CDR: pointer event description vs. z-order

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

CDR: pointer event description vs. z-order

Maciej Stachowiak


"The target element for pointer device events is established by first  
determining the most deeply nested child document which has content  
that intersects with the geometric (x,y) location of the event"

- This excludes z-ordering. Therefore it seems to claim that, for  
instance, a click on a portion of a parent document that overlaps  
included child content should go through to the child content. This  
is contrary to behavior of many current UAs.

Regards,
Maciej


Reply | Threaded
Open this post in threaded view
|

Re: CDR: pointer event description vs. z-order

L. David Baron
On Monday 2006-01-02 01:49 -0800, Maciej Stachowiak wrote:
> "The target element for pointer device events is established by first  
> determining the most deeply nested child document which has content  
> that intersects with the geometric (x,y) location of the event"
>
> - This excludes z-ordering. Therefore it seems to claim that, for  
> instance, a click on a portion of a parent document that overlaps  
> included child content should go through to the child content. This  
> is contrary to behavior of many current UAs.

In our face-to-face meeting, the group decided to change the quoted
sentence to:

  The target element for pointer device events is the topmost object in
  the paint order at the pointer location.


A slight addendum of my own, not from the group:  Now that I notice we
were editing based on only the first part of the quoted sentence;
preserving the more detailed part of the end of the sentence "that
intersects with the geometric (x,y) location of the event" might be
preferable, but I'd consider the difference editorial.  Although now
that I'm looking at that I'm not sure whether this sentence is trying to
determine which document gets the events or which element gets the
events, and that distinction should probably be clear.

I'd also note that this doesn't resolve any differences in the way SVG
and HTML+CSS define paint order (particularly with whether
'visibility:hidden' elements receive events).

-David

--
L. David Baron                                <URL: http://dbaron.org/ >
           Technical Lead, Layout & CSS, Mozilla Corporation

attachment0 (196 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: CDR: pointer event description vs. z-order

Maciej Stachowiak


On Jan 29, 2006, at 5:14 PM, L. David Baron wrote:

> On Monday 2006-01-02 01:49 -0800, Maciej Stachowiak wrote:
>> "The target element for pointer device events is established by first
>> determining the most deeply nested child document which has content
>> that intersects with the geometric (x,y) location of the event"
>>
>> - This excludes z-ordering. Therefore it seems to claim that, for
>> instance, a click on a portion of a parent document that overlaps
>> included child content should go through to the child content. This
>> is contrary to behavior of many current UAs.
>
> In our face-to-face meeting, the group decided to change the quoted
> sentence to:
>
>   The target element for pointer device events is the topmost  
> object in
>   the paint order at the pointer location.

This satisfies my concern, thanks.

> I'd also note that this doesn't resolve any differences in the way SVG
> and HTML+CSS define paint order (particularly with whether
> 'visibility:hidden' elements receive events).

Does the CDF working group plan to tackle this issue in any of the  
specs?

Regards,
Maciej


Reply | Threaded
Open this post in threaded view
|

RE: CDR: pointer event description vs. z-order

Jon Ferraiolo


...snip...

> I'd also note that this doesn't resolve any differences in the way SVG
> and HTML+CSS define paint order (particularly with whether
> 'visibility:hidden' elements receive events).

Does the CDF working group plan to tackle this issue in any of the  
specs?

Regards,
Maciej

----------

I think that processing model issues around paint order and event
targeting are language-specific. I don't see a need to unify HTML+CSS
and SVG in this area. For the SVG viewport, look to the SVG spec for how
to deal with paint order and event targeting. For the areas outside of
the SVG viewport, look to the CSS spec. The only conflict is when there
is a CSS region that has a higher z-index that the html:object tag that
references some SVG content (or inline SVG instead of the html:object),
but I think in that case it is fairly obvious that the CSS region with
the higher z-index would be above the SVG content.

Jon



Reply | Threaded
Open this post in threaded view
|

Re: CDR: pointer event description vs. z-order

Maciej Stachowiak


On Jan 30, 2006, at 7:35 PM, Jon Ferraiolo wrote:

>
> ...snip...
>
>> I'd also note that this doesn't resolve any differences in the way  
>> SVG
>> and HTML+CSS define paint order (particularly with whether
>> 'visibility:hidden' elements receive events).
>
> Does the CDF working group plan to tackle this issue in any of the
> specs?
>
> Regards,
> Maciej
>
> ----------
>
> I think that processing model issues around paint order and event
> targeting are language-specific.

If this is the case, then the CDF working group should be careful to  
avoid any normative references to event targeting and paint order.  
However, the current CDR section is purely informative, so whatever  
it says had better be correct per other specs.

> I don't see a need to unify HTML+CSS
> and SVG in this area. For the SVG viewport, look to the SVG spec  
> for how
> to deal with paint order and event targeting. For the areas outside of
> the SVG viewport, look to the CSS spec. The only conflict is when  
> there
> is a CSS region that has a higher z-index that the html:object tag  
> that
> references some SVG content (or inline SVG instead of the  
> html:object),
> but I think in that case it is fairly obvious that the CSS region with
> the higher z-index would be above the SVG content.

It might be good to state this explicitly in some spec (whether CSS  
or SVG or CDR or WICD) instead of just letting it remain implicit.  
Even if it is believed to already be required, it would be useful as  
an informative note.

Regards,
Maciej