W3C home > Mailing lists > Public > www-svg@w3.org > September 2010

RE: pointer-events test case

From: Kevin Ar18 <kevinar18@hotmail.com>
Date: Thu, 9 Sep 2010 17:34:19 -0400
Message-ID: <SNT110-W27FB5CA394E6663A61EAD4AA730@phx.gbl>
To: <jeremie.patonnier@gmail.com>, <www-svg@w3.org>

It seems like the the real question is whether pointer-events should inherit if not explicitly told to?
 
Here goes:
<g pointer-events="none">
    <rect x="20" y="20" width="200" height="200" pointer-events="auto" />
</g>
 
1. According to CSS: http://www.w3.org/TR/CSS2/syndata.html#illegalvalues 
pointer-events="auto" will be ignored as though it doesn't exist.
 
 
2. Result:
<g pointer-events="none">
    <rect x="20" y="20" width="200" height="200" />
</g>
 
3. Question: what pointer-events property does rect take on?  inherit "none"? or use default "visiblePainted"?
4. So I guess your real question is whether pointer-events should be inherited even if not explicitly declared to "inherit"... or if they should be "visiblePainted" instead?  Would that sum it up better?
 
 
Someone else should be able to answer that ... as I'm not 100% sure.
 

 


Date: Thu, 9 Sep 2010 22:41:26 +0200
From: jeremie.patonnier@gmail.com
To: www-svg@w3.org
Subject: pointer-events test case

Hello,

Here is a test case for some interactions with the pointer-events property

This test case show how implementation handled the pointer-events property when an element is nested in another element with a different pointer-events property value.

At that point, my test show that Opera has a problem when an element with an invalid pointer-events' value is wrap by an element with the pointer-events' value 'none'. According to the CSS 2 specification, Opera should ignore the invalid value and therefore use the default value for pointer-events which is "visiblePainted". It seams that Opera act like the pointer-events' value were "inherit" and does not fire pointer events on the child element.

Firefox and Chrome handle it that way.

Could someone explain to me how is it possible to turn that test case into a formal test case for the SVG spec?

Thanks
Jeremie Patonnier

 		 	   		  
Received on Thursday, 9 September 2010 21:34:52 GMT

This archive was generated by hypermail 2.3.1 : Friday, 8 March 2013 15:54:46 GMT