[webkit-dev] New web-facing canvas feature: "opaque" attribute

[Kenneth Russell]

On Fri, Mar 15, 2013 at 1:35 AM, Maciej Stachowiak <mjs at apple.com> wrote:

>
> On Mar 15, 2013, at 1:19 AM, Gregg Tavares <gman at google.com> wrote:
>
>
> I don't understand the opposition to "alpha"
>
> You set colors in Canvas2d with rgb or rgba. That 'a' in rgba stands for
> alpha.
> You can set a global alpha for drawing with context.globalAlpha
> You read data from getImageData that returns red, green, blue, alpha
> You write data to putImageData wit red, green, blue, alpha
>
>
> I think made the case for why "alpha: false" is confusing and ambiguous.
> It does not disable any of the above features, so far as I know.
>
> I also don't understand why if getContext("webgl", { alpha: false }) has
> been shipping for 3 years there's a need to make it different on canvas 2d?
>
>
> Is there a nontrivial amount of content depending on it? That would be
> more relevant precedent than merely being in the code base.
>

How would one determine that? Searching for "getcontext webgl alpha false"
returns many web pages on the topic, but doesn't seem to find JavaScript
source code.

For one high-profile example, Angry Birds HD for the browser uses this
flag; it uncovered a bug in the Mac OS X OpenGL driver on AMD GPUs.

-Ken



>
> Picking a different name or picking a different method (a function, an
> attribute on canvas or context) seems like bikeshedding rather than some
> functional objection.
>
>
> Besides naming, some of the suggested alternatives make a functional
> difference.
>
> Regards,
> Maciej
>
> P.S. One thing I'd like to know - if an opaque alpha/false canvas is
> itself composited (maybe CSS opacity is set on an ancestor), is it still
> somehow magically forced to ignore alpha, or does it composite as usual? If
> it's not forced opaque in the face of ancestor compositing, then I think
> that breaks the premise of using opaqueness as a trigger for subpixel
> anti-aliasing, which I thought was one of the main points of this. Note
> that opacity can change, so you can't just check CSS opacity of ancestors
> at the time of drawing into the canvas.
>
>
> _______________________________________________
> webkit-dev mailing list
> webkit-dev at lists.webkit.org
> https://lists.webkit.org/mailman/listinfo/webkit-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-dev/attachments/20130315/8dd3a1ad/attachment.html>