[00:00:01] RoanKattouw: https://gerrit.wikimedia.org/r/174016 [00:00:18] (03PS13) 10Trevor Parscal: Eventify TemplateDataGenerator and use oojs-ui [extensions/TemplateData] - 10https://gerrit.wikimedia.org/r/167046 (owner: 10Mooeypoo) [00:00:24] mooeypoo: see that ^^^ [00:00:28] I fixed a few things [00:00:38] (03CR) 10Catrope: [C: 032 V: 032] Update VE core for cherry-picks (7819523) [extensions/VisualEditor] (wmf/1.25wmf7) - 10https://gerrit.wikimedia.org/r/174016 (owner: 10Jforrester) [00:01:03] mooeypoo: https://gerrit.wikimedia.org/r/#/c/167046/12..13/modules/ext.templateDataGenerator.ui.js [00:01:15] oo [00:01:22] and https://gerrit.wikimedia.org/r/#/c/167046/12..13/modules/ext.templateDataGenerator.ui.tdDialog.js [00:01:37] yes! I actually fixed that with Roan in the *other* fix (smart me :\ ) [00:01:38] thanks! [00:01:38] so, first off, append window managers to the body [00:01:49] The icons were tiny [00:01:54] the text, I mean [00:01:56] * mooeypoo nods [00:02:24] Yeah I told her to not put the window manager where it was [00:03:02] IIRC [00:03:05] I remember something like that [00:03:17] RoanKattouw, you did, and I accidentally fixed it in the *dependent* commit [00:03:24] The one where I added the drag/drop widget. Meh. [00:04:13] TrevorParscal, I had an idea but I'm not sure if it fits your design schema; could we have a sort of panel above where the buttons are (at the bottom) for kind of "status" stuff or dialog messages? [00:04:27] So, it could show things like "current language: blah" [00:04:40] or in the case of the image dialog, "current image name: Foo.jpg" [00:04:43] stuff like that. [00:06:08] I'm not sure how to properly think about the design of the TemplateData dialog with the languages enabled. Krinkle and I brainstormed ,but we did on the previous style of the dialog which had the entire list of parameters as one big table. [00:08:59] Krinkle: BTW, four new JSCS options have been added. Do we want to consider them for the Wikimedia preset? [00:10:00] (03CR) 10Catrope: [C: 04-1] "-1 per Ed" [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/173296 (owner: 10Jforrester) [00:11:39] mooey, you can put anything you want in a Dialog's $foot [00:11:53] and it will automatically be visible, and should look nice [00:12:08] (03PS1) 10Jforrester: build: Update variosu devDependencies to latest [oojs/core] - 10https://gerrit.wikimedia.org/r/174022 [00:12:20] the footer buttons are added to the window by just appending, the sizing stuff is done during setup [00:12:52] (03PS2) 10Jforrester: build: Update various devDependencies to latest [oojs/core] - 10https://gerrit.wikimedia.org/r/174022 [00:13:12] mooeypoo: also, I switched your button select widget to a dropdown [00:13:25] note: that means the dropdown has to be duck typed atm [00:13:38] I did this to avoid assuming all selects would be dropdown widgets [00:13:46] see code changes for details [00:14:48] I agree that the UI could be /better/ but it's already getting pretty nice [00:15:36] There's a drop down? [00:15:45] mooeypoo: For type. [00:15:56] yeah, you were using a ButtonSelectWidget [00:15:56] TrevorParscal, my biggest worry is the language thing. I think it's extremely confusing [00:16:00] It used to be called InlineMenuWidget [00:16:03] Now it's called DropdownWidget [00:16:08] 3VisualEditor / 3ContentEditable: VisualEditor: discrepancy in displaying text with in Edit and Read modes - 10https://bugzilla.wikimedia.org/73536 (10etonkovidova) 3NEW p:3Unprio s:3normal a:3None Created attachment 17154 --> https://bugzilla.wikimedia.org/attachment.cgi?id=17154&action=e... [00:16:10] Ohh. I missed that. Neat! Thanks TrevorParscal [00:16:51] TrevorParscal, you see how you can change language, and there you can add your own (hopefully something that makes sense, like 'de' or 'he' etc) and then also choose which language you're dealing with? --> isn't that whole process very confusing? That's my main worry. I wasn't sure how to do it properly. [00:16:51] 3VisualEditor / 3ContentEditable: VisualEditor: discrepancy in displaying text with in Edit and Read modes - 10https://bugzilla.wikimedia.org/73536#c1 (10etonkovidova) Created attachment 17155 --> https://bugzilla.wikimedia.org/attachment.cgi?id=17155&action=edit Edit mode - only the first line is... [00:16:58] ok, so you should have the language be a dropdown, with an add button next to it which allows adding a language by language code, and a remove button? [00:17:18] oooh wait so then it can be in the main panel instead of its own panel? [00:17:22] or leave it on its own? [00:17:32] basically, I think that changing, adding and removing languages should be done in-place if possible [00:17:38] * mooeypoo nods [00:17:54] if you make it have it's own panel [00:18:08] I added the panel because otherwise I was worried that if you have a bunch of languages it may get cluttered. [00:18:21] you could perhaps show the change language button next to the label specifying which language is current [00:18:25] but I'm not sure how many times we have *many* languages. Krinkle might know. This is mostly for templates that are shared (in commons i think?) [00:18:40] how do you remove a language? [00:18:47] Oh [00:18:49] ha [00:18:51] uh [00:18:54] :-) [00:18:55] ... you... don't... [00:18:58] lol [00:19:06] what if you make a mistake and add the wrong code? [00:19:14] hehe well technically if you delete the info of the language it'll delete the language [00:19:15] but yes. [00:19:38] mooeypoo: So you can't have 'de' show a blank label for something, but 'de-formal' show one? [00:19:41] maybe removing a language with existing labels should show the number of pieces of information being deleted in the confirm dialog [00:19:44] TrevorParscal, oh, so yes, we can't really control for that :\ I am thinking of using CLDR but that is external and I am not sure we can trust it being in all wikis [00:20:22] what if you typed a valid but incorrect code? [00:20:44] es instead of en, both valid, but maybe you wanted es not en [00:20:56] TrevorParscal, not sure how to fix for that, though? [00:21:04] TrevorParscal, same question for the language input widget, too [00:21:07] (03PS1) 10Jforrester: build: Update various devDependencies to latest [oojs/ui] - 10https://gerrit.wikimedia.org/r/174027 [00:21:28] so, you need it support select, add and remove [00:21:32] we kinda have to give the user a blank input -- unless we want to use the language data. [00:21:51] which we can do but it's external. Wait, i think ed remodeled it for VE's language widget. I'll look at that. [00:21:55] * mooeypoo nods [00:22:11] so, I feel also that you may want to change the language separately from selecting a field [00:22:20] James_F: Meh, I forgot we needed to put https://gerrit.wikimedia.org/r/#/c/173977/ in the SWAT too. I'll do the legwork and put it in tomorrow's SWAT [00:22:41] RoanKattouw: Oops. Thanks! [00:22:47] imagine you are adding a field, then you want to provide labels for the other languages you know - would be nice if we could switch the language at any point [00:22:53] mooeypoo: TrevorParscal: context? Not sure how far back to read. [00:22:59] OOUI cherry-picks are always laborious [00:23:32] so, my suggestion here is to make language choice separate from adding/removing [00:23:38] TrevorParscal, it would be awesome, yes. that's why I thought of the little bottom-panel thing for the dialog [00:23:54] yeah, so you could use the right side of the foot [00:23:59] you don't need another panel [00:24:01] The "using this language" appears in all panels, with the select in all panels [00:24:01] you have the foot [00:24:06] 3VisualEditor / 3MediaWiki integration: VisualEditor: It's not possible to "Apply changes" from the transclusion dialog after it has been emptied - 10https://bugzilla.wikimedia.org/72191#c9 (10etonkovidova) 5RESO/?>3VERI Verified the fix in production. [00:24:16] TrevorParscal, can I add non-buttons to the footer? [00:24:20] Using English [change] [00:24:26] oh [00:24:30] change takes you to the panel [00:24:49] (03CR) 10Esanders: [C: 04-1] Eventify TemplateDataGenerator and use oojs-ui (034 comments) [extensions/TemplateData] - 10https://gerrit.wikimedia.org/r/167046 (owner: 10Mooeypoo) [00:25:01] probably use a select widget (like you do with params) [00:25:23] don't know about adding and removing yet... [00:25:30] Yeah I used the select widget with button option widgets [00:25:35] kaity: you 'round? [00:25:38] hm. [00:25:47] TrevorParscal: yep! [00:26:06] do you have any cycles to work with Moriel on the TemplateData user interface? [00:26:23] TrevorParscal: sure [00:26:31] TrevorParscal: what is it? :) [00:26:36] I can show you where we are at now (if you come by my desk) and then perhaps you could provide some basic mocks for a couple of workflows [00:26:38] Krinkle, we're talking about the template data dialog design, especially how we can create the UX for the language change better [00:26:42] maybe come by now-ish? [00:26:50] mooeypoo: TrevorParscal: I'm confused, wheres those dozen or so lines in answer to my question? [00:27:18] OK, I've missed to much. Let me know if you want my input, I'll need a little more context as I'm head-down into something else atm [00:27:21] Krinkle, ^^ sorry, I was distracted, answered right above. [00:27:42] Krinkle, basically, we're going over the new ooui-fied TemplateData edit dialog. [00:27:43] maybe by e-mail thread or next time after a standup [00:28:06] The discussion is about the design -- and specifically about how to work out the language issue. How there are fields that allow multiple languages, so how we can provide a good UX for them. [00:28:09] Krinkle: yeah, no worries [00:28:18] Krinkle: That would require you to attend one. :-) [00:28:26] oooo [00:28:30] burn! [00:28:36] haha [00:29:48] lol - dude, Krinkle is a free man, don't try and pin him down with your imperialist soul-crushing meeting schedules! [00:30:14] * James_F hugs Krinkle for being awesome. [00:30:22] If, you know, infrequently around. :-) [00:30:26] James_F: I got a house now, and no travels for hackathons any time soon. [00:30:38] Should be able to raise standup frequency now [00:30:49] :) [00:30:53] Krinkle: That would be nice. :-) [00:31:01] TrevorParscal, another a little side-question i have about the design -- I have a couple of error/notice messages, one per panel, that I am using to display notices or errors, etc, because having a sort of global one didn't work -- it appeared *behind* the panels, because the panels are absolutely positioned. [00:31:10] Krinkle: Also, ARGH at breaking changes in grunt tasks. [00:31:22] TrevorParscal, so, I thought that the best way is to have a message per panel, but I was wondering if you might have a better idea. [00:31:39] James_F: where? [00:31:48] grunt-css-url-embed 1.1.0 [00:31:58] Krinkle: Am working out a patch and going to submit a fix. [00:32:27] Krinkle: It used to work fine if you ran it on a file with nothing to embed; it now exits with non-zero status. [00:33:04] Ah, I like that. [00:33:18] Krinkle: You like the new behaviour? [00:33:18] It bothered me that it output all of them too. Rather verbose. [00:33:23] TrevorParscal, these are used for anything from "This language already exists" when you add a language that is already in the list, to "This parameter name already exists" which are in different panels -- and down to displaying a notice something like "please verify the name of the parameter doesn't include {{ | ]] " etc. So it *could* be 1 element that displays messages, or (as I do it now) multiple elements per panel. [00:33:47] It should be distinguishable from an actual error ideally, but it's nice to detect that I think, yes. [00:34:08] Krinkle: No, it still spams you. [00:34:16] Krinkle: But it exits as an error. [00:36:50] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf4] All inspector dialog UI is broken when added inside Reference/Media Settings dialog box - 10https://bugzilla.wikimedia.org/64761#c10 (10James Forrester) 5PATC>3RESO/FIX This bug didn't recur, it was a new bug with a totally different cau... [00:37:05] Krinkle: How do you think we should use it, then? [00:38:09] Krinkle: Manually adding files with @embed to the build script seems seriously sub-optimal. [00:40:22] 3VisualEditor / 3ContentEditable: VisualEditor: Discrepancy in displaying text with in Edit and Read modes - 10https://bugzilla.wikimedia.org/73536 (10James Forrester) 5NEW>3ASSI p:5Unprio>3Low s:5normal>3trivia [00:45:03] James_F: You're saying behaviour changed and since this version it fails when one of the input css files has 0 @embed's in it? [00:50:13] Krinkle: Yes. [00:50:23] (03CR) 10Catrope: [C: 04-2] "Blocked until OOUI change is merged. This will also require rebuilding then (also because another OOUI pull-through was merged in the mean" [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/173912 (owner: 10Trevor Parscal) [00:50:42] Krinkle: Current is 1.0.3; in 1.0.4 current behaviour is kept, but as of 1.1.0 there was an unannounced breaking change in this behaviour. [00:51:13] OK we're descending so I'm gonna go [00:51:19] James_F: You good to do verification for the SWAT? [00:51:49] RoanKattouw: Yeah. [00:55:06] 3VisualEditor / 3MediaWiki integration: VisualEditor: Toolbar should update fixed position after resize if MediaWiki layout changed (e.g. Vector skin goes below 982px SD/HD boundary) - 10https://bugzilla.wikimedia.org/68916#c4 (10etonkovidova) 5RESO/?>3VERI Verified the fix in production. [01:23:08] 3VisualEditor / 3ContentEditable: VisualEditor: Safari DOM Exception - should be handled? - 10https://bugzilla.wikimedia.org/73540 (10etonkovidova) 3NEW p:3Unprio s:3normal a:3None After a prolonged session of editing in Safari, the following error was displayed: [Error] Error: QuotaExceededError:... [01:55:22] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf4] All inspector dialog UI is broken when added inside Reference/Media Settings dialog box - 10https://bugzilla.wikimedia.org/64761#c11 (10Rummana Yasmeen) Ah I see, Sorry we didnt know that before the fix, opening a different one and tagging t... [01:55:25] (03PS53) 10Paladox: WikiEditor: Re add SVG [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/151611 (https://bugzilla.wikimedia.org/35342) [01:57:50] 3VisualEditor / 3Editing Tools: VisualEditor:[Regression wmf8] All inspector dialog UI is broken when added inside Reference/Media Settings dialog box - 10https://bugzilla.wikimedia.org/73542#c1 (10Rummana Yasmeen) 5NEW>3RESO/FIX From Ritu Swain: Reopen this issue. In test2 and beta only. steps- 1> op... [01:58:21] 3VisualEditor / 3Editing Tools: VisualEditor:[Regression wmf8] All inspector dialog UI is broken when added inside Reference/Media Settings dialog box - 10https://bugzilla.wikimedia.org/73542#c2 (10Rummana Yasmeen) Fixed by Followup ca6a16f: use parent selector both for hiding and showing windows https://... [01:58:50] 3VisualEditor / 3Editing Tools: VisualEditor:[Regression wmf8] All inspector dialog UI is broken when added inside Reference/Media Settings dialog box - 10https://bugzilla.wikimedia.org/73542#c3 (10Rummana Yasmeen) Verified the fix in Betalabs and test2 [02:09:21] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression pre-wmf4] All inspector dialog UI is broken when added inside Reference/Media Settings dialog box - 10https://bugzilla.wikimedia.org/64761#c12 (10James Forrester) (In reply to Rummana Yasmeen from comment #11) > Ah I see, Sorry we didnt know that befo... [02:09:50] 3VisualEditor / 3Editing Tools: VisualEditor:[Regression wmf8] All inspector dialog UI is broken when added inside Reference/Media Settings dialog box - 10https://bugzilla.wikimedia.org/73542 (10James Forrester) p:5Unprio>3Highes [03:46:08] (03CR) 10Jforrester: Provide a FileDropHandler for CSV and TSV files (031 comment) [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/173296 (owner: 10Jforrester) [03:47:38] (03PS4) 10Jforrester: Provide a FileDropHandler for CSV and TSV files [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/173296 [03:49:34] (03PS1) 10Jforrester: build: Update various devDependencies to latest [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/174061 [03:51:42] (03PS1) 10Jforrester: build: Update grunt-jscs devDependency to latest [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174062 [09:57:52] 3MediaWiki / 3Page editing: Auto-save to increase chances of lost edits recovery - 10https://bugzilla.wikimedia.org/73241#c5 (10Nemo) Merged in ContentTranslation: 00ebcc6e3e31b278092abde755cbd9009ef65cf4 No idea how much of its code/idea is reusable. [10:45:03] (03CR) 10Hashar: "recheck" [oojs/ui] - 10https://gerrit.wikimedia.org/r/172722 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [10:50:03] (03CR) 10Hashar: [C: 032] Fixed RuboCop Style/AndOr offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172722 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [10:50:26] (03CR) 10jenkins-bot: [V: 04-1] Fixed RuboCop Style/AndOr offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172722 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:06:44] (03PS3) 10Hashar: Fixed RuboCop Style/AsciiComments offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172726 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:06:46] (03PS3) 10Hashar: Fixed RuboCop Style/NilComparison offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172739 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:06:48] (03PS2) 10Hashar: Fixed RuboCop Style/TrailingBlankLines offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172741 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:06:50] (03PS3) 10Hashar: Fixed RuboCop Style/AndOr offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172722 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:06:52] (03PS2) 10Hashar: Fixed RuboCop Style/SpaceAfterComma offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172740 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:06:54] (03PS3) 10Hashar: Fixed RuboCop Style/SpaceAroundEqualsInParameterDefault offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172734 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:06:56] (03PS3) 10Hashar: Fixed RuboCop Style/NegatedIf offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172732 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:06:58] (03PS3) 10Hashar: Fixed RuboCop Style/SignalException offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172733 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:07:00] (03PS3) 10Hashar: Fixed RuboCop Style/EmptyLineBetweenDefs offence [oojs/ui] - 10https://gerrit.wikimedia.org/r/172730 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:07:02] (03PS3) 10Hashar: Fixed RuboCop Style/EmptyLines offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172731 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:07:04] (03PS1) 10Hashar: Merge rubocop related changes [oojs/ui] - 10https://gerrit.wikimedia.org/r/174088 [11:09:04] (03CR) 10Hashar: "That merges most of the patch Zeljkof proposed to fix ruby linting with a single merge commit." [oojs/ui] - 10https://gerrit.wikimedia.org/r/174088 (owner: 10Hashar) [11:13:10] (03CR) 10jenkins-bot: [V: 04-1] Fixed RuboCop Style/EmptyLines offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172731 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:13:29] (03CR) 10jenkins-bot: [V: 04-1] Merge rubocop related changes [oojs/ui] - 10https://gerrit.wikimedia.org/r/174088 (owner: 10Hashar) [11:13:40] (03CR) 10jenkins-bot: [V: 04-1] Fixed RuboCop Style/AndOr offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172722 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:14:10] (03CR) 10jenkins-bot: [V: 04-1] Fixed RuboCop Style/AsciiComments offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172726 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:15:45] (03CR) 10Hashar: "recheck" [oojs/ui] - 10https://gerrit.wikimedia.org/r/172722 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:15:51] (03CR) 10Hashar: "recheck" [oojs/ui] - 10https://gerrit.wikimedia.org/r/172731 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:16:43] (03CR) 10Hashar: "recheck" [oojs/ui] - 10https://gerrit.wikimedia.org/r/174088 (owner: 10Hashar) [11:17:05] (03CR) 10Hashar: "recheck" [oojs/ui] - 10https://gerrit.wikimedia.org/r/172726 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:21:20] (03CR) 10Hashar: [C: 032] Merge rubocop related changes [oojs/ui] - 10https://gerrit.wikimedia.org/r/174088 (owner: 10Hashar) [11:23:38] (03CR) 10Hashar: [C: 032] Fixed RuboCop Style/TrailingBlankLines offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172741 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:23:50] (03CR) 10Hashar: [C: 032] Fixed RuboCop Style/SpaceAfterComma offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172740 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:24:01] (03CR) 10Hashar: [C: 032] Fixed RuboCop Style/NilComparison offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172739 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:24:14] (03CR) 10Hashar: [C: 032] Fixed RuboCop Style/SpaceAroundEqualsInParameterDefault offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172734 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:24:26] (03CR) 10Hashar: [C: 032] Fixed RuboCop Style/SignalException offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172733 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:24:35] (03CR) 10Hashar: [C: 032] Fixed RuboCop Style/NegatedIf offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172732 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:24:49] (03CR) 10Hashar: [C: 032] Fixed RuboCop Style/EmptyLines offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172731 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:24:57] (03CR) 10Hashar: [C: 032] Fixed RuboCop Style/EmptyLineBetweenDefs offence [oojs/ui] - 10https://gerrit.wikimedia.org/r/172730 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:25:07] (03CR) 10Hashar: [C: 032] Fixed RuboCop Style/AsciiComments offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172726 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:25:17] (03CR) 10Hashar: [C: 032] Fixed RuboCop Style/AndOr offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172722 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:25:38] (03CR) 10Hashar: [V: 032] Fixed RuboCop Style/AndOr offense [oojs/ui] - 10https://gerrit.wikimedia.org/r/172722 (https://bugzilla.wikimedia.org/72841) (owner: 10Zfilipin) [11:25:58] 3VisualEditor / 3Editing Tools: Layout breaks in keyboard shortcuts dialog if translated messages occupy slightly more space - 10https://bugzilla.wikimedia.org/73552 (10Dan Michael Heggø) 3UNCO p:3Unprio s:3normal a:3None Created attachment 17159 --> https://bugzilla.wikimedia.org/attachment.cgi?id... [12:09:35] 3MediaWiki extensions / 3WikiEditor: Media selector doesn't find local media in private wiki - 10https://bugzilla.wikimedia.org/63282#c6 (10Stephan Matthiesen) 5UNCO>3RESO/FIX In 1.25wmf8, it works now correctly without the (unsecure) workaround. I guess some other things with permissions were fixed whic... [12:48:05] (03CR) 10Krinkle: [C: 032] build: Update various devDependencies to latest [oojs/core] - 10https://gerrit.wikimedia.org/r/174022 (owner: 10Jforrester) [12:48:27] (03CR) 10Krinkle: [C: 032] build: Update various devDependencies to latest [oojs/ui] - 10https://gerrit.wikimedia.org/r/174027 (owner: 10Jforrester) [12:50:58] (03CR) 10Krinkle: [C: 032] build: Update various devDependencies to latest [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/174061 (owner: 10Jforrester) [12:51:10] (03Merged) 10jenkins-bot: build: Update various devDependencies to latest [oojs/ui] - 10https://gerrit.wikimedia.org/r/174027 (owner: 10Jforrester) [12:51:22] (03CR) 10Krinkle: [C: 032] build: Update grunt-jscs devDependency to latest [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174062 (owner: 10Jforrester) [12:52:30] (03Merged) 10jenkins-bot: build: Update various devDependencies to latest [oojs/core] - 10https://gerrit.wikimedia.org/r/174022 (owner: 10Jforrester) [12:52:59] (03Merged) 10jenkins-bot: build: Update grunt-jscs devDependency to latest [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174062 (owner: 10Jforrester) [12:53:42] (03Merged) 10jenkins-bot: build: Update various devDependencies to latest [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/174061 (owner: 10Jforrester) [13:16:27] (03PS1) 10Hashar: Use README.md as doxygen main page [oojs/ui] - 10https://gerrit.wikimedia.org/r/174108 [13:18:39] (03CR) 10Hashar: "Found USE_MDFILE_AS_MAINPAGE when crafting a Doxygen file for the 'cdb' repository at https://gerrit.wikimedia.org/r/#/c/174093/" [oojs/ui] - 10https://gerrit.wikimedia.org/r/174108 (owner: 10Hashar) [13:53:06] 3VisualEditor / 3Data Model: Cannot copy text when dm.surface.enabled === false - 10https://bugzilla.wikimedia.org/69515#c2 (10Ed Sanders) 5ASSI>3RESO/WON The surface model also tracks selections so disabling the surface model would also disable any meaningful copy/paste functionality. dm.Surface#disable... [14:02:44] (03CR) 10Krinkle: "Reverted in c6181662499ab8f44ae760dc4dcda651d8afd3c5." [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/97454 (https://bugzilla.wikimedia.org/35342) (owner: 10M4tx) [14:03:06] 3VisualEditor / 3Editing Tools: VisualEditor: [Regression wmf9] Making an image Frameless with out any wrap option deletes the content of the caption - 10https://bugzilla.wikimedia.org/66808#c3 (10Ed Sanders) Currently captions on inline images are only made available as wikitext by Parsoid. In order to pres... [14:03:35] 3VisualEditor / 3Editing Tools: VisualEditor: Making an image Frameless with out any wrap option deletes the content of the caption - 10https://bugzilla.wikimedia.org/66808 (10Ed Sanders) [15:16:21] 3MediaWiki / 3ResourceLoader: PHP Notice: Uninitialized string offset: includes/libs/JavaScriptMinifier.php on line 571 - 10https://bugzilla.wikimedia.org/73556 (10Bartosz Dziewoński) [15:45:35] 3OOjs UI: VisualEditor: Outline widget in booklet layout responds to mouse up events even if they started elsewhere - 10https://bugzilla.wikimedia.org/58580 (10Ed Sanders) [15:49:36] 3OOjs UI: VisualEditor: Outline widget in booklet layout responds to mouse up events even if they started elsewhere - 10https://bugzilla.wikimedia.org/58580#c1 (10Ed Sanders) 5NEW>3RESO/FIX Fixed by https://gerrit.wikimedia.org/r/#/c/133284/ [15:57:40] (03PS1) 10Cmcmahon: [BrowserTest] WIP: remove sleeps from shared tests [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174133 [15:58:11] Morning [15:59:02] (03PS2) 10Cmcmahon: i[BrowserTest] WIP: remove sleeps from shared tests [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174133 [15:59:55] (03CR) 10Jforrester: [C: 04-1] "Instead, OOUI should be aware of languages (not just LTR/RTL types), and implement this properly. Suggestions welcome." [oojs/ui] - 10https://gerrit.wikimedia.org/r/143407 (owner: 10Jforrester) [16:00:38] (03CR) 10jenkins-bot: [V: 04-1] i[BrowserTest] WIP: remove sleeps from shared tests [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174133 (owner: 10Cmcmahon) [16:00:54] Good morning [16:01:58] Question: how do I get VisualEditor to work with mediawiki 1.23.3? It looks like it requires 1.25wmf7. [16:03:41] (03PS1) 10Bartosz Dziewoński: Consistently use @return annotation [oojs/core] - 10https://gerrit.wikimedia.org/r/174136 [16:05:32] (03PS1) 10Bartosz Dziewoński: Consistently use @return annotation [oojs/ui] - 10https://gerrit.wikimedia.org/r/174137 [16:06:50] Erik__: probably your only option is to use an older version of VE :( [16:07:45] do you have a suggested tag or branch that I should checkout to try? [16:07:47] Erik__: try downloading a 1.23-compatible version from https://www.mediawiki.org/wiki/Special:ExtensionDistributor/VisualEditor , or check out the REL1_23 branch if you're using git [16:08:01] ah, awesome, thanks. [17:00:18] (03PS3) 10Cmcmahon: [BrowserTest] WIP: remove sleeps from shared tests [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174133 [17:18:11] (03PS4) 10Cmcmahon: [BrowserTest] WIP: remove sleeps from shared tests [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174133 [17:21:22] (03PS1) 10Jforrester: build: Create a 'typos' task and fix a remaining @returns to be @return [oojs/ui] - 10https://gerrit.wikimedia.org/r/174147 [17:21:27] Krinkle: ^^^ :-) [17:21:35] Krinkle: Finally got irritated by the typo fixes. [17:22:45] James_F: Hm.. no jscs? [17:23:02] James_F: does it support regexes already? [17:23:03] Oh it does [17:23:06] Krinkle: ? [17:23:15] it'll fail momentarily I hope [17:23:15] MatmaRex: No, I made it simple for now in the name of speed. [17:23:19] MatmaRex: Maybe later. [17:23:26] we have some visiblities and movablilties in code comments. :) [17:24:12] (03CR) 10Jforrester: [C: 04-1] "Needs re-basing. Can't remember for sure, but I /think/ this is good to go otherwise." [oojs/ui] - 10https://gerrit.wikimedia.org/r/164007 (owner: 10Jforrester) [17:24:34] (03CR) 10Jforrester: [C: 04-1] "Needs rebasing. Can't remember what was outstanding." [oojs/ui] - 10https://gerrit.wikimedia.org/r/163785 (owner: 10Jforrester) [17:26:23] 3VisualEditor: VisualEditor: Move IE CSS hacks to core - 10https://bugzilla.wikimedia.org/73565 (10Ed Sanders) 3NEW p:3Unprio s:3normal a:3Roan Kattouw Most of the fixes in ve.init.mw.IEHacks.css are not MW-specific and would be useful to have in core. We need to chose a method for selectively loading... [17:27:39] (03CR) 10Jforrester: [C: 032] Consistently use @return annotation [oojs/core] - 10https://gerrit.wikimedia.org/r/174136 (owner: 10Bartosz Dziewoński) [17:27:49] Krinkle: What will fail? [17:28:37] Krinkle: so, any more news on the great NPM vs Bower battle? [17:30:14] (03CR) 10Krinkle: build: Create a 'typos' task and fix a remaining @returns to be @return (033 comments) [oojs/ui] - 10https://gerrit.wikimedia.org/r/174147 (owner: 10Jforrester) [17:30:28] James_F: There is a space missing somewhere in typos.js [17:30:36] Corruption alert: visualeditor-needcheck on svwiki: https://sv.wikipedia.org/?diff=28755235 [17:30:36] Corruption alert: visualeditor-needcheck on simplewiki: https://simple.wikipedia.org/?diff=4940556 [17:30:36] Corruption alert: visualeditor-needcheck on hewiki: https://he.wikipedia.org/?diff=16222083 [17:30:36] Corruption alert: visualeditor-needcheck on hewiki: https://he.wikipedia.org/?diff=16222114 [17:30:36] Corruption alert: visualeditor-needcheck on hewiki: https://he.wikipedia.org/?diff=16222117 [17:30:49] Krinkle: There is? [17:30:49] James_F: it looked like there was no jscs in Gruntfile, but I missed it [17:31:01] Krinkle: So… what are you saying? [17:31:17] James_F: There is a missing space coding-stylewise in the if [17:31:24] and jscs didn;t find it [17:31:32] (03Merged) 10jenkins-bot: Consistently use @return annotation [oojs/core] - 10https://gerrit.wikimedia.org/r/174136 (owner: 10Bartosz Dziewoński) [17:31:41] Krinkle: Which if? [17:32:08] 3VisualEditor / 3ContentEditable: VisualEditor: Warn if editing any reference that is used multiple times - 10https://bugzilla.wikimedia.org/73566 (10etonkovidova) 3NEW p:3Unprio s:3normal a:3None The fix 52792 will display a warning "This reference is used [n] times on this page." when editing Basic... [17:32:20] James_F: the 'for' [17:32:48] Krinkle: Which for? Which file? You seem to be talking about… the Gruntfile? [17:33:06] : James_F: There is a space missing somewhere in typos.js [17:33:22] There is only one 'for' [17:33:23] 3VisualEditor / 3Editing Tools: VisualEditor: Warn if editing a reference that is used multiple times - 10https://bugzilla.wikimedia.org/52792#c3 (10etonkovidova) 5RESO/?>3VERI The warning "This reference is used [n] times on this page." appears for Basic citations only - verified the fix in beta, test2... [17:44:36] 3VisualEditor / 3ContentEditable: VisualEditor: Copy paste is not working for Math elements - 10https://bugzilla.wikimedia.org/58981#c6 (10Alex Monk) a:5Alex Monk>3None No, I mean it renders fine for me. [17:44:41] Krinkle: OK. [17:45:31] I have electricity today, yay. [17:45:58] Krenair: Impressive. [17:50:02] https://simple.wikipedia.org/?diff=4940556 is weird [17:51:06] 3VisualEditor: VisualEditor: Warn if editing any reference that is used multiple times - 10https://bugzilla.wikimedia.org/73566 (10Alex Monk) [17:53:49] (03PS2) 10Jforrester: build: Create a 'typos' task and fix a remaining @returns to be @return [oojs/ui] - 10https://gerrit.wikimedia.org/r/174147 [17:56:22] 3VisualEditor / 3Editing Tools: VisualEditor: Warn if editing a citation that is used multiple times - 10https://bugzilla.wikimedia.org/73566 (10James Forrester) 5NEW>3ASSI p:5Unprio>3Normal s:5normal>3enhanc [17:58:05] (03CR) 10Jforrester: build: Create a 'typos' task and fix a remaining @returns to be @return (033 comments) [oojs/ui] - 10https://gerrit.wikimedia.org/r/174147 (owner: 10Jforrester) [18:02:24] 3VisualEditor / 3ContentEditable: VisualEditor: Pawn ♙ appears after deleting the text preceding and following a link by "undo" action - 10https://bugzilla.wikimedia.org/61521#c8 (10Alex Monk) a:5Alex Monk>3None I'm not taking this bug either, but I can reproduce it -if I make the link in the block slug-... [18:22:15] James_F, near the end the audio was terrible (even on audio-only), tried to reconnect and then it wouldn't work [18:22:20] did you mention something about a list of typos? [18:22:28] I found enwikipedia has one of those the other day [18:24:38] Krenair: I did, yeah. [18:25:12] edsanders: Did something in the Math extension break VE again? [18:25:24] edsanders: Beta has a fatal (undefined is not a function)… [18:25:37] edsanders: On insert. [18:28:21] https://en.wikipedia.org/wiki/Wikipedia:Lists_of_common_misspellings/For_machines [18:28:25] pf [18:28:52] 3VisualEditor / 3Technical Debt: VisualEditor: Move IE CSS hacks to core - 10https://bugzilla.wikimedia.org/73565 (10James Forrester) 5NEW>3ASSI p:5Unprio>3High s:5normal>3enhanc a:5Roan Kattouw>3Alex Monk [18:29:18] Krenair: Interesting. [18:29:23] Krenair: For AWB, it looks like? [18:29:46] um, can i aquaint you with https://en.wikipedia.org/wiki/Wikipedia:AutoWikiBrowser/Typos ? [18:30:20] MatmaRex: So… no, a fork of efforts. How very enwiki-like. :-) [18:33:13] James_F, I get no console errors, but "Failed to parse (Cannot store math image on filesystem)" [18:43:35] 3VisualEditor / 3Editing Tools: VisualEditor: Implement a find/search and replace tool - 10https://bugzilla.wikimedia.org/48368 (10Ed Sanders) [18:48:06] 3VisualEditor / 3ContentEditable: VisualEditor: [Regression wmf6] deleting text formatted as computer code produces: Failed to load resource: The operation couldn’t be completed. (kCFErrorDomainCFNetwork error -10.) in Safari - 10https://bugzilla.wikimedia.org/72698 (10Roan Kattouw) [18:59:06] 3VisualEditor / 3ContentEditable: VisualEditor: Link inspector opens in a upper left corner in Safari - 10https://bugzilla.wikimedia.org/73336 (10James Forrester) a:3Ed Sanders [19:00:24] 3VisualEditor / 3Editing Tools: Layout breaks in keyboard shortcuts dialog if translated messages occupy slightly more space - 10https://bugzilla.wikimedia.org/73552#c1 (10etonkovidova) Created attachment 17161 --> https://bugzilla.wikimedia.org/attachment.cgi?id=17161&action=edit IE Keyboard shortcuts in... [19:00:25] 3VisualEditor / 3MediaWiki integration: VisualEditor: Provide keyboard shortcut of Esc key to quit the editor, bring back to read mode (or trigger "are you sure?" confirmation dialog if needed) - 10https://bugzilla.wikimedia.org/73363 (10James Forrester) a:3Alex Monk [19:01:53] 3VisualEditor / 3Editing Tools: VisualEditor:[Regression wmf8] All inspector dialog UI is broken when added inside Reference/Media Settings dialog box - 10https://bugzilla.wikimedia.org/73542 (10James Forrester) a:3Roan Kattouw [19:02:37] 3MediaWiki / 3Page editing: of editing page doesn't respect DISPLAYTITLE - 10https://bugzilla.wikimedia.org/49667 (10James Forrester) a:3Alex Monk [19:02:52] <wikibugs> 3VisualEditor / 3Editing Tools: Layout breaks in keyboard shortcuts dialog if translated messages occupy slightly more space - 10https://bugzilla.wikimedia.org/73552#c2 (10etonkovidova) 5UNCO>3NEW There are two cases for which the bug has been confirmed: - IE11 will display the keyboard short cuts in t... [19:06:51] <grrrit-wm> (03CR) 10Esanders: build: Create a 'typos' task and fix a remaining @returns to be @return (031 comment) [oojs/ui] - 10https://gerrit.wikimedia.org/r/174147 (owner: 10Jforrester) [19:07:29] <grrrit-wm> (03CR) 10Jforrester: [C: 04-1] build: Create a 'typos' task and fix a remaining @returns to be @return (031 comment) [oojs/ui] - 10https://gerrit.wikimedia.org/r/174147 (owner: 10Jforrester) [19:08:49] <grrrit-wm> (03CR) 10Catrope: [C: 04-1] Provide a FileDropHandler for CSV and TSV files (031 comment) [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/173296 (owner: 10Jforrester) [19:12:41] <grrrit-wm> (03CR) 10Jforrester: Provide a FileDropHandler for CSV and TSV files (031 comment) [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/173296 (owner: 10Jforrester) [19:12:47] <grrrit-wm> (03PS5) 10Jforrester: Provide a FileDropHandler for CSV and TSV files [VisualEditor/VisualEditor] - 10https://gerrit.wikimedia.org/r/173296 [19:15:37] <grrrit-wm> (03CR) 10Esanders: build: Create a 'typos' task and fix a remaining @returns to be @return (031 comment) [oojs/ui] - 10https://gerrit.wikimedia.org/r/174147 (owner: 10Jforrester) [19:16:58] <James_F> edsanders: That's me. ;-) [19:17:52] <wikibugs> 3OOjs UI: VisualEditor: Outline widget in booklet layout responds to mouse up events even if they started elsewhere - 10https://bugzilla.wikimedia.org/58580 (10James Forrester) p:5Unprio>3Low a:3Trevor Parscal [19:18:22] <wikibugs> 3VisualEditor / 3Technical Debt: VisualEditor: Calling buildNodeTree() manually should not be needed - 10https://bugzilla.wikimedia.org/73569 (10Roan Kattouw) 3NEW p:3Unprio s:3normal a:3Ed Sanders We explicitly call buildNodeTree() in a few places. Ideally, we would not need to do this outside of ve... [19:20:51] <wikibugs> 3VisualEditor / 3Technical Debt: VisualEditor: Calling buildNodeTree() manually should not be needed - 10https://bugzilla.wikimedia.org/73569 (10James Forrester) 5NEW>3ASSI p:5Unprio>3Normal [19:21:36] <BladeBronson> hey guys, i have a question about eventemitter [19:21:53] <James_F> BladeBronson: Hey. [19:21:54] <RoanKattouw> BladeBronson: Hey man, good to see you around here [19:21:56] <RoanKattouw> Shoot [19:22:01] <BladeBronson> i've subclassed select and option widgets and added more elements. [19:22:11] <BladeBronson> clicking (mouseup more specifically) emits 'selected' [19:22:36] <BladeBronson> i want to emit another event when other dom elements inside optionwidget are clicked [19:23:11] <RoanKattouw> Right, OptionWidget emits 'selected', and then the SelectWidget that contains it emits 'select' [19:23:20] <BladeBronson> i have one buttonwidget that seems to emit propery (and prevent the 'selected') emit. however, i'm creating some dom on the fly (this.mything = this.$('<div>')) [19:23:38] <BladeBronson> and i'm using dom events this.mything.on('click', ... ) [19:23:47] <RoanKattouw> ryasmeen: AndyRussG in #wikimedia-dev [19:23:59] <BladeBronson> so, what i've figured out is that i can use this.mything.on('mousedown', ... return false; ) [19:24:11] <BladeBronson> and it prevents the option/select from emitting 'selected' [19:24:15] <BladeBronson> does that sound like the right way to go? [19:24:30] <RoanKattouw> Ahm, hold on [19:24:46] <AndyRussG> RoanKattouw: eh? [19:24:49] <RoanKattouw> So you want to have certain things in your OptionWidget that, when you click them, they prevent 'selected' from being emitted and instead emit something else? [19:24:52] <RoanKattouw> AndyRussG: Oh you're here too! [19:24:54] <BladeBronson> right [19:25:14] <RoanKattouw> AndyRussG: ryasmeen noticed some weirdness in beta labs that looks like it's CentralNotice-related, so I told her to talk to you about it [19:25:49] <AndyRussG> RoanKattouw: OK thank you, yes we did deploy new stuff there yesterday [19:25:56] <ryasmeen> AndyRussG: We are getting this on top of every page when we open it with VE:<div id="centralnotice_testbanner_name">two</div> [19:26:48] <RoanKattouw> BladeBronson: Ahm... I don't see any code where OptionWidgets emit 'selected'. Instead, I see SelectWidget doing the mousedown/up binding [19:26:55] <AndyRussG> ryasmeen: hmm, OK... I know a campaign was enabled there precisely to test new CentralNotice features [19:27:13] <RoanKattouw> But in any case, if you prevent mousedown within the OptionWidget itself, then the SelectWidget will never be notified about it [19:27:14] <RoanKattouw> So that sounds legit [19:27:30] <AndyRussG> ryasmeen: You're getting that displayed as text? [19:27:45] <AndyRussG> We didn't actually touch the code that injects the banners per se [19:27:45] <ryasmeen> AndyRussG: yeah [19:28:08] <AndyRussG> K I'll have a look, thank you!! :) [19:28:18] <ryasmeen> Sure! :) [19:29:01] <BladeBronson> i should have a look at the latest oojs-ui to verify, but in our slightly older version, selectwidget has onMouseUp which calls chooseItem which calls selectItem which emits 'select' [19:29:50] <RoanKattouw> OK yeah that's what I have too [19:29:52] <wikibugs> 3VisualEditor / 3Technical Debt: VisualEditor: Improve FileDropHandler API - 10https://bugzilla.wikimedia.org/73570 (10Roan Kattouw) 3NEW p:3Unprio s:3normal a:3None From comments at https://gerrit.wikimedia.org/r/#/c/173296/1/src/ui/filedrophandlers/ve.ui.DSVFileDropHandler.js : * Right now every c... [19:30:19] <RoanKattouw> So yeah as you might have noticed, it's the SelectWidget binding the mouse events, but the thing you're actually clicking is an OptionWidget [19:30:30] <RoanKattouw> So it's relying on the event bubbling up from the OptionWidget to its parent the SelectWidget [19:30:46] <RoanKattouw> If you return false;, that bubbling is prevented. You can also call e.stopPropagation(); to prevent bubbling [19:31:00] <BladeBronson> right, okay [19:31:28] <RoanKattouw> So which of those two you need also depends on whether or not you need to preventDefault on the mousedown [19:32:05] <RoanKattouw> ( return false; is equivalent to doing both e.preventDefault(); and e.stopPropagation(); ) [19:32:12] <BladeBronson> yep [19:32:34] <BladeBronson> okay, then i'll just keep going on what i have [19:32:40] <BladeBronson> thanks! [19:34:22] <wikibugs> 3VisualEditor / 3Editing Tools: VisualEditor: Suggestion lists (of templates/links/categories/etc.) differ over time with the same input (e.g. between browsers/users) - 10https://bugzilla.wikimedia.org/53231#c2 (10James Forrester) *** Bug 70219 has been marked as a duplicate of this bug. *** [19:34:22] <wikibugs> 3VisualEditor / 3Editing Tools: Order of suggested link targets is random - 10https://bugzilla.wikimedia.org/70219#c1 (10James Forrester) 5NEW>3RESO/DUP Merging with bug 53231. *** This bug has been marked as a duplicate of bug 53231 *** [19:37:40] <grrrit-wm> (03PS5) 10Cmcmahon: [BrowserTest] WIP: remove sleeps from shared tests [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174133 [19:37:47] <grrrit-wm> (03PS6) 10Catrope: Add instrumentation for edit schema [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/171499 [19:37:52] <wikibugs> 3VisualEditor / 3Technical Debt: VisualEditor: Improve FileDropHandler API - 10https://bugzilla.wikimedia.org/73570 (10James Forrester) 5NEW>3ASSI p:5Unprio>3High s:5normal>3enhanc [19:42:54] <wikibugs> 3VisualEditor / 3Editing Tools: VisualEditor: After updating a page <br /> gets added after the timestamp of last edit made of that page - 10https://bugzilla.wikimedia.org/73571 (10Rummana Yasmeen) 3NEW p:3Unprio s:3normal a:3None Created attachment 17163 --> https://bugzilla.wikimedia.org/attachme... [19:43:05] <wikibugs> 3VisualEditor / 3Editing Tools: VisualEditor: After updating a page <br /> gets added after the timestamp of last edit made on that page - 10https://bugzilla.wikimedia.org/73571 (10Rummana Yasmeen) [19:43:19] <ryasmeen> Hey Krenair! [19:44:29] <ryasmeen> RoanKattouw says you know a lot about this : https://bugzilla.wikimedia.org/show_bug.cgi?id=73571 :P [19:45:20] <BladeBronson> ah, i see that ButtonedElement has a static property to cancel mousedown events, which in turn returns false. that explains why the buttonwidget that i have inside of option wasn't triggering selectwidget to emit 'select' [19:46:36] <wikibugs> 3VisualEditor / 3ContentEditable: VisualEditor: Preserve `data-parsoid` attribute on internal copy-paste so that Parsoid preserves e.g. syntax layout - 10https://bugzilla.wikimedia.org/72426#c18 (10ssastry) Parsoid can handle the preserved data-parsoid .. but, i haven't verified that nothing breaks if data-p... [19:47:40] <AndyRussG> ryasmeen: I'm not getting any issues with banners in VE on http://en.wikipedia.beta.wmflabs.org [19:48:09] <etonkovi_> edsanders: bug 73336 Resolved-fixed -Change 173007 merged by jenkins-bot - cannot see the fix in UI... [19:48:23] <AndyRussG> ryasmeen: Can you tell me which browser/platform/article, or if that's the right beta site, or any other info that would help me replicate? [19:48:40] <ryasmeen> AndyRussG: What happens when you switch to edit source ? [19:49:03] <ryasmeen> because I just got that on the same page using chrome [19:49:43] <AndyRussG> ryasmeen: With edit source, I did get the banner... [19:50:29] <AndyRussG> ...displayed as a banner, not showing the <div> tag as text or anything [19:50:44] <ryasmeen> yeah, if you make any change using VE and save the page, you will see it on top of it in readmode too [19:52:20] <AndyRussG> ryasmeen: it is expected that you'll get the banners in readmode. They are tests of banners like the ones used for fundraising. You should be getting one, two or three 31% of the time, and nothing 7% of the time. [19:52:35] <grrrit-wm> (03PS6) 10Cmcmahon: [BrowserTest] WIP: remove sleeps from shared tests [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174133 [19:52:49] <AndyRussG> I mean, one, two or three 31% of the time each, that is, one of them 93% of the time [19:52:59] <ryasmeen> Okay , I got that <div> tag in this page:http://en.wikipedia.beta.wmflabs.org/wiki/User:RYasmeen_(WMF) [19:54:12] <AndyRussG> ryasmeen: I am getting the div in the HTML, but not the actual "text" displayed "<div>" displayed [19:55:10] <ryasmeen> ah no! I am not getting that either! Sorry I misunderstood your question [19:55:18] <AndyRussG> Ah OK [19:55:27] <AndyRussG> Yes you should get those words, inside a div [19:55:34] <ryasmeen> it is appearing as "One", "two" on the page but not with tags right. [19:55:43] <AndyRussG> OK [19:55:47] <RoanKattouw> AndyRussG: https://i.imgur.com/vU7JyBr.png [19:55:52] <AndyRussG> I'm just checking if you're expected to get it when editing [19:56:05] <AndyRussG> RoanKattouw: ryasmeen: yes that's correct [19:56:15] <AndyRussG> no issuz there :) [19:56:35] <AndyRussG> I'm surprised that it shows when you edit but I'm not sure about the expected behaviour, lemme check [19:57:29] <AndyRussG> (on another note, what I'm slightly concerned about is the apparent greater page bump with the new banner system, but that may be because of beta labs inferior infrastructure? or my imagination...) [19:59:03] <RoanKattouw> AndyRussG: It does not show during editing, but it appears after saving [19:59:05] <wikibugs> 3VisualEditor / 3Technical Debt: VisualEditor: Calling buildNodeTree() manually should not be needed - 10https://bugzilla.wikimedia.org/73569#c1 (10Ed Sanders) The node tree is lazily evaluated by buildNodeTree the first time you call getDocumentNode. In the examples your provide (paste and drag-drop) we cal... [19:59:25] <AndyRussG> RoanKattouw: well, appearing after saving is normal since you're viewing the page [19:59:32] <RoanKattouw> Yeah I suppose that's right [19:59:38] <RoanKattouw> I see it's there before I start editing too [19:59:45] <AndyRussG> That's also correct [19:59:46] <grrrit-wm> (03PS7) 10Cmcmahon: [BrowserTest] remove sleeps from shared tests [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174133 [20:00:01] <RoanKattouw> We often go directly to the editing view so we don't always see or pay much attention to what's there before [20:00:08] <AndyRussG> You should be getting one of them randomly 93% of the time on most page views [20:00:09] <AndyRussG> Ah OK [20:00:14] <RoanKattouw> Yeah it looks right [20:00:14] <grrrit-wm> (03CR) 10Cmcmahon: [C: 032] "maintenance" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174133 (owner: 10Cmcmahon) [20:00:20] <RoanKattouw> "Nothing to see here, move along" then :) [20:00:28] <AndyRussG> I'm just checking if you're expected to get it when editing source! that surprised me [20:01:01] <RoanKattouw> Hmm it's also there then, I see [20:01:05] <RoanKattouw> That ... seems right though? [20:02:11] <grrrit-wm> (03Merged) 10jenkins-bot: [BrowserTest] remove sleeps from shared tests [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174133 (owner: 10Cmcmahon) [20:08:00] <AndyRussG> RoanKattouw: ryasmeen: yes, confirmed that banners should show up when you edit source! thanks so much for pointing this out [20:09:03] <AndyRussG> It's such a single point of failure and at the same time substantial changes like yesterday's could potentially bork other things in so many unforseen ways [20:09:20] <grrrit-wm> (03PS1) 10Cmcmahon: [BrowserTest] forgot a bit of dead code [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174231 [20:09:40] <AndyRussG> So please don't hesitate to reach out if you see anything that feels a bit off, even if you're think it might be expected... thanks again!! :) [20:09:44] <grrrit-wm> (03CR) 10Cmcmahon: [C: 032] "maintenance" [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174231 (owner: 10Cmcmahon) [20:10:21] <wikibugs> 3VisualEditor / 3Editing Tools: VisualEditor: Making an image Frameless with out any wrap option deletes the content of the caption - 10https://bugzilla.wikimedia.org/66808 (10James Forrester) [20:11:15] <grrrit-wm> (03Merged) 10jenkins-bot: [BrowserTest] forgot a bit of dead code [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174231 (owner: 10Cmcmahon) [20:15:13] <MatmaRex> TrevorParscal: around? here's a thought. CheckboxInputWidget should be labelled, rather than relying on a FieldLayout for the label. what do you think? [20:15:19] <TrevorParscal> hey man [20:15:30] <TrevorParscal> hmm [20:15:37] <TrevorParscal> well, we've sort of been through this before [20:15:44] <James_F> MatmaRex: When I created it TrevorParscal said no. :-) [20:16:05] <TrevorParscal> originally James created CheckboxWidget in addition to CheckboxInputWidget [20:16:06] <MatmaRex> well, i wasn't through it! :) [20:16:15] * James_F grins. [20:16:28] <TrevorParscal> CheckboxInputWidget obviously should not have a label, it's only the input [20:17:03] <TrevorParscal> but when you actually write the code for CheckboxWidget you quickly see that it gets messy [20:17:13] <TrevorParscal> one of the reasons is it's duplicative of FieldLayout [20:17:54] <wikibugs> 3VisualEditor / 3MediaWiki integration: VisualEditor: After updating a page <br /> gets added after the timestamp of last edit made on that page - 10https://bugzilla.wikimedia.org/73571 (10James Forrester) 5NEW>3ASSI a:3Alex Monk [20:18:04] <TrevorParscal> the other is that it either hacks CheckboxInputWidget (making styling tricky) or it embeds a CheckboxInputWidget inside, making access to it less convenient - much the way DropdownWidget causes issues by nesting it's menu [20:18:06] <wikibugs> 3VisualEditor / 3MediaWiki integration: VisualEditor: After updating a page <br /> gets added after the timestamp of last edit made on that page - 10https://bugzilla.wikimedia.org/73571 (10James Forrester) p:5Unprio>3Normal s:5normal>3minor [20:20:07] <wikibugs> 3VisualEditor / 3ContentEditable: VisualEditor: Preserve `data-parsoid` attribute on internal copy-paste so that Parsoid preserves e.g. syntax layout - 10https://bugzilla.wikimedia.org/72426#c19 (10Roan Kattouw) (In reply to ssastry from comment #18) > Parsoid can handle the preserved data-parsoid .. but, i... [20:24:22] <wikibugs> 3VisualEditor / 3ContentEditable: VisualEditor: DOM Exception (no user-facing loss of functionality?) after a long editing session in Safari - 10https://bugzilla.wikimedia.org/73540#c1 (10James Forrester) 5NEW>3ASSI p:5Unprio>3Low s:5normal>3minor We should find out what this is before WONTFIXing... [20:24:41] <mooeypoo> Gaaaaaaa position() and offset(). I hate you. [20:27:07] <wikibugs> 3OOjs UI: OOjs UI: OO.ui.ButtonWidget default target should not be "_blank" - 10https://bugzilla.wikimedia.org/73452#c2 (10James Forrester) 5NEW>3ASSI p:5Unprio>3Normal a:3Richhiey (In reply to Richhiey from comment #1) > Created attachment 17148 [details] > Changed the config for target to '_self'.... [20:27:36] <wikibugs> 3VisualEditor / 3ContentEditable: VisualEditor: shortcut(Ctrl+1) for Page Title does not work on Mac - 10https://bugzilla.wikimedia.org/73340 (10James Forrester) p:5Unprio>3High [20:28:10] <TrevorParscal> MatmaRex: any further thoughts on the checkbox thing? [20:28:37] <wikibugs> 3OOjs UI: OOjs UI: trouble with CheckboxInputWidget in Opera - 10https://bugzilla.wikimedia.org/73517 (10James Forrester) 5NEW>3ASSI p:5Unprio>3Normal a:3Bartosz Dziewoński [20:28:59] <MatmaRex> TrevorParscal: i need to think about it. you're probably right, but i still don't like how it works [20:29:22] <TrevorParscal> i think you have good instincts, and I'm open to your thoughts on how to improve it [20:29:36] <TrevorParscal> feel free to bring it up again - at least you now know my arguments [20:33:40] <mooeypoo> RoanKattouw, I found my issue. It's screen vs. document position again. It was easier in horizontal because the only "top" position changing wasn't quite fom dragging. Now that top is being dragged up and down, the document vs screen position is very important. GAAAH. [20:33:53] <mooeypoo> clientX vs screenX again. Yeepy. [20:34:22] <wikibugs> 3VisualEditor / 3Editing Tools: VisualEditor: Blank link inspector UI appears when trying to open it inside the Media Settings while selecting another image in the caption - 10https://bugzilla.wikimedia.org/68130#c1 (10James Forrester) 5NEW>3RESO/FIX The inspector now "works", but due to bug 72431 immedi... [20:35:19] <RoanKattouw> Krinkle: Are there any plans to fix https://bugzilla.wikimedia.org/show_bug.cgi?id=72334 ? [20:37:22] <wikibugs> 3VisualEditor: VisualEditor: Performance issues (tracker) - 10https://bugzilla.wikimedia.org/49685 (10James Forrester) [20:37:22] <wikibugs> 3VisualEditor / 3Data Model: VisualEditor and Parsoid: When saving a page, only send back the parts of the page that have been changed - 10https://bugzilla.wikimedia.org/68042 (10James Forrester) [20:39:37] <wikibugs> 3OOjs UI: OOjs UI: trouble with CheckboxInputWidget in Opera - 10https://bugzilla.wikimedia.org/73517#c2 (10Bartosz Dziewoński) p:5Normal>3Lowest Assuming this is about Opera 12, then this is a Opera bug where SVG background-image sometimes disappears in presence of border-radius and the right phase of the... [20:42:21] <wikibugs> 3OOjs UI: OOjs UI: trouble with CheckboxInputWidget in Opera - 10https://bugzilla.wikimedia.org/73517#c3 (10Bartosz Dziewoński) (In reply to Bartosz Dziewoński from comment #2) > The solution is to stop serving SVG to Opera 12, like Thiemo has figured out > and did it for this new checkbox style back when it w... [20:44:52] <wikibugs> 3VisualEditor / 3Technical Debt: VisualEditor: Spin out UnicodeJS - 10https://bugzilla.wikimedia.org/60731#c1 (10James Forrester) 5ASSI>3RESO/FIX a:3James Forrester Done at Wikimania 2014. [20:47:08] <wikibugs> 3VisualEditor / 3Technical Debt: VisualEditor: Load visualeditor modules in a closure mapping $/mw/ve to their globals - 10https://bugzilla.wikimedia.org/48885 (10James Forrester) a:3Krinkle [20:50:22] <wikibugs> 3VisualEditor / 3Editing Tools: VisualEditor: Better, automatic reference adding within VE (including parsing destination URL, access timestamps, and intelligent defaults) - 10https://bugzilla.wikimedia.org/60768 (10James Forrester) a:3mvolz [20:50:41] <mooeypoo> RoanKattouw, so, if I use screenX for horizontal dragging and pageY for vertical dragging it means something's not done right, right? :\ 'cause it works perfectly like that, but it's two different coordinate systems. [20:51:22] <wikibugs> 3VisualEditor / 3Initialisation: VisualEditor: Pre-fetch modules to improve speed for end-users - 10https://bugzilla.wikimedia.org/57952 (10James Forrester) [20:51:22] <wikibugs> 3VisualEditor: VisualEditor: Performance issues (tracker) - 10https://bugzilla.wikimedia.org/49685 (10James Forrester) [20:52:22] * MatmaRex stabs SelectWidget [20:52:31] <grrrit-wm> (03PS1) 10Catrope: Fix back button not exiting VE [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174270 [20:52:48] <MatmaRex> the wretched thing emits four different events that all look like they do the same thing [20:52:54] <RoanKattouw> mooeypoo: That sounds very fishy [20:53:05] <mooeypoo> It does. [20:53:06] <RoanKattouw> mooeypoo: Have you tried using pageX instead of screenX? [20:53:12] <mooeypoo> RoanKattouw, I am now [20:53:21] <mooeypoo> but some of the calculations will be adjusted [20:53:27] <RoanKattouw> Also, with the code that uses screenX, have you tried breaking it by scrolling *horizontally* (rather than vertically)? [20:53:33] <mooeypoo> why would it *work* though for X and not Y is another interesting question [20:53:41] <mooeypoo> I suspect it's because we don't really scroll sideways [20:53:45] <mooeypoo> ... minor detail [20:53:52] <RoanKattouw> (e.g. by adding <div style="width:1000px; height:2px;"></div> and making your browser window narrow) [20:54:00] <mooeypoo> RoanKattouw, yeah I am doing that now [20:54:10] <mooeypoo> and yes, I think you're correct in saying it's broken. [20:54:49] <mooeypoo> I asked more as a "wtf just happened" and an addendum to your fascination with web coordinates than an actual "should I keep it this way" question. [20:55:18] <MatmaRex> (ooookay, so three of them are useless and i only want one.) [20:55:47] <RoanKattouw> MatmaRex: There is a subtle difference between 'choose' and 'select'. What are the other two? [20:56:11] <RoanKattouw> MatmaRex: Also if this is an area that should be documented better, you should flag that to TrevorParscal and kirstenMA [20:56:12] <MatmaRex> yeah, i figured out the subtleties [20:56:42] <MatmaRex> well, right now figuring them requires some grepping for uses, so documentation would help. [20:56:52] <wikibugs> 3OOjs UI: OOjs UI: Toolbar should collapse items rather than span multiple lines when the window is narrow - 10https://bugzilla.wikimedia.org/50227 (10James Forrester) [20:57:13] <MatmaRex> there's also press and highlight, i've figured out what these do already, too. [20:57:43] <RoanKattouw> Aah OK [20:57:52] <RoanKattouw> Yeah there's another subtle difference between highlighted and selected [20:58:05] <RoanKattouw> Trevor knows exactly what these differences are, but they should be documented [20:58:16] <kirstenMA> I'm reading this all, btw [20:58:36] <kirstenMA> will note [20:59:22] <TrevorParscal> MatmaRex: which even do you need? [20:59:38] <TrevorParscal> there's highlight, select, and choose [20:59:54] <MatmaRex> TrevorParscal: i'm doing the radio thing, and i need choose, probably [20:59:58] <TrevorParscal> yeah [21:00:04] <TrevorParscal> choose is deliberate user action [21:00:10] <TrevorParscal> choose causes select [21:01:09] <TrevorParscal> select can be programatic though, so if you are only wanting to close a menu when the user chooses an item, for instance, you want to use choose, but if you are programatically changing the selected item and don't want there to be UI side effects other than showing which item is selected, use select [21:01:28] <TrevorParscal> they are separate because otherwise we have to track where the event originated, and that pattern falls apart quickly [21:01:34] <TrevorParscal> instead of we have separate events [21:01:52] <TrevorParscal> we probably need to also do something similar to input widgets tbh [21:01:58] <RoanKattouw> Yes we do [21:02:00] <RoanKattouw> We really do [21:02:02] <RoanKattouw> Please do [21:02:16] <TrevorParscal> change (could be programatic or user driven) and edit (caused by the user) [21:02:19] <TrevorParscal> for instance [21:03:06] <TrevorParscal> HTML of course doesn't really give this to us, so we won't know the difference between the user typing something and $( input ).val( newVal ) [21:03:11] <RoanKattouw> Sure [21:03:16] <TrevorParscal> but we can know the difference between typing and setValue [21:03:20] <RoanKattouw> But we can at least distinguish between... yeah [21:03:20] <TrevorParscal> so, we should be fine [21:03:59] <TrevorParscal> DOM events fall very very short of what is needed to build good UI, so we have to abstract, as usual [21:04:05] <TrevorParscal> welcome to OOjs UI [21:06:52] <mooeypoo> RoanKattouw, you're going to *LOVE* this: I have to use clientX and clientY for .elementFromPoint() but I will use pageX and pageY for all the position calculations. Ain't dom positioning grand!? :D [21:07:15] <RoanKattouw> Yeah that sounds about right [21:07:49] <RoanKattouw> What you could do, though, and what might be more efficient, is use client coords for everything [21:08:10] <RoanKattouw> Use clientX and clientY, and for adjusting those relative to the draggable container, get the client coords of the draggable container [21:08:11] <mooeypoo> RoanKattouw, that doesn't work, because I can't really get the client coordinates for the element itself properly [21:08:29] <RoanKattouw> $draggableContainer[0].getBoundingClientRect(); [21:08:32] <mooeypoo> .offset() and .position() give me relative and document positions [21:08:33] <mooeypoo> oh [21:08:45] <mooeypoo> hmm. [21:08:48] <RoanKattouw> It'll be cheaper to compute [21:08:54] <mooeypoo> let me see if I can easily set it up ---oh. It is? [21:08:58] <mooeypoo> if it is, I might as well [21:09:16] <mooeypoo> 'cause this is done while you move the mouse in drag, so it's probably better to go that route [21:09:16] <RoanKattouw> I mean, you know, don't melt your brain with it [21:09:19] <RoanKattouw> But if you can pull it off... [21:09:33] <RoanKattouw> Yeah exactly, it's a tight code path [21:09:47] * mooeypoo shall try that [21:10:49] <mooeypoo> wait, I have to do $obj[0].getBoundingClientRect() ? because I go $obj.offset() [21:11:00] <mooeypoo> without the [0] ... this always confuses me. [21:11:28] <RoanKattouw> Oh it's because .offset() is a jQuery thing whereas .getBoundingClientRect() is a native DOM thing [21:11:33] <RoanKattouw> [0] on a jQuery object gets you the native DOM node [21:11:48] <mooeypoo> ohh [21:11:50] <RoanKattouw> (All jQuery objects are really jQuery *collections*, and they behave in a roughly array-like fashion) [21:11:56] * mooeypoo nods [21:12:04] <mooeypoo> I knew that, I think, but the when to use either is always confusing. [21:12:21] <mooeypoo> Okay, so this will be cheaper also because of that? [21:12:29] <mooeypoo> it bypasses jQuery calculations, I assume? [21:12:44] <RoanKattouw> That's a minor factor I thin [21:12:55] <RoanKattouw> The main thing is that they need to go and compute multiple things [21:13:06] <mooeypoo> I'd suspect the pageX thing would be the cheapest [21:13:10] <RoanKattouw> Especially .position() needs to find the offset parent and compute *its* .offset() in addition to your own .offset() [21:13:21] <mooeypoo> since it's document position, which is global [21:13:33] <RoanKattouw> pageX is probably computed as screenX + scrollTop or something [21:13:34] <Krenair> RoanKattouw, I've been wondering about the way we do citations... to change the group of one, the only way is to select it, open the menu and select Basic, right? [21:13:41] <mooeypoo> ha [21:13:42] <Krenair> You can do the same thing to get the warning for https://bugzilla.wikimedia.org/show_bug.cgi?id=73566 [21:13:45] <RoanKattouw> Krenair: Yeah, which sucks [21:13:49] <RoanKattouw> both suck [21:14:01] <Krenair> I wonder if we have a bug for the first [21:14:27] <RoanKattouw> mooeypoo: Similarly, jQuery .offset() is probably something like getClientRects() + scrollTop(/Left) [21:14:35] <mooeypoo> oi vey [21:14:57] <RoanKattouw> Yup [21:14:59] <RoanKattouw> box=elem.getBoundingClientRect(); [21:15:07] <RoanKattouw> return{top:box.top+(win.pageYOffset||docElem.scrollTop)-(docElem.clientTop||0),left:box.left+(win.pageXOffset||docElem.scrollLeft)-(docElem.clientLeft||0)};} [21:15:20] <RoanKattouw> (From dumping $.fn.offset into my console and picking out relevant-looking parts) [21:15:45] <mooeypoo> ha [21:16:10] <mooeypoo> so wait, it seems like the page coordinates are more straight forward, no? [21:16:16] <RoanKattouw> Anyway, one of the general principles here is that if you want to get the distance between two points, all you have to do is subtract two coordinates in the same system [21:16:18] <mooeypoo> this uses pageXOffset [21:16:23] <RoanKattouw> And it doesn't matter which system you use [21:16:34] <mooeypoo> right but in this case I need the actual position to put my little indicator in [21:16:41] <RoanKattouw> mooeypoo: That's a different pageXOffset, it's set on the window (as an alias-ish for scrollLeft), you're dealing with one that's set on an event [21:16:47] <mooeypoo> oh [21:16:58] <RoanKattouw> Right but your indicator position is a relative position, relative to the draggable container or its offset parent or whatever [21:17:01] <mooeypoo> because why *not* make another *different* offset coordinates. [21:17:12] <mooeypoo> right [21:17:15] <mooeypoo> that's true [21:17:18] <RoanKattouw> And a "relative coordinate" is really the same as "distance" [21:17:23] <mooeypoo> Indeed [21:17:35] <RoanKattouw> "Position of A relative to B" === A - B [21:17:36] <mooeypoo> The Roan Theory of Browser Relativity [21:17:45] <mooeypoo> Does it change if we move really fast? [21:17:51] <RoanKattouw> hahaha [21:18:17] <mooeypoo> But yeah, I see what you mean. It's just super confusing, especially when things like "offset" and "position" mean different things that are not necessarily intuitive [21:18:42] <RoanKattouw> And when the definition of the word "offset parent" includes the word "position" but not the word "offset" [21:18:53] <mooeypoo> Indeed [21:19:20] <wikibugs> 3VisualEditor / 3Editing Tools: VisualEditor: Warn if editing a citation that is used multiple times - 10https://bugzilla.wikimedia.org/73566#c1 (10Alex Monk) This is the same issue as editing the group of a non-basic citation. It's possible by simply selecting the reference, opening the menu and selecting... [21:20:42] <RoanKattouw> Ugh [21:20:50] <RoanKattouw> I was just about to ping Krinkle|detached with https://gerrit.wikimedia.org/r/174270 [21:21:01] <RoanKattouw> And while I typed my ping he left [21:21:57] <mooeypoo> That happened to me a couple of times. [21:21:58] <mooeypoo> The Krinkle Effect ? [21:22:32] <mooeypoo> We should start a study book. The Roan Theory of General Browser Relativity and the Krinkle Effect will be the first two chapters. [21:22:37] <mooeypoo> ah, food is ready. Brb. [21:32:53] <MatmaRex> TrevorP|Away: awwww poop [21:33:15] <MatmaRex> remember when we made FieldsetLayout a <label/> and wondered really hard what could go wrong and couldn't come up with anything? [21:33:54] <grrrit-wm> (03CR) 10TheDJ: [C: 04-1] "Please correct the indentation of the block." [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/98249 (owner: 10Dev-zero) [21:34:05] <grrrit-wm> (03PS8) 10Alex Monk: Follow normalisation response from category status query [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/166863 (https://bugzilla.wikimedia.org/72080) [21:34:05] <MatmaRex> well, it all goes wrong when you nest more than one <input/> inside a FieldsetLayout [21:37:25] <grrrit-wm> (03PS5) 10Alex Monk: [WIP] Make category widget items draggable [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/173360 (https://bugzilla.wikimedia.org/50809) [21:38:03] <TrevorParscal> MatmaRex: i'm back, reading your msgs [21:38:24] <TrevorParscal> hmm [21:38:29] <TrevorParscal> what does it do? [21:38:36] <TrevorParscal> which input gets focused when you click on the label? [21:38:40] <grrrit-wm> (03CR) 10jenkins-bot: [V: 04-1] [WIP] Make category widget items draggable [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/173360 (https://bugzilla.wikimedia.org/50809) (owner: 10Alex Monk) [21:38:56] <MatmaRex> TrevorParscal: the first one, it seems (but i doubt that this is well-defined) [21:39:41] <TrevorParscal> interesting [21:39:45] <MatmaRex> TrevorParscal: my use case is sticking a RadioSelectWidget into the layout, which naturally has a few radio buttons inside. (i should probably push that code, one minute) [21:39:47] <TrevorParscal> what's the case where you are nesting inputs? [21:40:26] <TrevorParscal> I thought we were talking about not having radio select, but instead having sparse radio inputs [21:40:27] <TrevorParscal> ? [21:40:53] <MatmaRex> yeah, but that sucks even more [21:41:01] <MatmaRex> you can't track the state of a lone radio input [21:41:10] <MatmaRex> since you only get events when it is selected, but not when it is unselected [21:41:36] <MatmaRex> and i figured we'd want something that would work in JS, not just PHP [21:42:10] <MatmaRex> http://i.imgur.com/K2U86Ku.png [21:42:14] <MatmaRex> let me just rebase the code [21:44:11] <MatmaRex> (it's easy to track the selection state yourself, provided you somehow "connect" the inputs together; and i even wrote most of that, and then realized i'm reimplementing SelectWidget) [21:45:54] <grrrit-wm> (03PS1) 10Bartosz Dziewoński: [WIP] Implement radio button widgets [oojs/ui] - 10https://gerrit.wikimedia.org/r/174290 (https://bugzilla.wikimedia.org/72586) [21:46:12] <MatmaRex> TrevorParscal: ^ see what happens on the demo page. (RadioSetWidget is the attempt to implement what i mentioned above, without SelectWidget) [21:47:56] <grrrit-wm> (03CR) 10jenkins-bot: [V: 04-1] [WIP] Implement radio button widgets [oojs/ui] - 10https://gerrit.wikimedia.org/r/174290 (https://bugzilla.wikimedia.org/72586) (owner: 10Bartosz Dziewoński) [21:53:08] <TrevorParscal> MatmaRex: hmm... [21:53:13] <TrevorParscal> doesn't it emit change? [21:53:27] <TrevorParscal> there are some more DOM events to use there no? [21:53:35] <MatmaRex> TrevorParscal: <input type=radio> only emits change when it is selected [21:53:47] <TrevorParscal> input, change, mouseup/keyup + timeout [21:53:47] <MatmaRex> there are no events at all emitted when it is unselected [21:54:02] <MatmaRex> (there is just the 'change' even on the other input) [21:54:05] <MatmaRex> event* [21:54:05] <TrevorParscal> dude, browsers are stupid [21:54:14] <mooeypoo> ^^ [21:54:31] <MatmaRex> heh [21:54:42] <MatmaRex> the spec is stupid, in this case [21:55:03] <TrevorParscal> so, yeah, I guess the trick with radio buttons then will be that when a select occurs, they need to be able to find the others in the group (via a $this.element.data( 'widget', this ) or something) [21:55:10] <MatmaRex> (i would like to take this opportunity to state that just about everything about the <input> tag is, in fact, stupid.) [21:55:21] <TrevorParscal> then when one is selected, it can notify others that this happene [21:55:24] <TrevorParscal> *happened [21:55:56] <MatmaRex> TrevorParscal: yeah, the only issue is that you need to link them somehow [21:56:05] <RoanKattouw> Why are we not using SelectWidget again [21:56:06] <MatmaRex> TrevorParscal: which basically means that you can't have just one checkbox [21:56:06] <RoanKattouw> ? [21:56:07] <MatmaRex> by the way [21:56:12] <MatmaRex> another fun nugget from the spec [21:56:39] <Krenair> sigh, categorywidget [21:56:40] <MatmaRex> if there is an <input type=radio> in the document / containing form, it must not be the only one with given value of 'name' attribute. [21:56:41] <TrevorParscal> MatmaRex: but, c'mon you know, like isn't it better to just use a few CSS rules and pure semantic HTML? that's totally enough, and if you can't make that work you aren't a good programmer bro! who want's another round of shots?! [21:57:18] <MatmaRex> :D [21:57:28] <TrevorParscal> (tha't's my best impression of a hipster brogrammer) [21:58:02] <MatmaRex> what also bothers me, is that the spec defines the concept of "radio button group", which is, all the radios with the same name in the same form [21:58:10] <MatmaRex> but it's not accessible from JS [21:58:16] <mooeypoo> I think you're missing a couple more "like" and "yeah, bro" in there. [21:58:30] <RoanKattouw> Well, other than $('input[type=radio][name='+ foo +']') right? [21:58:41] <RoanKattouw> Oh in the same form [21:58:49] <RoanKattouw> So $thing.closest( 'form' ).find( ... ) [21:59:03] <MatmaRex> RoanKattouw: too easy, you can have inputs outside of the form, but connected to it [21:59:06] <RoanKattouw> But yeah it's surprising there isn't some super-legacy document.all-style API for it [21:59:08] <Krenair> Hmm [21:59:09] <MatmaRex> <input form=somethingsomething> [21:59:21] <Krenair> Is oo.EventEmitter.emit supposed to just error with "<error>" ? [21:59:36] <MatmaRex> RoanKattouw: also, we don't even have a wrapping form in most cases [21:59:49] <MatmaRex> and in fact, we probably don't even have any names [21:59:50] <MatmaRex> :D [21:59:50] <Krenair> oh I made it infinite loop. hah [22:00:08] <MatmaRex> fun times [22:00:10] <MatmaRex> so, anyway [22:00:20] <RoanKattouw> MatmaRex: Wait, WHAT?! [22:00:28] <RoanKattouw> <input form=blah> ? YOU CAN DO THAT? [22:00:41] <MatmaRex> or something like that, can't remember the syntax [22:00:51] <MatmaRex> i don't guarantee that it works, but it's specced alright [22:01:22] <MatmaRex> http://www.w3.org/TR/html-markup/input.radio.html#input.radio.attrs.form [22:02:35] <MatmaRex> so. the sanest way to do radio inputs does seem to be a SelectWidget [22:03:09] <MatmaRex> and if we ever need them in a non-contiguous block, we can probably just detach the individual OptionWidgets and attach them elsewhere [22:04:11] <RoanKattouw> That doesn't work [22:04:19] <RoanKattouw> SelectWidget assumes its OptionWidgets are its descendants [22:04:29] <MatmaRex> well, then the individual <input type=radio> elements, whatever [22:04:49] <RoanKattouw> I mean that wouldn't be too hard to fix I suppose [22:04:58] <MatmaRex> probably nobody will ever need to do that anyway. [22:04:59] <RoanKattouw> The only such assumption I can think of is the use of event bubbling [22:05:06] <wikibugs> 3VisualEditor / 3Editing Tools: VisualEditor: [Regression wmf5] Have to undo twice to reverse comment creation - 10https://bugzilla.wikimedia.org/72315#c7 (10etonkovidova) Verified the fix in beta. [22:06:03] <MatmaRex> so, assuming we want this, i did it and it works; now we have the separate problem of nesting multiple inputs in one label. i wonder if you can nest labels? [22:06:22] <Krenair> mooeypoo, RoanKattouw: So one of the issues I've found with the draggable category widget stuff is that we need to remove/add the meta item [22:06:41] <MatmaRex> or we could capture the click event before it bubbles up to the label? [22:06:51] <Krenair> However, in the current system the "category meta item was inserted/remove" events are what triggers addItems/removeItems [22:06:56] <Krenair> Not the other way around [22:07:27] <RoanKattouw> Right [22:07:32] <RoanKattouw> Flipping that seems reasonable [22:07:34] <mooeypoo> Krenair, let me guess -- you're probably running into the same issue I did with extending addItems/removeItems [22:07:41] <Krenair> i think [22:07:59] <mooeypoo> which you can't comfortably do because the selectWidget is instatiated directly [22:08:02] <mooeypoo> ? [22:08:10] <Krenair> onLookupMenuItemChoose is what currently deals with the removing/inserting stuff when moving [22:09:13] <Krenair> If it already exists, remove the meta item. Then emit a newCategory event so CategoriesPage will insert the new meta item. Which will eventually trigger addItems to be called [22:09:54] <Krenair> but addItems/removeItems should do this, and is what the draggable stuff would expect [22:10:01] <RoanKattouw> I was thinking we could possibly listen to events from the draggable stuff [22:10:04] <RoanKattouw> There will be add/remove events [22:10:14] <RoanKattouw> The nasty thing there is that you have to prevent a loop [22:10:23] <RoanKattouw> and not listen to add/remove events you yourself caused to happen [22:10:35] <mooeypoo> Or we can get the SelectWidget to instnatiate like link widget does with static choice [22:10:39] <mooeypoo> like RoanKattouw suggested [22:10:45] <mooeypoo> and then you can extend addItems/removeItems [22:10:55] <RoanKattouw> mooeypoo: Does DraggableWidget emit any events, or does it just call addItems/removeItems? [22:11:00] <Krenair> I was going to say I was hoping to just fix CategoryWidget [22:11:05] <mooeypoo> it calls them directly [22:11:08] <mooeypoo> addItems [22:11:12] <Krenair> But then I realised I should probably say: I hope someone else will fix CategoryWidget :) [22:11:15] <mooeypoo> no removeItems, that's done automatically from addItems() [22:11:42] <mooeypoo> RoanKattouw, but it's the second time we run into the need to extend addItems() [22:13:43] <RoanKattouw> mooeypoo: I'm not sure he will need to [22:13:50] <RoanKattouw> And I think he's already got his own subclass anyway [22:13:59] <RoanKattouw> This is not the SearchWidget situation [22:14:07] * mooeypoo nods [22:14:13] <mooeypoo> I just looked [22:14:21] <mooeypoo> 'cause it sounded like he needs to extend addItems() again [22:14:24] <mooeypoo> but yeah, it seems to be there [22:14:41] <RoanKattouw> But ahm, maybe if DraggableElement emitted events when things were dragged, then he could listen to that? [22:14:46] <mooeypoo> Anyways, the draggable group element calls its child item elements directly [22:14:52] <RoanKattouw> The other option I see is flipping the relationship between the model and the view [22:14:57] <mooeypoo> and it uses its own addItems() and removeItems() directly [22:15:09] <RoanKattouw> Right now the widget (view) listens to model events, and a choose handler modifies the model which then causes the view to update [22:15:16] <RoanKattouw> But drag/drop modifies the view directly [22:15:19] <mooeypoo> no events. The only events triggered are fom the items themselves related to drag/drop -- and the widget emits "change" event when the order changed [22:15:26] <mooeypoo> whch btw I should probably replace with "reorder" event [22:15:35] <RoanKattouw> So integrating that properly would require making the view (i.e. addItems/removeItems) drive the model [22:15:43] <RoanKattouw> Which I'm not sure I like [22:15:45] <mooeypoo> since "change" is too broad and can eat up another event if this is used as a mixin on something that has its own change event. [22:15:49] <RoanKattouw> TrevorParscal: Can we pick your brain about this? [22:15:59] <TrevorParscal> yeah, how far shall I read back? [22:16:17] <RoanKattouw> TrevorParscal: Start with [14:14] RoanKattouw But ahm, maybe if DraggableElement emitted events when things were dragged, then he could listen to that? [22:16:24] <mooeypoo> Why is that a problem RoanKattouw ? We do the addItems() to change order a lot? [22:16:33] <RoanKattouw> TrevorParscal: There are a few relevant things above that but that's the short of it [22:17:31] <TrevorParscal> ok, first off [22:18:21] <TrevorParscal> I think that once you start using a drag/drop interface, it's no different than the add/remove/move interface that OutlineWidget has - in that we should emit events saying "the user is asking to do this" but not do anything in the view at all [22:18:47] <TrevorParscal> so, dropping will trigger an event saying "the user has dropped this item at this index" and that's it [22:18:50] <mooeypoo> It's not doing anything to the view. [22:19:01] <RoanKattouw> mooeypoo: It is, it's calling addItems()/removeItems() directly [22:19:06] <mooeypoo> Yes [22:19:09] <mooeypoo> Oh [22:19:11] <mooeypoo> well, yes. [22:19:18] <mooeypoo> But that changes the model first [22:19:21] <RoanKattouw> My reading of Trevor's suggestion is that an event describing the user's intent should be emitted [22:19:31] <mooeypoo> intent, or action? [22:19:34] <TrevorParscal> then the model can decide if that's a good idea, fix things up if needed (maybe the index needs to be modified a bit) and then the change occurs in the model, triggers a change in the view, the user sees the thing they dropped where they dropped it, job done [22:19:34] <RoanKattouw> And then let the listener to that event do what it will [22:19:43] <mooeypoo> should I emit "user has started dragging" or only after user has dropped the item? [22:19:50] <RoanKattouw> Emit when they drop it [22:20:07] <RoanKattouw> I mean the event is like "user asked to move item X to position P", so you don't know until drop [22:20:14] <mooeypoo> Okay, so that's what I do, but I am not using this event to actually do anything -- everything is done through addItems () [22:20:27] <mooeypoo> and I need to add the data (it is available, just didn't attach it to the emitted event) [22:20:50] <RoanKattouw> Yeah, so... maybe there needs to be some sort of default implementation or demo implementation that moves items in response to these events? [22:20:54] <TrevorParscal> look at OutlineWidget [22:20:56] <mooeypoo> Thing is -- addItems is meant for this, isn't it? For changing the items in a group element ? [22:21:14] <RoanKattouw> mooeypoo: Yes, but we're trying to avoid having a self-modifying view [22:21:26] <RoanKattouw> Instead, we're trying to have: [22:21:27] <mooeypoo> If that's the case, we'll need to fix *all* of GroupElement [22:21:33] <TrevorParscal> well, OutlineControlsWidget [22:22:10] <RoanKattouw> (user interacts with view) --> (view emits event with user intent) --> (controller code decides how to modify model) --> (model is modified) --> (model emits event) --> (view updates itself to match model) [22:22:30] <RoanKattouw> mooeypoo: What in GroupElement would need changing? [22:22:53] <TrevorParscal> mooeypoo: no no, it's OK for GroupElement to be changed due to model changes, we just don't want to actually change the group because of drag-and-drop, we want to use drag and drop the same way we would use OutlineControlsWidget to let the model know that the user wants to do something [22:23:14] <TrevorParscal> RoanKattouw has it right [22:23:21] <mooeypoo> RoanKattouw, if I understand correctly, you want to separate the model and view changes -- but the addItems() does both, no? it adds items to the item array *and* changes/appends them to the view [22:23:21] <TrevorParscal> mooeypoo: is that making sense at all? [22:23:24] <RoanKattouw> to let the /controller code/ know that the user wants to do something, which then /decides/ what to do to the model [22:23:33] <RoanKattouw> mooeypoo: The item array is not a model [22:23:43] <RoanKattouw> When I say "model" I mean something like "the order of the categories in the DM" [22:23:53] <RoanKattouw> Some things don't really have a model as such [22:23:55] <TrevorParscal> yeah, a dm object [22:23:56] <mooeypoo> Oh you mean the dm model itself. I was so confused. [22:23:57] <RoanKattouw> Most SelectWidgets don't [22:24:26] <TrevorParscal> but they really should [22:24:29] <grrrit-wm> (03PS1) 10Jforrester: ViewPageTarget#onSaveErrorSpamBlacklist: Use mw.language.listToText now it exists [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174301 [22:24:31] <grrrit-wm> (03PS1) 10Jforrester: ve.ce.MWTransclusionNode: Replace last use of ', ' in VE-MW with <comma-seperator> [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174302 (https://bugzilla.wikimedia.org/66894) [22:24:34] <RoanKattouw> Or at least most SelectWidgets (or even GroupElements) don't have the kind of model in which order matters [22:24:37] <TrevorParscal> when they don't, it's a corner we cut [22:24:38] <mooeypoo> right that's what I was confused about -- if we change this, should we change them all? [22:24:41] <TrevorParscal> and it causes problems later on [22:24:49] <RoanKattouw> In many cases the individual OptionWidgets have models [22:24:58] <mooeypoo> right now, as far as I understand, the draggableGroupElement acts more or less the same as GroupElement [22:25:28] <TrevorParscal> also, if you want to write code that just says "when the user drags something, modify the GroupElement" that's fine [22:25:36] <wikibugs> 3VisualEditor / 3Technical Debt: VisualEditor: Use comma-separator interface message - 10https://bugzilla.wikimedia.org/66894 (10James Forrester) a:3James Forrester [22:25:38] <TrevorParscal> you don't HAVE to write a separate model and rig events [22:26:02] <mooeypoo> it ... extends it, really. So all events that groupElement emits will come out of draggable group element. That's the source of my confusion -- if we change the draggable group element to emit events because we *generally* want the view and data to be separate, isn't that true for *all* our use of the group element? [22:26:51] <TrevorParscal> no, because usually when you are inserting and removing items in a groupElement it's because of the model [22:27:11] <mooeypoo> I'm having touble understanding why that behavior is changed between GroupElement and DraggableGroupElement in this specific aspect [22:27:14] <TrevorParscal> this is unique because we are now giving the user a direct control over the content of the model [22:27:33] <TrevorParscal> GroupElement can only be changed using a method in JavaScript [22:27:34] <mooeypoo> ... oh.. you just answered this. Gotcha. Okay, yes. GroupElement is done because of the model, but in DraggableGroupElement the *user* changes something. [22:27:38] * mooeypoo nods [22:27:41] <TrevorParscal> DraggableGroupElement can be changed with a mouse [22:27:44] <TrevorParscal> that's the difference [22:27:46] * mooeypoo nods [22:27:53] <mooeypoo> yes. Sorry for the slowness, I got confused. [22:27:55] <TrevorParscal> cool [22:27:57] <TrevorParscal> no worries at all [22:28:21] <mooeypoo> Okay, so do we "just" add events, or do we need to completely separate "addItems" out to an event listener [22:28:30] <TrevorParscal> this sort of view -> controller -> model -> view thing is often difficult to use properly [22:28:46] <TrevorParscal> we just add events [22:28:49] <RoanKattouw> mooeypoo: I'd say, move the call to addItems() that happens in response to drag to an event listener [22:29:01] <mooeypoo> But it sounds like we may need to have addItems be in an event listener that responds to a change in the model [22:29:07] <TrevorParscal> on drop, we emit an event that describes what the user tried to do, like "this item should be moved to this index" [22:29:17] <mooeypoo> which means we need to emit an event after drop, and then *wait* for the model to emit its change event to go with the "addItems()" [22:29:18] <mooeypoo> right? [22:29:35] <RoanKattouw> Well [22:29:42] <RoanKattouw> DraggableGroupElement will emit a drop event [22:29:49] <RoanKattouw> Describing what the user indicated they want [22:29:55] <RoanKattouw> Then it does nothing after that [22:30:05] <TrevorParscal> widget emits drop event, something may or may not be listening to that, that thing may or may not call addItems [22:30:08] <RoanKattouw> The code using DraggableGroupElement will receive that event and decide to do something [22:30:24] <RoanKattouw> That something will hopefully be a model modification that is what the user wanted, or close to it [22:30:39] <RoanKattouw> Then that model modification will cause an event to be emitted which eventually causes addItems/removeItems to be invoked [22:31:00] <RoanKattouw> So all you need to do in Draggable is provide that event with user intent information [22:31:05] <RoanKattouw> And your caller is responsible for the rest [22:31:07] <RoanKattouw> Make sense? [22:31:18] <mooeypoo> Hmm [22:31:30] <mooeypoo> Yes, but it will create a potential for a weird UX i think [22:31:39] <TrevorParscal> potential yes [22:31:43] <TrevorParscal> if you don't hook it up right [22:31:53] <TrevorParscal> if you do hook it up right, it will work as expected [22:31:54] <mooeypoo> So, if something happens and the order didn't get changed in the model your drag/drop will seem to work and then not [22:31:57] * mooeypoo nods [22:32:13] <TrevorParscal> mooeypoo: one sec [22:32:29] <mooeypoo> yeah, these are likely going to be bugs if they happen. It's not as intuitive to think about, but I think I get what you are saying. It makes sense. [22:32:41] <TrevorParscal> if the event doesn't trigger the item being moved, then the user will see that their drop did nothing [22:33:22] <TrevorParscal> if the event does trigger the item being moved, the user will be fooled into believing that their drop instantly moved the item - even though all it did was emit an event saying they wanted to move the item [22:33:31] <TrevorParscal> the UX is fine [22:33:37] <mooeypoo> Yeah this makes sense. I was thinking in the not-VE way of changing the order at the end (on save) which is *not* the way to go. You're right. [22:34:08] <Krenair> James_F, https://bugzilla.wikimedia.org/show_bug.cgi?id=73571 is very weird. On my wiki, mw.msg( 'lastmodifiedat' ) returns "This page was last modified on $1, at $2.". On beta, it returns "This page was last modified on $1 at $2.<br />" [22:34:10] <TrevorParscal> cool - if you aren't clear, feel free to ask more questions [22:34:20] <TrevorParscal> we are happy to help you [22:35:22] <MatmaRex> Krenair: http://en.wikipedia.beta.wmflabs.org/w/index.php?title=MediaWiki:Lastmodifiedat&action=edit [22:36:29] <Krenair> grr, imports [22:36:31] <mooeypoo> TrevorParscal, only quick question -- how do I do it for the demo? Just have a dummy model trigger the event ? [22:36:48] <RoanKattouw> mooeypoo: You could bypass having a model and have the event hook into addItems()/removeItems() [22:37:01] * mooeypoo nods [22:37:09] <mooeypoo> okay I just solved the positioning issue -- I'll work on that next [22:37:48] <mooeypoo> RoanKattouw, btw, re positioning, 99% is screen positions, except for the *actual* positioning of the separator/placeholder/indicator thing which is the offset (because position:absolute) [22:37:57] <mooeypoo> seems to work okay now [22:38:39] <RoanKattouw> Yeah that's what I was thinking, you can work in the client coordinate system until the last minute where you have to compute stuff to put in CSS [22:38:50] <RoanKattouw> And that you should be able to compute by subtracting client coords from each other [22:41:31] <Krenair> MatmaRex, although... why doesn't the <br /> show up in the server's version? [22:41:41] <grrrit-wm> (03PS7) 10Mooeypoo: [wip] Adding DraggableGroupElement and DraggableElement mixins [oojs/ui] - 10https://gerrit.wikimedia.org/r/172325 [22:41:53] <Krenair> ->text() strips it? [22:42:16] <mooeypoo> TrevorParscal, RoanKattouw, ^^ this is the positioning fix, plus allowing for vertical vs horizontal drag/drop. I put it back into [wip] so I can work on the events [22:43:04] <TrevorParscal> cool - will peep in a min [22:43:16] <mooeypoo> ... what the heck [22:43:25] <mooeypoo> something weird is happening now to the horizontal drag/drop [22:43:46] <mooeypoo> if you drag an element from one side to the other (say, item 1 after item 4) it drops a line, as if you have 2 blocks [22:43:50] <mooeypoo> this is weird. [22:44:24] <Krenair> MatmaRex, I see mw.message( 'lastmodifiedat' ).text() leaves it in... [22:45:38] <MatmaRex> Krenair: wait, and wfMessage() doesn't? that'd be pretty weird [22:46:05] <Krenair> mw.message( 'lastmodifiedat' ).text() -> "This page was last modified on $1 at $2.<br />" [22:46:17] <MatmaRex> Krenair: nah, the <br/> is there in PHP too. [22:46:39] <MatmaRex> my fix for whatever bug this is caused would be to delete the stupid local customization page [22:46:41] <MatmaRex> causing* [22:46:55] <Krenair> haha [22:47:03] <MatmaRex> i dunno if mw.message( 'lastmodifiedat' ).parse() in JS will work? it might [22:47:22] <MatmaRex> (work = keep the <br/> alone, while still sanitizing weird HTML) [22:48:31] <Krenair> could just use .html rather than .text in the js [22:50:38] <Krenair> but then we'll just have one more message for admins to add malicious stuff [22:50:59] <grrrit-wm> (03PS1) 10Krinkle: build: Clean up fileExists config [oojs/ui] - 10https://gerrit.wikimedia.org/r/174306 [23:06:43] <grrrit-wm> (03CR) 10Jforrester: "This doesn't actually error if a file in src/styles or src/themes is deleted… but neither does HEAD." [oojs/ui] - 10https://gerrit.wikimedia.org/r/174306 (owner: 10Krinkle) [23:10:38] <Krenair> MatmaRex, "This page was last modified on $1 at $2.<br />" [23:11:13] <Krenair> mediawiki.jqueryMsg: lastmodifiedat: Parse error at position 40 in input: This page was last modified on $1 at $2.<br /> [23:13:09] <MatmaRex> Krenair: hm [23:13:12] <MatmaRex> allowedHtmlElements: [ [23:13:12] <MatmaRex> 'b', [23:13:12] <MatmaRex> 'i' [23:13:12] <MatmaRex> ], [23:13:27] <MatmaRex> also, // Self-closing tags are not currently supported. [23:14:59] <Krenair> yeah, mediawiki.jqueryMsg tries to parse wikitext [23:15:48] <Krenair> among other things [23:15:54] <Krenair> Hi ChanServ [23:16:23] <Krenair> (I actually hadn't noticed ChanServ shut down) [23:18:37] <grrrit-wm> (03PS1) 10Cmcmahon: [BrowserTest] stop rubocop complaints for now [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174314 [23:18:51] <James_F> Krenair: It's rare. :-) [23:20:00] <grrrit-wm> (03CR) 10Cmcmahon: [C: 032] [BrowserTest] stop rubocop complaints for now [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174314 (owner: 10Cmcmahon) [23:21:25] <grrrit-wm> (03Merged) 10jenkins-bot: [BrowserTest] stop rubocop complaints for now [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174314 (owner: 10Cmcmahon) [23:32:51] <wikibugs> 3VisualEditor / 3MediaWiki integration: VisualEditor: After updating a page <br /> gets added after the timestamp of last edit made on that page - 10https://bugzilla.wikimedia.org/73571#c1 (10Alex Monk) This is due to some weird enwiki customisation of the message that got imported by Petrb. [23:33:26] <Krenair> MatmaRex, ignore what I said earlier about how it could be abused - it should be fine [23:37:22] <grrrit-wm> (03PS1) 10Alex Monk: Insert last modified text as HTML [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174318 (https://bugzilla.wikimedia.org/72571) [23:42:30] <grrrit-wm> (03PS2) 10Alex Monk: Insert last modified text as HTML [extensions/VisualEditor] - 10https://gerrit.wikimedia.org/r/174318 (https://bugzilla.wikimedia.org/73571) [23:42:45] <wikibugs> 3VisualEditor / 3Mobile: VisualEditor Mobile: [Regression wmf5] The wikitext editor does not show any wikitext for any page when switching from VE in test2 and Beta Labs - 10https://bugzilla.wikimedia.org/72571#c2 (10Alex Monk) 5PATC>3ASSI Oops, wrong bug number. [23:45:05] <wikibugs> 3VisualEditor / 3Mobile: VisualEditor Mobile: For long drop down menus the options are not at all accessible when the screen is set to landscape in iOS device - 10https://bugzilla.wikimedia.org/67340#c7 (10Rummana Yasmeen) Confirming that, it is still occurring. [23:46:13] <grrrit-wm> (03PS8) 10Mooeypoo: Adding DraggableGroupElement and DraggableElement mixins [oojs/ui] - 10https://gerrit.wikimedia.org/r/172325 [23:46:27] <mooeypoo> TrevorParscal, RoanKattouw that's what you meant, I hope? ^^ [23:47:17] <TrevorParscal> lookin [23:47:28] <mooeypoo> Drop emits a "reorder" event. The widget then has "reorderItem" which in the demo is triggered from inside the widget, but can be triggered outside [23:48:04] <mooeypoo> not sure how to make it listen to a specific event from the dm if that's what you meant? so, the "second stage" isn't listening to events. Wasn't entirely sure how to do that part correctly. I figured it should be triggered from outside the widget? if not, what should it listen to? [23:52:11] <grrrit-wm> (03CR) 10jenkins-bot: [V: 04-1] Adding DraggableGroupElement and DraggableElement mixins [oojs/ui] - 10https://gerrit.wikimedia.org/r/172325 (owner: 10Mooeypoo) [23:52:53] <mooeypoo> O.o eh? [23:53:22] <wikibugs> 3VisualEditor / 3ContentEditable: VisualEditor: Insert Media - "Error creating thumbnail" - 500 server error - 10https://bugzilla.wikimedia.org/73584 (10etonkovidova) 3NEW p:3Unprio s:3normal a:3None 1. In VE insert Media - enter N1814 as a criteria to search(some other searches produce server errors...