Re: [matrix][cssom-view] DOMPoint, DOMPointLiteral definitions

On Wed, 25 Sep 2013 04:23:42 +0200, Tab Atkins Jr. <jackalmage@gmail.com>  
wrote:

> On Tue, Sep 24, 2013 at 5:46 PM, Robert O'Callahan  
> <robert@ocallahan.org> wrote:
>> On Wed, Sep 25, 2013 at 11:52 AM, Tab Atkins Jr. <jackalmage@gmail.com>
>> wrote:
>>> On Tue, Sep 24, 2013 at 4:34 PM, Robert O'Callahan  
>>> <robert@ocallahan.org>
>>> wrote:
>>> > Who said anything about DOMRect inheriting from DOMQuad? It  
>>> shouldn't,
>>> > IMHO.
>>> > Constructing a DOMQuad from a DOMRect is good enough.
>>>
>>> I think I did - anything that can take a DOMQuad should be able to
>>> take a DOMRect as well, and user-level code should be able to interact
>>> with multiple Quads/Rects in a consistent way (that is, as Quads).  We
>>> could maybe do this latter by just letting the DOMQuad constructor
>>> take a DOMRect (and vice versa, to obtain the aligned bounding rect?).
>>
>>
>> I feel like things are spinning out of control here. Should we make  
>> DOMPoint
>> inherit from DOMRect since a point is just a degenerate rectangle?
>>
>> APIs that take quads (of which we have zero right now) can take  
>> DOMRects as
>> well via overloading or union types. That, plus a DOMQuad constructor  
>> that
>> takes a DOMRect, and DOMQuad.bounds, should be enough to make things
>> arbitrarily convenient.
>
> Sure, I'm fine with that.  (And DOMQuad.bounds is a readonly DOMRect,  
> right?)

OK, I've removed the inheritance.

https://dvcs.w3.org/hg/csswg/rev/1720a4a9f3bd

-- 
Simon Pieters
Opera Software

Received on Wednesday, 25 September 2013 11:58:45 UTC