Qt SVG renderer

[Matthew Woehlke]

Tor Arne Vestbø wrote:
> Matthew Woehlke wrote:
>> As Rafael said, I'm a bit surprised that WebKit seems to be *worse* in 
>> several ways, it looks like maybe it does not support object opacity? 
> 
> That was my thinking too. Remember though that this is using WebKit from 
> Qt 4.4. I tried a couple of them in WebKit trunk (which Qt 4.5 will be 
> based on), and they look a lot better. I'll try to add an entry in the 
> chart for trunk when I get the chance.

Right, I'm not familiar enough with webkit to have any expectations how 
much it's changed vs. 4.4.

>> I think it would be nice if your "compare" frame was gradiated so that 
>> the dithering artifacts aren't so visible (after all, it's acceptable 
>> if QtSvg and librsvg use different dithering algorithms!).
> 
> I agree. The reason the comparison is b/w is because these results come 
> from our internal test system, which was built for regression testing of 
>  pixel perfect drawing. I will see if I can hack something together for 
> this specific case.

Makes sense, but as mentioned, I'd say the test program shouldn't demand 
perfection, but rather limit permitted deviation. I'll grant that there 
may be a legitimate reason to use the exact same dithering as librsvg, 
but I consider support of blur far more important than the exact 
dithering pattern used when drawing gradients, which seems to be what a 
lot of the artifacts correspond to.

> I'm going on a two week sick leave now though, so feel free to enhance 
> on these results.

Is the code for the test program available somewhere?

-- 
Matthew
Did you hear about the pig that makes footwear for cows? He's a real moo 
shoe pork.