FIX [19:10:41] sorry for the delay getting on IRC [19:10:47] mooeypoo: I'm happy to help you now [19:10:47] (03Merged) 10jenkins-bot: Follow-up 07251d0: Add mobile target to RangeFix [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171306 (https://bugzilla.wikimedia.org/73038) (owner: 10Catrope) [19:15:12] TrevorParscal, okay, I have a broken demo up on github [19:15:20] I'm trying to figure out which event can give me what. [19:15:38] cool [19:15:42] repo link? [19:15:43] TrevorParscal, right now, I'm trying to figure out how I can discern which object i'm "on top of" while dragging. [19:16:05] the drag event (which you can listen to at the group level) should have that inside it [19:16:23] So here's the thing [19:16:39] the drag event on the top widget? only fires when you're *not* on top of individual elements [19:16:41] fun, right? [19:16:43] dragOver [19:16:50] yes, dragover [19:17:03] that on the parent widget only fires when you're OUTSIDE the internal images [19:17:10] 3OOjs / 3EventEmitter: EventEmitter and events such as 'watch' - 10https://bugzilla.wikimedia.org/73034#c7 (10James Forrester) Will go into OOjs v1.1.2, which we'll release today and get into MW for 1.25wmf8. Is this sufficient, or would you need it back-ported? [19:17:11] so I have to use individual elements' dragover [19:17:18] they emit their own "dragover" the parent collects [19:17:22] so far so good, actually, i think [19:17:29] but then I need to figure out which one of them will "expand" [19:17:48] TrevorParscal, so, right now, my widget recognizes which element was dragged, which it is dragged *over* and where it was dropped [19:18:00] but my problem is while you're dragging oer stuff -- where to expand the margins [19:18:46] looking at my code, it appears that I'm actually calculating the "what is being dragged over" https://github.com/collabkit/future/blob/master/brie/lib/client/ux.gridList/ux.gridList.js#L432 [19:18:49] (03PS1) 10Krinkle: EventEmitter: Use hasOwn check in #off [oojs/core] - 10https://gerrit.wikimedia.org/r/171311 [19:18:51] (03PS1) 10Krinkle: EventEmitter: Simplify #validateMethod and increase test coverage [oojs/core] - 10https://gerrit.wikimedia.org/r/171312 [19:18:51] TrevorParscal, so after this whole intro, here's the repo -- it's using 'animate' on margins to try and get the elements to expand, only -- as you will see -- it's doing it all wrong. I call it "the dragging dance." Should sell tickets. [19:19:06] TrevorParscal, yes, I also looked at document.elementFromPoint [19:19:23] That works great except all items have multiple nested elements in them; span->div->div etc. [19:19:36] i'm iterating through every item, checking the bounding boxes to see if the drag left and top are inside them, and even deciding if it's to the left or right of center (you may need to decide if it's to the top or bottom of center if it's vertical stacking) [19:19:37] I *could* calculate the parent item from that, but I saved that idea for "omg i have nothing else working" [19:19:45] (03CR) 10Catrope: [C: 04-1] EventEmitter: Use hasOwn check in #off (031 comment) [oojs/core] - 10https://gerrit.wikimedia.org/r/171311 (owner: 10Krinkle) [19:19:57] you want to only iterate over the option widgets, not their children [19:20:01] TrevorParscal, https://github.com/mooeypoo/ooui-reorder <-- the repo. Full of commented ou stuff and weird behavior, but you can see where I'm trying to go with this [19:20:02] (03PS2) 10Krinkle: EventEmitter: Simplify #validateMethod and increase test coverage [oojs/core] - 10https://gerrit.wikimedia.org/r/171312 [19:20:04] (03PS2) 10Krinkle: EventEmitter: Use hasOwn check in #off [oojs/core] - 10https://gerrit.wikimedia.org/r/171311 [19:20:22] this approach is hurt badly by increasing the number of elements being iterated over [19:20:26] (03CR) 10Catrope: [C: 032] EventEmitter: Use hasOwn check in #off [oojs/core] - 10https://gerrit.wikimedia.org/r/171311 (owner: 10Krinkle) [19:20:31] (03CR) 10Krinkle: EventEmitter: Use hasOwn check in #off (031 comment) [oojs/core] - 10https://gerrit.wikimedia.org/r/171311 (owner: 10Krinkle) [19:20:44] (03CR) 10Catrope: [C: 032] EventEmitter: Simplify #validateMethod and increase test coverage [oojs/core] - 10https://gerrit.wikimedia.org/r/171312 (owner: 10Krinkle) [19:21:08] TrevorParscal, aye, so, if you want to calculate which element you're over on the widget level, it seems to be a pain here. It's either with elementFromPoint which gives you all the underlying elements the OptionWidget is *made* of, or it's not updating itself properly (in 'real time' movement) when the elements are moving [19:21:16] and they are moving, because we need to shove something in there [19:21:43] So instead, the element fires up an event notifying the widget that it's dragging over them [19:21:46] that seems to work, actually [19:21:55] It recognizes where it lands correctly [19:22:18] if you comment out the idiotic animate() thing, you'll see in the #status output that the dragging information works fine [19:22:25] 3OOjs / 3EventEmitter: OOjs: EventEmitter fails when emitting unbound event with name of Object prototype - 10https://bugzilla.wikimedia.org/73034 (10Krinkle) [19:22:31] the problem right now seems to be the "shoving it aside" effect [19:22:50] (03CR) 10Krinkle: "Good to go (after re-base to include the extra bugfix)" [oojs/core] - 10https://gerrit.wikimedia.org/r/171304 (owner: 10Jforrester) [19:23:09] TrevorParscal, I'm also a bit confused about how you're making the dragging effect itself, you seem to append something to the image? *BUT* before I go into that, I wanted to see if I can calculate the position of the drag over the item *correctly* which at the moment is the biggest problem [19:23:12] * Krinkle does that [19:23:24] Something's wrong with my calculation. It's wonky, to say the least. [19:24:10] * mooeypoo waits for TrevorParscal to see 'the dragging dance' [19:24:23] Dance, OptionWidgets, dance. [19:25:24] (03Merged) 10jenkins-bot: EventEmitter: Use hasOwn check in #off [oojs/core] - 10https://gerrit.wikimedia.org/r/171311 (owner: 10Krinkle) [19:25:48] (03Merged) 10jenkins-bot: EventEmitter: Simplify #validateMethod and increase test coverage [oojs/core] - 10https://gerrit.wikimedia.org/r/171312 (owner: 10Krinkle) [19:26:01] (03PS2) 10Krinkle: Tag v1.1.2 [oojs/core] - 10https://gerrit.wikimedia.org/r/171304 (owner: 10Jforrester) [19:26:10] (03PS3) 10Krinkle: Tag v1.1.2 [oojs/core] - 10https://gerrit.wikimedia.org/r/171304 (owner: 10Jforrester) [19:26:16] Krinkle: Bah, I was in the middle of fixing that. [19:27:14] (03CR) 10Jforrester: [C: 04-1] Tag v1.1.2 [oojs/core] - 10https://gerrit.wikimedia.org/r/171304 (owner: 10Jforrester) [19:30:19] mooeypoo: if you use elementFromPoint, you can then just use $( this.getElementDocument().elementFromPoint( drag.left, drag.top ) ).closest( '.oo-ui-optionWidget' ); [19:30:28] yeah? [19:30:29] yeah that's what I'm doing now, hang on [19:30:33] ok [19:30:37] TrevorParscal, pushing another version [19:31:05] ok new version pushed [19:32:24] obj = this.getElementDocument().elementFromPoint( pageX, pageY ); [19:32:24] $optWidget = $( obj ).closest( '.reorderItemWidget' ); [19:32:37] Seems to work so far. [19:33:15] I tried to avoid that since the Items already emit events when the drag is on top of them, and this is a bit ugly [19:33:23] But it does seem to be the best way [19:35:30] (03CR) 10Bartosz Dziewoński: [C: 04-1] "These are not all instances, there's stuff like "border-bottom-left-radius: 0.3em" too. Grep for "radius: 0.3em"?" [oojs/ui] - 10https://gerrit.wikimedia.org/r/171240 (owner: 10Prtksxna) [19:36:24] TrevorParscal, I think my issue now is calculating the actual position. I'm getting weird results, but it's probably not figuring out the offset vs position properly. [19:38:22] hm. This seems to be more or less the same as the events triggered from the first version. The issue is the position calculations. [19:38:27] * mooeypoo will work on figuring that out [19:40:55] 3OOjs: OOjs: v8 warns it won't optimize oo.Factory.create or oo.EventEmitter.emit - 10https://bugzilla.wikimedia.org/72933#c2 (10Roan Kattouw) EventEmitter.emit also seems to be because of arguments. I checked, though, and there's very little "self" time spent in EventEmitter.emit and Factory.create: 25ms and... [19:41:13] TrevorParscal, I'm looking at GroupElement#removeItem [19:41:36] where is itemEvent set? [19:46:32] mooeypoo: always use offset in this sort of case [19:46:41] otherwise you will just get into really confusing situations [19:46:50] edsanders: looking [19:47:17] TrevorParscal, yeah, I'm learning that. I was under the mistaken impression that since I have pageX and pageY (which are global page locations) I should also get the "global" position of the element. [19:47:24] I think there's a foreach loop in #aggregate, but we used a regular for loop in the other functions [19:47:28] Apparently not. [19:48:21] (03PS1) 10Esanders: Use OO.hasOwnProperty where appropriate [oojs/ui] - 10https://gerrit.wikimedia.org/r/171320 [19:49:18] Welcome kirsten :) [19:49:41] hi [19:50:16] Hey kirsten. [19:51:06] mooey|bbiab: no, read offset vs positon, offset IS the global coords, position is relative to the nearest positioned element [19:51:12] hi kirsten, welcome [19:51:19] * Krenair waves [19:51:31] (03CR) 10Krinkle: [C: 04-1] "OO.hasOwnProperty ?" [oojs/ui] - 10https://gerrit.wikimedia.org/r/171320 (owner: 10Esanders) [19:51:47] everyone, kirsten is our new technical writer [19:52:20] (03PS4) 10Jforrester: Tag v1.1.2 [oojs/core] - 10https://gerrit.wikimedia.org/r/171304 [19:52:39] she is bravely helping document OOjs UI [19:52:56] (03CR) 10Jforrester: Tag v1.1.2 (031 comment) [oojs/core] - 10https://gerrit.wikimedia.org/r/171304 (owner: 10Jforrester) [19:53:09] hi! [19:53:29] * James_F waves. [19:53:31] hi, everyone. [19:53:54] and thanks, trevor, for the introduction [19:54:27] edsanders: looks like a total oversight [19:54:33] yup [19:54:35] undersight? [19:54:40] hah [19:54:41] non-sight? [19:54:45] I'm guessing that code isn't used anywhere [19:54:47] dude, what was I smoking? [19:55:03] well, mooey was using aggregate the other day [19:55:06] see this is why we need to rewrite everything in java [19:55:16] ToolGroup [19:55:33] edsanders: Hush, you. [19:55:38] so, I guess we aren't really removing items from toolgroups in practice [19:55:40] so, yeah [19:56:30] I guess we aren't hitting that path in production [19:57:10] Java: Solving all your run-time errors with a simple extra thousand classes [19:58:36] edsanders: clearItems has the same mistake [19:58:38] fyi [19:58:42] yes [19:58:46] cool [20:00:00] (03PS1) 10Catrope: Remove nonfunctional $icon/$indicator code from OptionWidget [oojs/ui] - 10https://gerrit.wikimedia.org/r/171322 [20:00:23] (03CR) 10Trevor Parscal: [C: 032] Remove nonfunctional $icon/$indicator code from OptionWidget [oojs/ui] - 10https://gerrit.wikimedia.org/r/171322 (owner: 10Catrope) [20:01:17] (03CR) 10Trevor Parscal: "This variable also seems very generically named, are you quite sure that there's not even a little more specificity that can be added?" [oojs/ui] - 10https://gerrit.wikimedia.org/r/171240 (owner: 10Prtksxna) [20:02:14] (03Merged) 10jenkins-bot: Remove nonfunctional $icon/$indicator code from OptionWidget [oojs/ui] - 10https://gerrit.wikimedia.org/r/171322 (owner: 10Catrope) [20:04:17] 3VisualEditor / 3Editing Tools: VisualEditor: Tools to set/unset text with annotations (tracker) - 10https://bugzilla.wikimedia.org/47780 (10James Forrester) [20:04:19] 3VisualEditor / 3Editing Tools: VisualEditor: Tool to set/unset background colour - 10https://bugzilla.wikimedia.org/73041 (10James Forrester) 3ASSI p:3Unprio s:3enhanc a:3None For completion. [20:04:29] 3VisualEditor / 3Editing Tools: VisualEditor: Tool to set/unset background colour - 10https://bugzilla.wikimedia.org/73041 (10James Forrester) p:5Unprio>3Lowest [20:04:42] 3VisualEditor / 3Editing Tools: VisualEditor: Tool to set/unset text colour - 10https://bugzilla.wikimedia.org/52645#c9 (10James Forrester) (In reply to Elitre from comment #8) > Would we need to file a separate bug for background text color? Background colour isn't really a thing we want to encourage. :-)... [20:05:07] Hey, does anyone have any examples of a template for a single table row that would be VE friendly? (I've got some strange sorting issues, so trying to use templates to make it easier on the end user). [20:05:45] Rosencrantz: I'm not sure I understand the question? [20:05:56] 3OOjs / 3EventEmitter: OOjs: EventEmitter fails when emitting unbound event with name of Object prototype - 10https://bugzilla.wikimedia.org/73034#c8 (10Jon) Doesn't seem to have made it out to mediawiki.org... [20:06:08] (03CR) 10Trevor Parscal: "Lgtm." [oojs/ui] - 10https://gerrit.wikimedia.org/r/171320 (owner: 10Esanders) [20:06:16] Krinkle: ^^^^ [20:06:21] I've got a template that generates a single table row (for sort keys). [20:06:39] Right. [20:06:42] 3OOjs / 3EventEmitter: OOjs: EventEmitter fails when emitting unbound event with name of Object prototype - 10https://bugzilla.wikimedia.org/73034#c9 (10James Forrester) (In reply to Jon from comment #8) > Doesn't seem to have made it out to mediawiki.org... OOjs 1.1.2 isn't built yet, and MW 1.25wmf8 won't... [20:06:51] TrevorParscal: Wat? How does that relate to *OO*.hasOwn*Property* [20:07:03] dude, you are using the same EXACT trick in OOjs [20:07:14] Right now it kinds of displays funkily in VE. [20:07:15] TrevorParscal: No, this is radically different [20:07:20] oo = {}, hasOwn = oo.hasOwnProperty [20:07:25] TrevorParscal: Yes, I know it's the same reference [20:07:26] oo === OO [20:07:44] (03PS11) 10Divec: Refactor SurfaceObserver pollOnceInternal [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/166188 [20:07:51] so why is it when he uses OO.hasOwnProperty that's somehow different? [20:07:59] he's using .call [20:08:00] Rosencrantz: Hmm. We've just landed table editing, so in a week it will probably behave rather differently. [20:08:08] TrevorParscal: The radical difference being that 1) hasOwnProperty is *NOT* an API, 2) I'm using it there as short cut to Object.prototype in a private scope as optimisation to not create a new object [20:08:16] referencing it repeatedly defeats the optimisation [20:08:20] Ah! Didn't know that. (I'm also still on the 1.23 branch) [20:08:31] TrevorParscal: following my lead would be using the same local hasOwn reference. [20:08:35] or creating one in oojs-ui [20:08:40] he's not writing code inside a closure [20:09:05] I don't care, this is is insane. [20:09:14] It's not a huge issue. Just mostly a curiosity question. [20:09:20] hold on, you are suggesting he alias it to use it one time? [20:09:26] No [20:09:27] look at the uses, once per method [20:09:27] Rosencrantz: Aha. Good things to come. Templates inside tables didn't work very well at all until just recently. The REL1_23 branch version doesn't have those improvements, sadly. [20:09:38] what do you suggest? [20:09:41] TrevorParscal: If you need it somewhere, just get it from Object.prototype as we do everywhere else [20:09:51] we do that for all prototype methods [20:09:58] slice, splice, push, bind, hasOwnProperty, all the time [20:10:11] Having a local reference is just an optimisation I did since we use it a few times in EE [20:10:14] that's all. [20:11:15] OK, so to be clear, I also prefer to use Object.prototype.hasOwnProperty because it's clear, but my point is that it appears to be equally evil that you are using this "shortcut" [20:11:19] James_F: Ah ha. thanks! [20:12:19] and when I saw the code you wrote back in Jun 2013, I thought it was a micro-optimization that made the code less clear [20:12:22] so, pick a side [20:12:53] (03CR) 10Catrope: [C: 032] Node method cleanups [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171269 (owner: 10Esanders) [20:13:28] (03CR) 10Catrope: [C: 032] Add isAllowed(Child|Parent)NodeType to ve.Node [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171270 (owner: 10Esanders) [20:13:30] (03CR) 10Krinkle: "OO.hasOwnProperty is not an API and not even declared anywhere. Assigment of that in oojs-core is not a pattern. This is the programming e" [oojs/ui] - 10https://gerrit.wikimedia.org/r/171320 (owner: 10Esanders) [20:13:39] TrevorParscal: clarified^ [20:13:41] 3VisualEditor / 3Mobile: VisualEditor Mobile: [Regression pre-wmf7] "Uncaught Error: Unknown dependency: rangefix" appears while trying to load mobile VE - 10https://bugzilla.wikimedia.org/73038#c4 (10Rummana Yasmeen) Verified the fix in Betalabs [20:13:45] edsanders: Dude thanks for cleaning up that ridiculous ve.Node vs ve.dm.Node crap, that had been bothering me recently [20:14:06] TrevorParscal: I'm happy to elaborate or change oojs-core if you think it's inconsistent. [20:14:14] well I went to add a new method and it was just all over the place [20:14:31] But whatever the case, OO.hasOwnProperty doesn't make sense. If you're accessing globals, might as well just use Object.hasOwnProperty [20:14:35] which is 20x less confusing. [20:14:50] I was rereading the implementations of getRange() and getOuterRange() while reviewing your table code a few weeks ago and I was very surprised that they were in different places [20:15:35] (03Merged) 10jenkins-bot: Node method cleanups [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171269 (owner: 10Esanders) [20:15:37] (03Merged) 10jenkins-bot: Add isAllowed(Child|Parent)NodeType to ve.Node [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171270 (owner: 10Esanders) [20:16:29] (03PS1) 10Esanders: Use hasOwnProperty instead of 'key in obj' [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171326 [20:16:50] (03CR) 10Catrope: [C: 032] Replace block drop logic with isAllowedChildNodeType check [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171271 (https://bugzilla.wikimedia.org/72995) (owner: 10Esanders) [20:18:04] (03CR) 10Krinkle: [C: 031] Use hasOwnProperty instead of 'key in obj' [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171326 (owner: 10Esanders) [20:18:28] 3VisualEditor / 3ContentEditable: VisualEditor: Drag-and-drop an inserted image slices a table - 10https://bugzilla.wikimedia.org/72995 (10James Forrester) 5PATC>3RESO/FIX p:5Unprio>3High a:3Ed Sanders [20:18:42] (03Merged) 10jenkins-bot: Replace block drop logic with isAllowedChildNodeType check [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171271 (https://bugzilla.wikimedia.org/72995) (owner: 10Esanders) [20:18:51] (03PS1) 10Esanders: Use hasOwnProperty instead of 'key in obj' [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171328 [20:19:52] (03CR) 10Krinkle: [C: 031] "LGTM. Will improve support from breaking on any inherited property key (e.g. "watch", "valueOf", etc.) to only breaking if the key is name" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171328 (owner: 10Esanders) [20:20:00] (03CR) 10Krinkle: "LGTM. Will improve support from breaking on any inherited property key (e.g. "watch", "valueOf", etc.) to only breaking if the key is name" [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171326 (owner: 10Esanders) [20:20:29] Krinkle, does OO-core provide some sort of polyfill to hasOwnProperty? [20:21:02] edsanders: What do you mean? It's available in all engines. [20:21:11] why does it exist then> [20:21:25] there is no polyfill. [20:21:28] 'hasOwn = oo.hasOwnProperty' [20:21:33] oo = {} [20:21:35] one line up [20:21:39] why not just hasOwn = Object.prototype.hasOwnProperty [20:21:47] global object lookup, property lookup [20:21:50] minor optimisation, that's all [20:22:35] oh [20:22:37] that's confusing [20:23:00] edsanders: The reason we do hasOwn = foo; is so we can then do hasOwn.call( object, key ); , which is better than object.hasOwnProperty( key ); because the latter breaks if you did object['hasOwnProperty'] = 'ed'; at any point [20:23:07] whereas the former is resistant against that [20:23:21] I know [20:23:29] Using OO as a global as you did in that oojs-ui commit wouldn't be the same, since that's just another global. In that case you can also use Object.hasOwnProperty (without .prototype in between, since Object, like OO inherits from prototype) [20:23:33] my point is that we should do away with this minor optimization because it's very confusing [20:23:59] Confusing or just something you're learning now? [20:24:01] when you read oojs code, and sett oo.hasOwnProperty, at a glance it appears to be something it's not [20:24:14] dude, it didn't trick me, I'm saying I don't like how unclear it is [20:24:29] I know *you* knew it. [20:24:31] and if Ed mistook it, then I think it's overly clever [20:25:15] Krinkle: we agree that OO.hasOwnProperty is not a good idea [20:25:34] it's simple, we either have a global lookup with inheritance, global lookup with property, or local lookup with no inheritance or property [20:25:53] e.g. Object.hasOwnProperty, Object.prototype.hasOwnProperty, or {}.hasOwnProperty [20:26:06] where you'd only do the latter if there is one already. [20:26:51] This is quite common in libraries as small optimisation in rarely used code (e.g. local references to native helpers set once for an entire code base, like I did in the intro of oojs) [20:27:44] https://github.com/wikimedia/mediawiki/blob/master/resources/lib/jquery/jquery.js#L46-L61 [20:27:53] I think that your use of oo.hasOwnProperty has been demonstrated to be confusing, so either comment what you are doing to avoid confusion, or use something clearer [20:28:18] I can add to comment to oojs-core if that helps [20:28:20] and no offsense to jquery devs, but jquery is not known for being easy to understand code [20:28:28] in fact, it's quite dense and obscure [20:29:04] and, maybe they can defend that, but I would set the balance a little farther toward clarity [20:30:14] Krinkle: thank you for adding a comment [20:30:24] and hashing this out with me [20:30:51] I appreciate your push for performance, and your willingness to compromise when needed [20:31:04] a comment will do just fine [20:31:16] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression wmf3] In Firefox, Cursor is appearing in Read mode at the end of the document after saving a page and after discarding the current edit in Firefox - 10https://bugzilla.wikimedia.org/71852#c10 (10Rummana Yasmeen) Verified the fix in Betalabs [20:31:22] I was surprised we now pay attention to v8 optimisation warnings [20:31:33] (03PS39) 10Paladox: WikiEditor: Fix issue with SVG [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/151611 [20:31:36] (03CR) 10jenkins-bot: [V: 04-1] WikiEditor: Fix issue with SVG [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/151611 (owner: 10Paladox) [20:31:36] Not sure that's a good use of our time at this point. There's way bigger fish to catch. [20:31:43] (03PS40) 10Paladox: WikiEditor: Fix issue with SVG [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/151611 (https://bugzilla.wikimedia.org/35342) [20:31:46] (03CR) 10jenkins-bot: [V: 04-1] WikiEditor: Fix issue with SVG [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/151611 (https://bugzilla.wikimedia.org/35342) (owner: 10Paladox) [20:32:36] Krinkle: I'm paying a small amount of attention to them to the extent that they affect hot code paths [20:32:49] But the only combination of those two that I've found so far is oo.copy [20:32:56] ok [20:33:52] Krinkle: I agree, highly optimized code often is tricky to read and write and wastes a lot of engineer hours to save computer milliseconds [20:33:56] :P [20:34:21] (03PS2) 10Esanders: Use OO.hasOwnProperty where appropriate [oojs/ui] - 10https://gerrit.wikimedia.org/r/171320 [20:34:22] TrevorParscal: To clarify, I wouldn't normally make a minor optimisation like that. Not inside a regular method or "interesting" part of the code. But it seemed sensible to do in the bootstrapping for a library as a whole when that reference is used a fair bit. [20:36:05] i'm just giving you a hard time [20:36:06] http://www.theoreticallogic.com/2011/07/computation-vs-engineering-efficiency/ [20:36:23] edsanders: Hm.. Just curious, why obj.hasOwnProperty in ve and hasOwnProperty.call(obj) in oojs-ui? Both is fine, but wondering if we're more concerned about storing arbitrary keys in one then the other. [20:36:34] Since one covers all cases except hasOwn, and the other covers all cases. [20:36:48] going to fix the VE ones [20:37:13] OK [20:37:27] I think he is just hunting down cases where we should have been using hasOwn and murdering them to death [20:37:31] (03PS3) 10Krinkle: Use Object.prototype.hasOwnProperty where appropriate [oojs/ui] - 10https://gerrit.wikimedia.org/r/171320 (owner: 10Esanders) [20:37:53] (03CR) 10Krinkle: [C: 031] Use Object.prototype.hasOwnProperty where appropriate [oojs/ui] - 10https://gerrit.wikimedia.org/r/171320 (owner: 10Esanders) [20:38:03] I think Israel is going to rescend my citizenship, but here it is: Accounting for RTL in position calculations sucks. [20:38:11] 3VisualEditor / 3Mobile: VisualEditor Mobile: [Regression wmf4] Link suggestions are appearing in a broken way inside Link Inspector in mobile - 10https://bugzilla.wikimedia.org/72402#c3 (10Rummana Yasmeen) Created attachment 17044 --> https://bugzilla.wikimedia.org/attachment.cgi?id=17044&action=edit brok... [20:38:12] If only we didn't have to do LTR too... [20:38:16] ..... [20:38:18] ........ [20:38:21] .okay, okay, no. [20:38:25] TrevorParscal: BTW, https://phabricator.wikimedia.org/T857 was the Phabricator issue I mentioned yesterday. [20:38:26] 3VisualEditor / 3Mobile: VisualEditor Mobile: [Regression wmf4] Link suggestions are appearing in a broken way inside Link Inspector in mobile - 10https://bugzilla.wikimedia.org/72402 (10Rummana Yasmeen) 5RESO/FIX>3REOP [20:38:41] 3VisualEditor / 3Mobile: VisualEditor Mobile: [Regression wmf4] Link suggestions are appearing in a broken way inside Link Inspector in mobile - 10https://bugzilla.wikimedia.org/72402 (10Rummana Yasmeen) 5REOP>3ASSI [20:38:42] mooeypoo: * rescind. [20:38:48] James_F, *take away. [20:38:49] :D [20:38:56] but thank you. [20:39:27] James_F: I emailed quim and the response was basically "it's really hard, may be impossible, too time consuming to know for sure, couldn't be fussed" [20:39:46] TrevorParscal: :-( [20:39:52] mooeypoo: Yeah. [20:41:49] aaaaa. Okay, TrevorParscal do you have any idea maybe? In the parent widget, I have 'parentOffset = this.$element.offset().left' which I calculate while I drag. This shouldn't *change*; the specific items I hover over do change (as they should, since I hover over different items) ... and yet, parentOffset *CHANGES* ?? [20:41:53] how is it changing! [20:41:56] the parent widget isn't moving [20:42:02] * mooeypoo tosses popcorn on Javascript [20:42:48] mooeypoo: why are you using the parent offset? [20:43:03] sorry, isn't drag.left and drag.top relative to the screen? [20:43:24] TrevorParscal, at the moment I'm just trying to see all the details so I don't go mad. [20:43:25] always use offsets for everything, then you don't have to translate them by their parent's all the way up the chain (as you do with position) [20:43:41] Plus, I'm trying to account for both LTR and RTL [20:44:09] and screen positioning while mixed with "offset" is a big big blackhole of boo [20:44:49] TrevorParscal, I found my issue, though, the offset is correct now. [20:44:50] Krinkle, we only need to fix the cases where the object can be modified by the user [20:44:58] mooeypoo: glad to hear it [20:45:10] edsanders: Yeah [20:45:25] e.g. registries [20:45:46] but not simple lookups like {td:1, th:2}[style] [20:45:59] edsanders: I suppose in other cases localobject.hasOwnProperty would be sufficient (to still avoid non-standard traps with "watch" which is a reasonable name, as opposed to "hasOwnProperty" which is not reasonable except in user input) [20:46:25] TrevorParscal, my bugging you seems to take the shape of "panic! -> oh, I'm silly -> forget it, found the solution!" more often than not.... Apologies. [20:46:29] yes - if the key search is user input you need hasOwn [20:46:34] Though doing it everywhere would be fine too, might be clearer. Whichever you prefer. [20:49:11] 3VisualEditor / 3Mobile: VisualEditor Mobile: [Regression pre-wmf15] Pressing backspace to delete the link target inside link inspector closes the inspector - 10https://bugzilla.wikimedia.org/68195#c8 (10James Forrester) (In reply to Alex Monk from comment #7) > Can't reproduce... Can reproduce here. [20:51:34] (03PS1) 10Catrope: Remove the loading spinner at the end, not halfway through loading [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171429 [20:52:13] 3VisualEditor / 3Mobile: VisualEditor Mobile: [Regression wmf4] Link suggestions are appearing in a broken way inside Link Inspector in mobile - 10https://bugzilla.wikimedia.org/72402#c4 (10Rummana Yasmeen) In an actual mobile device,it appears broken for a split second, then again gets back to normal after... [20:52:56] (03PS2) 10Esanders: Use hasOwnProperty instead of 'key in obj' [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/171326 [20:55:06] (03PS2) 10Esanders: Use hasOwnProperty instead of 'key in obj' [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171328 [20:58:33] mooey|food: if me hashing things out with you has a positive effect, then I'm happy [20:58:52] I don't care if it's because you realized you made a mistake or I solved your problem for you [20:59:06] just want to help you get past your road blocks [20:59:11] 3VisualEditor / 3Editing Tools: VisualEditor: Tool to set/unset text as highlighted () - 10https://bugzilla.wikimedia.org/64376#c3 (10Elitre) Will this only allow marking the text in yellow? [20:59:11] brb - food [20:59:34] thanks, TrevorParscal :) Just hope I'm not bugging you too much with my occasional frustrated cries for help [20:59:36] (03CR) 10SuchetaG: [C: 032] Remove the loading spinner at the end, not halfway through loading [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171429 (owner: 10Catrope) [21:00:54] (03Merged) 10jenkins-bot: Remove the loading spinner at the end, not halfway through loading [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171429 (owner: 10Catrope) [21:01:22] Thanks sucheta [21:17:55] RoanKattouw: 72948 Citations do not work in wmf7 - Basic and Re-use are fine though [21:18:00] no js errors [21:21:39] OK I'll check in a minute [21:21:43] I'm in a meeting now [21:21:59] ok [21:22:50] OK confirmed locally; probably due to Ed's tool->command refactor [21:24:56] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf7] Page Settings dialog box - an opacity issue - 10https://bugzilla.wikimedia.org/72838#c3 (10etonkovidova) Verified the fix in betalabs and test2. [21:48:42] (03PS12) 10Divec: Refactor SurfaceObserver pollOnceInternal [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/166188 [21:48:50] (03PS1) 10Catrope: Followup 31bafa1: forgot to rename 'data' to 'args' for citation commands [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171447 (https://bugzilla.wikimedia.org/73001) [21:52:52] edsanders: https://gerrit.wikimedia.org/r/171447 [21:53:27] 3VisualEditor / 3ContentEditable: VisualEditor: Tables - using keyboard to move cursor throws "TypeError: surfaceView.getNativeRangeBoundingClientRect is not a function" - 10https://bugzilla.wikimedia.org/72978#c4 (10etonkovidova) Verified the fix in betalbs and test2. [21:55:02] (03CR) 10Esanders: [C: 032] Followup 31bafa1: forgot to rename 'data' to 'args' for citation commands [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171447 (https://bugzilla.wikimedia.org/73001) (owner: 10Catrope) [21:56:23] (03Merged) 10jenkins-bot: Followup 31bafa1: forgot to rename 'data' to 'args' for citation commands [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171447 (https://bugzilla.wikimedia.org/73001) (owner: 10Catrope) [21:56:57] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf7] Cite command not working (but cite-tranclusion does?) - 10https://bugzilla.wikimedia.org/73001 (10James Forrester) 5PATC>3RESO/FIX a:5Ed Sanders>3Roan Kattouw [21:56:58] (03PS1) 10Jforrester: Followup 31bafa1: forgot to rename 'data' to 'args' for citation commands [extensions/VisualEditor] (wmf/1.25wmf7) - 10https://gerrit.wikimedia.org/r/171451 (https://bugzilla.wikimedia.org/73001) [21:58:39] (03CR) 10Trevor Parscal: [C: 032] Use Object.prototype.hasOwnProperty where appropriate [oojs/ui] - 10https://gerrit.wikimedia.org/r/171320 (owner: 10Esanders) [22:01:05] (03CR) 10Esanders: Refactor SurfaceObserver pollOnceInternal (031 comment) [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/166188 (owner: 10Divec) [22:01:51] (03Merged) 10jenkins-bot: Use Object.prototype.hasOwnProperty where appropriate [oojs/ui] - 10https://gerrit.wikimedia.org/r/171320 (owner: 10Esanders) [22:07:32] MatmaRex: I'm going to ask Chris for a security review of OOjs UI's PHP code [22:07:53] That way we can get it added to MediaWiki core [22:08:01] if there's anything we need to tweak before then, let me know [22:08:09] I have a meeting right now, be back in about 20 [22:09:38] TrevorParscal: nothing i'm aware of, apart from outstanding feature requests [22:22:43] 3VisualEditor / 3Editing Tools: VisualEditor: Matching pages list in link dialog should support right-click / ctrl+click to open page in a new window/tab - 10https://bugzilla.wikimedia.org/51205 (10Roan Kattouw) [22:22:43] 3VisualEditor / 3ContentEditable: VisualEditor: Internal links should display the page's URL in the status bar on the browsers that do that - 10https://bugzilla.wikimedia.org/50646#c3 (10Roan Kattouw) 5ASSI>3NEW a:5Sucheta Ghoshal>3None Removing See Also and unassigning; I thought this was about link... [22:26:56] 3VisualEditor / 3ContentEditable: VisualEditor: Internal links should display the page's URL in the status bar on the browsers that do that - 10https://bugzilla.wikimedia.org/50646#c4 (10Roan Kattouw) As for the issue of browsers not displaying the link target in the status bar when you hover them: that's a... [22:28:26] 3VisualEditor / 3Editing Tools: VisualEditor: Matching pages list in link dialog should support right-click / ctrl+click to open page in a new window/tab - 10https://bugzilla.wikimedia.org/51205#c4 (10Roan Kattouw) (In reply to Roan Kattouw from comment #3) > Both this and bug 50646 Forget about bug 50646, m... [22:28:41] 3VisualEditor / 3Editing Tools: VisualEditor: Parameter add icon remains enabled in empty transclusion dialog box - 10https://bugzilla.wikimedia.org/63158#c4 (10Rummana Yasmeen) 5RESO/?>3VERI Verified the fix in production [22:30:41] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf6] "TypeError: Cannot read property 'isDefault' of undefined" appears when Media Settings dialog appears while adding a new image - 10https://bugzilla.wikimedia.org/72492#c4 (10Rummana Yasmeen) 5RESO/?>3VERI Verified the fix in production [22:30:56] 3VisualEditor / 3ContentEditable: VisualEditor: Internal links should display the page's URL in the status bar on the browsers that do that - 10https://bugzilla.wikimedia.org/50646#c5 (10James Forrester) 5NEW>3RESO/WOR Re-marking as WORKSFORME. [22:32:42] 3VisualEditor / 3Editing Tools: VisualEditor: Tool to set/unset text as highlighted () - 10https://bugzilla.wikimedia.org/64376#c4 (10James Forrester) (In reply to Elitre from comment #3) > Will this only allow marking the text in yellow? It will mark the text with . It's up to users' browsers t... [22:33:34] (03CR) 10Esanders: "No, the test just asserts something that isn't true anymore (tables getting exploded)." [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171288 (https://bugzilla.wikimedia.org/72990) (owner: 10Esanders) [22:34:27] 3VisualEditor / 3Editing Tools: VisualEditor: Tables drop their class when pasted in IE11 - 10https://bugzilla.wikimedia.org/72599#c3 (10etonkovidova) Verified the fix in test2 - IE11, FF, and Chrome. [22:34:32] (03PS2) 10Esanders: Allow headings in table cells [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171288 (https://bugzilla.wikimedia.org/72990) [22:35:46] RoanKattouw: https://gerrit.wikimedia.org/r/171460 [22:38:26] 3VisualEditor / 3Editing Tools: VisualEditor: Tool to set/unset text as highlighted () - 10https://bugzilla.wikimedia.org/64376#c5 (10Elitre) I'm unhappy. I demand the ability to pick my favorite color when highlighting. [22:40:08] (03PS2) 10Cmcmahon: QA: sort remaining page elements that belong in the iframe [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171295 [22:42:12] (03CR) 10Cmcmahon: [C: 032] "maintenance, interim step" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171295 (owner: 10Cmcmahon) [22:42:41] 3VisualEditor / 3Editing Tools: VisualEditor: Tool to set/unset text as highlighted () - 10https://bugzilla.wikimedia.org/64376#c6 (10James Forrester) (In reply to Elitre from comment #5) > I'm unhappy. I demand the ability to pick my favorite color when > highlighting. Talk to W3C. ;-) [22:43:39] (03Merged) 10jenkins-bot: QA: sort remaining page elements that belong in the iframe [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171295 (owner: 10Cmcmahon) [22:47:11] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression wmf3] In Firefox, Cursor is appearing in Read mode at the end of the document after saving a page and after discarding the current edit in Firefox - 10https://bugzilla.wikimedia.org/71852#c11 (10Rummana Yasmeen) Verified the fix in test2 [22:48:16] (03PS1) 10Alex Monk: Show category items as red if they don't have description pages [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171464 (https://bugzilla.wikimedia.org/65517) [22:49:26] 3VisualEditor / 3Editing Tools: VisualEditor: Tool to set/unset text as highlighted () - 10https://bugzilla.wikimedia.org/64376#c7 (10Elitre) No! Google Docs allows highlighting text in the color I want. Please make that happen somehow? [22:49:42] 3VisualEditor / 3ContentEditable: VisualEditor: Please let me drop the cursor into the table caption if it exists. Please? - 10https://bugzilla.wikimedia.org/72724#c2 (10etonkovidova) Verified the fix in betalabs and test2(FF, Safari, Chrome). [22:49:42] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf7] The down arrow to bring up the context menu for Categories is not center-aligned, again - 10https://bugzilla.wikimedia.org/72962#c5 (10Rummana Yasmeen) Verified the fix in test2 [22:51:26] 3VisualEditor / 3Mobile: VisualEditor Mobile: [Regression pre-wmf7] "Uncaught Error: Unknown dependency: rangefix" appears while trying to load mobile VE - 10https://bugzilla.wikimedia.org/73038#c5 (10Rummana Yasmeen) Verified the fix in test2 [22:52:56] 3VisualEditor / 3Editing Tools: VisualEditor: Regular context menu trying to render on TableSelections - 10https://bugzilla.wikimedia.org/72802#c8 (10Rummana Yasmeen) Verified the fix in test2 [22:56:26] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf7] Link Suggestions are not appearing on placing the cursor inside it - 10https://bugzilla.wikimedia.org/72959#c4 (10Rummana Yasmeen) Verified the fix in test2 [22:57:41] 3VisualEditor / 3ContentEditable: VisualEditor: Overwriting the text inside a cell does not retain the text format applied to that cell previously - 10https://bugzilla.wikimedia.org/72831#c4 (10Rummana Yasmeen) Verified the fix in test2 [22:58:56] 3VisualEditor / 3Editing Tools: VisualEditor: When the cursor is active inside a content cell , cannot insert any additional row or column to the table (or should we disable the row/column operator while user is inside a content cell?) - 10https://bugzilla.wikimedia.org/72803#c4 (10Rummana Yasmeen) Verified... [23:01:11] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf7] Empty Add Template dialog appearing upon trying to open an existing template from a page - 10https://bugzilla.wikimedia.org/72719#c3 (10Rummana Yasmeen) Verified the fix in test2 [23:02:27] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf7] VisualEditor gets unresponsive after adding a template - 10https://bugzilla.wikimedia.org/72961#c4 (10Rummana Yasmeen) Verified the fix in test2 [23:03:29] 3VisualEditor / 3ContentEditable: VisualEditor: Tables Caption - moving a cursor with keyboard forces a cursor to the begininng of the page - 10https://bugzilla.wikimedia.org/73055 (10etonkovidova) 3NEW p:3Unprio s:3normal a:3None 1. Open a page in VE and insert a table with a caption. Make sure that... [23:08:55] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf7] The options under Page Settings and Advances Settings of Page Options dialog are appearing very cramped - 10https://bugzilla.wikimedia.org/72792#c5 (10Rummana Yasmeen) Verified the fix in test2 [23:10:11] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf7] The group names inside "Use this group" field are appearing in a short form like "g..." for both "References list" and "Basic Reference" dialog when you open these dialogs first time in a session - 10https://bugzilla.wikimedia.org/72783#c10 (... [23:14:26] 3VisualEditor / 3Mobile: VisualEditor Mobile: The label for "Re-use" tool is not greyed out when the tool is disabled (icon is greyed out) - 10https://bugzilla.wikimedia.org/72405#c6 (10Rummana Yasmeen) Verified the fix in test2 [23:20:37] James_F|Away: https://gerrit.wikimedia.org/r/#/c/171464 [23:21:00] (03CR) 10Catrope: [C: 031] "Code looks fine, James has some product input" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171464 (https://bugzilla.wikimedia.org/65517) (owner: 10Alex Monk) [23:22:04] (03CR) 10Jforrester: "Add a line at the top of the category thing saying "This category doesn't have a description page yet." or something, like we do for redir" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171464 (https://bugzilla.wikimedia.org/65517) (owner: 10Alex Monk) [23:22:12] 3VisualEditor / 3Editing Tools: VisualEditor: TitleInputWidget should validate inputs - 10https://bugzilla.wikimedia.org/71249#c8 (10etonkovidova) test2 has the same validation that is in betalabs: - cannot add Talk: as a template - 'Add template' is disabled - cannot add a template or a category with |(a... [23:22:56] 3VisualEditor / 3Editing Tools: VisualEditor: TitleInputWidget should validate inputs - 10https://bugzilla.wikimedia.org/71249#c9 (10James Forrester) (In reply to etonkovidova from comment #8) > test2 has the same validation that is in betalabs: > > - cannot add Talk: as a template - 'Add template' is disa... [23:24:01] Jforrester: happy to hear that! [23:25:49] etonkovidova: :-) [23:26:11] 3VisualEditor / 3Editing Tools: VisualEditor: Remove transclusion node if we were just going to set it to empty content - 10https://bugzilla.wikimedia.org/63341#c7 (10Rummana Yasmeen) Verified the fix in test2 [23:29:23] edsanders|away: did you see my comments on PS5? https://gerrit.wikimedia.org/r/#/c/170907 [23:32:12] 3VisualEditor / 3Editing Tools: VisualEditor: Be able to add or remove tables, table rows and table columns - 10https://bugzilla.wikimedia.org/52181#c8 (10etonkovidova) post-deploy testing in test2 - VisualEditor has added functiona lity to insert/remove tables and modify tables - insert columns and rows in... [23:34:36] (03PS1) 10Alex Monk: Disable review button while save dialog is pending [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171474 (https://bugzilla.wikimedia.org/65310) [23:35:56] (03CR) 10jenkins-bot: [V: 04-1] Disable review button while save dialog is pending [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171474 (https://bugzilla.wikimedia.org/65310) (owner: 10Alex Monk) [23:36:41] (03PS2) 10Alex Monk: Disable review button while save dialog is pending [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171474 (https://bugzilla.wikimedia.org/65310) [23:42:19] RoanKattouw, James_F: Sigh... So for adding that "This category doesn't have a description page yet." text, I'd have to check item.$label.hasClass( 'new' ) in MWCategoryPopupWidget#loadCategoryIntoPopup? [23:42:57] although I could pass that data through from whatever creates the CategoryItemWidget, like isHidden [23:45:03] Krenair: Yeah, maybe? [23:46:33] (03PS2) 10Alex Monk: Show category items as red if they don't have description pages [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171464 (https://bugzilla.wikimedia.org/65517) [23:46:35] (03PS2) 10Prtksxna: mediawiki: Replace instances of 0.3em border-radius with variable [oojs/ui] - 10https://gerrit.wikimedia.org/r/171240 [23:47:16] (03CR) 10Prtksxna: "> These are not all instances, there's stuff like "border-bottom-left-radius: 0.3em" too. Grep for "radius: 0.3em"?" [oojs/ui] - 10https://gerrit.wikimedia.org/r/171240 (owner: 10Prtksxna) [23:50:44] cscott: by the way, I'm getting OOjs UI's PHP stuff into core ASAP [23:50:49] it's ready, just needs security review [23:51:08] (03CR) 10Jforrester: [C: 032] Allow headings in table cells [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171288 (https://bugzilla.wikimedia.org/72990) (owner: 10Esanders) [23:51:12] shooting for having it in production next Wednesday [23:51:39] MatmaRex: by the way, we need to split the styles and the scripts for OOjs UI into 2 separate modules, the scripts depending on the styles [23:51:47] 3VisualEditor / 3ContentEditable: VisualEditor: Tables - applying headings format to table cells will split the table - 10https://bugzilla.wikimedia.org/72990 (10James Forrester) 5PATC>3RESO/FIX p:5Unprio>3Low a:3Ed Sanders [23:52:02] (03CR) 10Krinkle: "Where is this variable defined?" [oojs/ui] - 10https://gerrit.wikimedia.org/r/171240 (owner: 10Prtksxna) [23:52:25] (03Merged) 10jenkins-bot: Allow headings in table cells [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171288 (https://bugzilla.wikimedia.org/72990) (owner: 10Esanders) [23:52:28] MatmaRex: that way PHP usage can add the styles only, and statically [23:52:35] yeah [23:52:45] but no, we don't need to split anything [23:52:59] we don't have to split the styles up [23:53:00] we can just do $out->addModuleStyles('oojs-ui') right now [23:53:15] if we split them up, we won't be able to handle dependencies correctly anyway [23:53:19] right, well we don't want to ship the styles AGAIN when you add the scripts [23:53:25] 3VisualEditor / 3Editing Tools: VisualEditor: Tool to set/unset text as highlighted () - 10https://bugzilla.wikimedia.org/64376#c8 (10James Forrester) (In reply to Elitre from comment #7) > No! Google Docs allows highlighting text in the color I want. Please make > that happen somehow? That's not high... [23:53:34] i realize this is a little hand-wavy right now [23:53:35] sure, but we can't express that using RL dependencies [23:53:44] (03CR) 10Prtksxna: "> Where is this variable defined?" [oojs/ui] - 10https://gerrit.wikimedia.org/r/171240 (owner: 10Prtksxna) [23:53:47] but perhaps you could help me solve this problem [23:54:08] we need the styles to be usable without JS (since the PHP version is designed to be useful without JS) [23:54:09] we can either possibly load twice, or assumie that we already have styles loaded (we can't check that) and load scripts only [23:54:30] how big is the payload we are adding if we always load? [23:54:33] pretty big right? [23:54:40] esp. because of embedded icons [23:54:55] no idea [23:55:31] yeah... I'll ask Roan for his ideas [23:56:37] unless we suddenly decide to implement support for JS-independent CSS-only modules in RL, which was a big WONTFIX for years, this will be messy no matter how we do it [23:57:02] if we want to avoid loading anything twice, then we indeed need two modules, one CSS-only and one with everything else [23:57:26] 3VisualEditor / 3MediaWiki integration: VisualEditor: Invocation of "lastmodifiedat" message is not updated when page is saved - 10https://bugzilla.wikimedia.org/52403#c7 (10etonkovidova) Verified in test2 and betalabs. The timestamp for "This page was last modified on 5 November 2014, at 23:54." seems to re... [23:57:47] they must not depend on each other, CSS one must always be loaded using addModuleStyles(), and the other one must always be loaded using addModules(), but only after the CSS one was loaded [23:58:46] it just gets messy for users who need to do ->addModuleStyles('oojs-ui-styles'); ->addModules('oojs-ui-scripts')