W3C home > Mailing lists > Public > public-fx@w3.org > October to December 2012

Re: [css-compositing] blending in canvas

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Wed, 14 Nov 2012 09:04:59 -0800
Message-ID: <CAAWBYDA-t6bFM48CQGEcHiwVBLfARLnJ42PJqXzAnikEiFOJmA@mail.gmail.com>
To: Rik Cabanier <cabanier@gmail.com>
Cc: "public-fx@w3.org" <public-fx@w3.org>
On Wed, Nov 14, 2012 at 8:45 AM, Rik Cabanier <cabanier@gmail.com> wrote:
> the Compositing and Blending spec has a chapter on adding blending to the
> Canvas 2d context [1].
> Blending is currently defined by overloading the 'globalCompositeOperator'
> [2] with additional values.
>
> People have raised this as an issue because CSS split compositing and
> blending into 2 properties and collapsing them for Canvas seems
> inconsistent.
> The main reason for this difference, is that Canvas is built upon Core
> Graphics on the hood and this API set can't implement compositing and
> blending separately. Because of this, splitting the properties can cause
> combinations that are not implementable (ie source-atop + multiply).
>
>
> What do people think?
> Is it more important for the spec to be consistent, or to have a feature
> that can be consistently implemented today?

I'm fine with the current proposal.  In the future, if it becomes more
palatable to mix arbitrary compositing and blending, we can simply
change the attribute to accept up to two keywords, one from the
compositing list and one from the blending list.

~TJ
Received on Wednesday, 14 November 2012 17:05:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 14 November 2012 17:05:50 GMT