[whatwg] clarification on createRadialGradient( x0, y0, r0, x1, y1, r1 )

[Ian Hickson]

On Mon, 27 Nov 2006, Mathieu HENRI wrote:
> 
> The current specification of the createRadialGradient( x0, y0, r0, x1, 
> y1, r1 ) [1] is a bit ambiguous about the colour to use in the disc 
> defined by x0, y0, r0 when a colorStop is set for the offset 0. Should 
> the disc be transparent black or should it be filled with the color 
> defined for the offset 0 ?
> 
> It seems more sensible to fill it ( the disc defined by x0, y0, r0 ) 
> with the color set at the offset 0. And if the developers want to have a 
> 'holed' radial gradient they can easily set the first colorStop to rgba( 
> R, G, B, 0 )
> 
> [1] http://www.whatwg.org/specs/web-apps/current-work/#createradialgradientx0

I've updated the spec to match the interoperable behaviour of Safari and 
Mozilla. Apologies for the error in the spec. Please let me know if you 
see anything else wrong (in particular, I didn't test what should happen 
outside the "start" circle).

Thanks,
-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'