Videos cannot be edited in CX2. Switching to "Use source text" does not help, you just get the original text that is not editable.
Example from en:Spinor:
For comparison, CX1 allows editing the caption below video:
• Petar.petkovic | |
Jul 26 2018, 5:31 PM |
F24429857: Screen Shot 2018-08-01 at 8.48.09 PM.png | |
Aug 2 2018, 3:58 AM |
F24163622: cx2-adding-video.png | |
Jul 26 2018, 5:31 PM |
F24163646: cx1-editing-video-caption.png | |
Jul 26 2018, 5:31 PM |
Videos cannot be edited in CX2. Switching to "Use source text" does not help, you just get the original text that is not editable.
Example from en:Spinor:
Two issues:
Change 448112 had a related patch set uploaded (by Esanders; owner: Esanders):
[mediawiki/extensions/ContentTranslation@master] Fix inheritance of CX nodes
The above patch fixes the client issue. You probably want to fix the server issue and merge that first.
Change 448412 had a related patch set uploaded (by Santhosh; owner: Santhosh):
[mediawiki/services/cxserver@master] linreardoc: Support video content parsing
Change 448112 merged by jenkins-bot:
[mediawiki/extensions/ContentTranslation@master] Fix inheritance of CX nodes
Merged patch, that was meant for fix inheritance of nodes, now causes following error, while trying to load en:Spinor in CX2:
jQuery.Deferred exception: Cannot read property 'getAttribute' of undefined TypeError: Cannot read property 'getAttribute' of undefined at Object.ve.dm.MWBlockImageNode.static.toDataElement (http://cx2-testing.wmflabs.org/extensions/VisualEditor/modules/ve-mw/dm/nodes/ve.dm.MWBlockImageNode.js?dcb55:83:14) at Object.ve.dm.CXBlockImageNode.static.toDataElement (http://cx2-testing.wmflabs.org/extensions/ContentTranslation/modules/ve-cx/dm/ve.dm.CXBlockImageNode.js?4a683:49:67) at VeDmConverter.ve.dm.Converter.createDataElements (http://cx2-testing.wmflabs.org/extensions/VisualEditor/lib/ve/src/dm/ve.dm.Converter.js?24f36:560:36) at VeDmConverter.ve.dm.Converter.getDataFromDomSubtree (http://cx2-testing.wmflabs.org/extensions/VisualEditor/lib/ve/src/dm/ve.dm.Converter.js?24f36:864:30) at VeDmConverter.ve.dm.Converter.getDataFromDomSubtree (http://cx2-testing.wmflabs.org/extensions/VisualEditor/lib/ve/src/dm/ve.dm.Converter.js?24f36:990:13) at VeDmConverter.ve.dm.Converter.getDataFromDomSubtree (http://cx2-testing.wmflabs.org/extensions/VisualEditor/lib/ve/src/dm/ve.dm.Converter.js?24f36:990:13) at VeDmConverter.ve.dm.Converter.getModelFromDom (http://cx2-testing.wmflabs.org/extensions/VisualEditor/lib/ve/src/dm/ve.dm.Converter.js?24f36:634:14) at new MwCxDmTranslation (http://cx2-testing.wmflabs.org/extensions/ContentTranslation/modules/dm/mw.cx.dm.Translation.js?8784a:43:35) at MwCXInitTranslation.<anonymous> (http://cx2-testing.wmflabs.org/extensions/ContentTranslation/modules/mw.cx.init.Translation.js?36335:84:27) at mightThrow (http://cx2-testing.wmflabs.org/load.php?debug=true&lang=en&modules=jquery%7Cmediawiki.base&only=scripts&skin=vector&version=069i1tc:3583:29) undefined
Somehow I forgot about that comment after reading it and testing the patch later :)
cxserver patch and 448112 work nicely and users can edit videos again. However, that is true only when you start the translation from scratch. If you try opening your in-progress translation, that was started before 448112 and 448412, you'll get the same error from my last comment and in-progress translation will never load, losing user's work.
We need to support already started translations too.
Change 448412 merged by jenkins-bot:
[mediawiki/services/cxserver@master] linreardoc: Support video content parsing
Not moving to QA as the problem with already started translations is yet to be resolved.
I think we can move to QA. Given that version 2 is presented as an experimental version (of a beta feature), and the issue affects only translations started with the video element, I think it is not worth it to make backwards compatibility efforts for the few translations that may have been started in that context. Seems more practical to spend these efforts in areas with more impact.