Re: [CSSOM] Revisiting transforms and getBoundingClientRect()

On Thu, Sep 8, 2011 at 10:26 AM, Boris Zbarsky <bzbarsky@mit.edu> wrote:
> On 9/8/11 1:03 PM, Erik Arvidsson wrote:
>>
>> On Wed, Sep 7, 2011 at 22:34, Robert O'Callahan<robert@ocallahan.org>
>>  wrote:
>>>
>>> It would be difficult to expose a synchronous "onlayout" event to Web
>>> content safely, and it would be pointless to make it asynchronous.
>>
>> I don't think it would be pointless to have an async event that would
>> be dispatched when we know that the layout is non dirty at some point
>> before the next paint. At this point getting the layout rects would be
>> cheap.
>
> This causes problems if the event handler then makes layout dirty, no?

Can we forbid this? That is, any mutation in the event handler throws.
 Then it could be sync.

(BTW I apologize to roc for hijacking his getQuad post).

jjb

Received on Thursday, 8 September 2011 17:51:21 UTC