- From: Ben Peters <Ben.Peters@microsoft.com>
- Date: Fri, 8 Aug 2014 19:03:05 +0000
- To: Johannes Wilm <johannes@fiduswriter.org>, Frederico Knabben <f.knabben@cksource.com>
- CC: Olivier F <teleclimber@gmail.com>, "public-editing-tf@w3.org" <public-editing-tf@w3.org>
- Message-ID: <679fe1a0e89549d5b46fc1655d54f241@BLUPR03MB437.namprd03.prod.outlook.com>
This is important and part of the conversation around the intention events we want to support. Once we decide on which specs we need, we can put each intention in a spec as appropriate. Personally I think this should be in Selection API. From: johanneswilm@gmail.com [mailto:johanneswilm@gmail.com] On Behalf Of Johannes Wilm Sent: Wednesday, August 6, 2014 9:45 AM To: Frederico Knabben Cc: Olivier F; public-editing-tf@w3.org Subject: Re: cursor movement Right now the Cursor cannot go between certain elements in both Firefox and Chrome. Even if one manually moves it there, it just cannot be drawn in certain places, such as between two stub elements, or here: <div contenteditable=true>|<span contenteditable=false>...</span>blablabla</div> ( | = cursor). So at the very least I think there should be a definition somewhere that the cursor can indeed go anywhere inside of contenteditable=minimal except of non-editable islands. Additionally I agree with Frederico that it would be a good idea to have a minimal cursor moving logic that can then be overridden if the editor developer chooses to do so. On Wed, Aug 6, 2014 at 5:33 PM, Frederico Knabben <f.knabben@cksource.com<mailto:f.knabben@cksource.com>> wrote: Still even in contenteditable=minimal, I assume that one of the "minimal" tasks of the browser is providing a *default* behavior for caret movements. After all, the possible combinations are mostly known and can be documented, guaranteeing a standard and expected behavior. Then, only special cases, usually result of specific features available in editors, should be handled by editors' developers through intentions. I agree that this is an important topic for standardization. On Wednesday, 6 August 2014 at 18:22, Olivier F wrote: My understanding is that cursor movement fires an "intent event" for "move caret"[1] which you can listen for and stop, thereby allowing you to place the cursor exactly where you want it based on the user's intent. The result is that we (editor developers) shouldn't have to worry about what the default browser behavior is or will be. [1] http://lists.w3.org/Archives/Public/public-editing-tf/2014Jul/0023.html On Wed, Aug 6, 2014 at 4:47 AM, Johannes Wilm <johannes@fiduswriter.org<mailto:johannes@fiduswriter.org>> wrote: Hey, it is my understanding that both in the case of the traditional contenteditable and contenteditable=minimal, the movement of the cursor will still be handled by the browser, at least in the case of using the arrow keys to move around. I am thinking of things when the cursor should move across "complex elements" and how it should treat SVGs, IMGs, and CANVAS (and other stub) elements. It may seem obvious that the cursor should just jump to the other side of such elements when using left/right and not inside of it when going up and down. But it is my understanding that browser makers feel that this is not an area they will fix unless there is a specification that clearly states how the cursor is supposed to move. So my question is: What document will be covering this in the future? It's not really covered by selection given that it's only relevant for contenteditable, correct? -- Johannes Wilm Fidus Writer http://www.fiduswriter.org<http://www.fiduswriter.org/> -- Johannes Wilm Fidus Writer http://www.fiduswriter.org<http://www.fiduswriter.org/>
Received on Friday, 8 August 2014 19:03:38 UTC