W3C home > Mailing lists > Public > www-svg@w3.org > December 2012

Re: SVGDefinitionElement implementing SVGTests

From: Cameron McCormack <cam@mcc.id.au>
Date: Sat, 29 Dec 2012 13:02:54 +1100
Message-ID: <50DE4F4E.8010000@mcc.id.au>
To: "www-svg@w3.org" <www-svg@w3.org>
I made a test (attached) for various definition elements to see whether 
implementations do in fact look at the conditional processing attributes 
on them.

According to the spec, failing conditional processing attributes on 
<mask>, <clipPath> and <pattern> should have no effect.  There is 
nothing mentioned of <filter> or gradient elements, but they seems 
similar to the others, so let's assume it should be possible to 
reference them too.

In Firefox, rendering of the referencing element is disabled for all of 
these elements if they have a failing conditional processing attribute. 
  This is wrong.

In Opera, you can reference all of these elements and the effect will be 
applied.  This matches the current spec.

In Chrome, a referenced <clipPath>, <mask> or <pattern> will not be 
applied if it has failing conditional processing attributes, but a 
<filter> or a gradient element does work.


Given the differences in implementations, I wanted to see whether we are 
still happy with the conditional processing attributes having no effect 
on <mask>, <clipPath> and <pattern>, and if so whether we should 
explicitly handle <filter> and gradient elements in the same way.


test.svg
(image/svg+xml attachment: test.svg)

Received on Saturday, 29 December 2012 02:03:24 GMT

This archive was generated by hypermail 2.3.1 : Friday, 8 March 2013 15:54:53 GMT