Re: [Selection] Should selection.getRangeAt return a clone or a reference?

2015-01-27 Thread Koji Ishii
So...we discussed 31 e-mails, time to try to reach a consensus? Please keep in mind, it's too obvious to everyone but you don't have to agree to build a consensus. We can build a consensus if all can live with single conclusion. 3 proposals so far: Proposal A: Leave it undefined. If it's not caus

Re: [Selection] Support of Multiple Selection (was: Should selection.getRangeAt return a clone or a reference?)

2015-01-27 Thread Koji Ishii
I thought a bit of the same thing as Tab when I saw the visual focus discussion in CSS WG, but my current conclusion is not to think the two the same thing. Focus may be possible, but selections is a different thing. It's true that you could use multi-range selections to select in visual order. Bu

Re: [Selection] Should selection.getRangeAt return a clone or a reference?

2015-01-24 Thread Koji Ishii
On Sat, Jan 24, 2015 at 9:52 PM, Olli Pettay wrote: > > Right now the liveness doesn't really cause issues, since only some UAs > support it. > But that doesn't mean getRangeAt should return cloned ranges. > Adding another getRange*At would just pollute the API. > > The more I think this, the more

Re: [Selection] Should selection.getRangeAt return a clone or a reference?

2015-01-23 Thread Koji Ishii
On Thu, Jan 22, 2015 at 12:20 AM, Mats Palmgren wrote: > >> If we really want authors to have convenience methods like >> setStartBefore() on Selection, we could add them to Selection. > > Selection methods wouldn't provide the same functionality though. > Selection.setStart* would presumably be e

Re: [Selection] Should selection.getRangeAt return a clone or a reference?

2015-01-13 Thread Koji Ishii
As in the thread Ben split, I think it's just Aryeh omitted "in the current spec focus". Visual Studio supported multiple selections in version 11 or so. MS Word, I don't remember, but probably version 9 or 10. I agree that it's a great feature, but also agree with Aryeh and Ben that we have a lot

Re: [Selection] Should selection.getRangeAt return a clone or a reference?

2015-01-09 Thread Koji Ishii
w arbitrary ranges, because > the author could call setStart() on the returned range with any value > you want, and they will expect that the range's new start will be > exactly what they set it to.) > > On Wed, Jan 7, 2015 at 12:08 PM, Koji Ishii wrote: >> I also guess t

Re: [Selection] Should selection.getRangeAt return a clone or a reference?

2015-01-07 Thread Koji Ishii
While I agree that it's nice, I have mild preference to return a clone. As Olii said, changing from clone to live would involve quite a bit of code. I feel sorry for already implemented people though. I also guess that we need to ask more work to the spec editor to support the liveness, such as: -