Re: [geometry] Remove liveness (was: Re: [blink-dev] Re: Intent to Implement: Geometry Interfaces)

If the objects are no longer live, is there a reason to keep around the
DOMRect/DOMRectReadOnly distinction?

On Wed, Nov 18, 2015 at 2:33 PM, Simon Pieters <simonp@opera.com> wrote:

> On Tue, 17 Nov 2015 11:48:02 +0100, Robert O'Callahan <
> robert@ocallahan.org> wrote:
>
> On Tue, Nov 17, 2015 at 11:35 PM, Simon Pieters <simonp@opera.com> wrote:
>>
>> I thought this had been discussed in the past, but I can't find anything
>>> now. Only
>>> https://lists.w3.org/Archives/Public/public-fx/2014JanMar/0012.html
>>> which
>>> isn't asking for non-liveness.
>>>
>>> The only live object is DOMQuad#bounds, I believe.
>>>
>>> https://drafts.fxtf.org/geometry/#associated-bounding-rectangle
>>>
>>> What are the pros and cons for live vs non-live for this object?
>>>
>>>
>> If it's not live, would you have an attribute that returns a new object
>> every time, or a method that returns a new object every time, or an
>> attribute (or method) that returns a new object every time the DOMQuad
>> changes, or what?
>>
>
> I don't know. Maybe we could remove it and add a .fromQuad() static method
> on DOMRect/DOMRectReadOnly, that gives the bounding rectangle (new object
> every time)?
>
>
> --
> Simon Pieters
> Opera Software
>

Received on Wednesday, 18 November 2015 22:42:15 UTC