Re: [CSSOM] Interaction of getBoundingClientRect/getClientRects with transforms

On Tue, Aug 31, 2010 at 9:19 AM, Simon Fraser <smfr@me.com> wrote:
> On Aug 31, 2010, at 8:52 am, Brad Kemper wrote:
>> It seems unexpected to me. I expect it to act more like position:relative. Can you explain why it makes more sense, and what use case would benefit by doing the child inlines separately?
>
> Position: relative only lets you offset the boxes horizontally or vertically, so I don't think it informs on this issue. What do you expect to happen when a split <span> has a rotation transform?

Indeed; translate()s should act indistinguishably from relpos whether
you use multiple boxes or one.  It's just the other transformations
that show the difference.

I think that transforming each box individually is the right thing to
do here, particularly given Boris' example of mixed-direction text.
An author that dislikes the effect can transform a wrapper element
instead.

~TJ

Received on Tuesday, 31 August 2010 16:30:55 UTC