RE: SVGt 1.2 Tests: animate-elem-41-t fill rule animation is incorrect and error in ref image

Hi,

Note that the test animate-elem-41-t, also use a single "to" animation
for the start (fill-rule) test case.
I'm not sure if I reported this one or not :-(

Julien 

> -----Original Message-----
> From: www-svg-request@w3.org [mailto:www-svg-request@w3.org] 
> On Behalf Of Dr. Olaf Hoffmann
> Sent: 06 September 2008 14:54
> To: www-svg@w3.org; Kalle Raita
> Subject: Re: SVGt 1.2 Tests: animate-elem-41-t fill rule 
> animation is incorrect and error in ref image
> 
> 
> Hello,
> 
> before I forget it, I'd like to mention, that it turned out 
> http://lists.w3.org/Archives/Public/www-smil/2008JulSep/0011.html
> , that the behaviour for discrete
> to-animations is indeed already defined in the old SMIL 
> animation recommendation 
> http://www.w3.org/TR/2001/REC-smil-animation-20010904/#comma-wspBNF
> , and therefore for SVG1.1 in the same way as for SVGT1.2, 
> SMIL3 and SMIL2.
> 
> This especially means, that already the SVG1.1 test is wrong, 
> the SVGT1.2 test is derived from and that user-agents like 
> Opera, Batik/Squiggle, Adobe plugin have currently the wrong 
> timing for discrete to-animations implemented.
> 
> Maybe this indicates, that this feature should be tested in 
> the SVG test suite, with calcMode explicitly set to discrete 
> and implicitely both by the attributeName value, by the 
> to-value and by the underlying value (if they are not 
> interpolable themselves, but the attribute or property has 
> interpolabe values to, as for example fill and stroke: 'none' etc)...
> 
> Having enough detailled tests for to-animation, maybe one day 
> to-animations really can be used by auhors, if most viewers 
> support them as specified ;o)
> 
> 
> Olaf
> 
> 
>

Received on Saturday, 6 September 2008 19:00:44 UTC