W3C home > Mailing lists > Public > public-fx@w3.org > January to March 2015

Re: Should filters on the root element affect the entire canvas?

From: Rik Cabanier <cabanier@gmail.com>
Date: Tue, 10 Feb 2015 14:08:41 -0800
Message-ID: <CAGN7qDAE-gzs+WuOfD5kUqONaG=2JPVZ97qfL3ac-Kj0654NQQ@mail.gmail.com>
To: Boris Zbarsky <bzbarsky@mit.edu>
Cc: "public-fx@w3.org" <public-fx@w3.org>
On Tue, Feb 10, 2015 at 12:25 PM, Boris Zbarsky <bzbarsky@mit.edu> wrote:

> Consider this document:
>
> <!DOCTYPE html>
> <html style="filter: invert(90%);">
> Some text
> </html>
>
> Note that in this document the <html> has a transparent background but a
> black text color by default in browsers.  The canvas has a white background
> by default in browsers.
>
> The spec text says:
>
>   Conceptually, any parts of the drawing are effected by filter
>   operations. This includes any content, background, borders, text
>   decoration, outline and visible scrolling mechanism of the element
>   to which the filter is applied, and those of its descendants.
>
> The canvas background is not part of the "element to which the filter is
> applied", so per the current spec this document should render with light
> gray text on a white background.
>
> This is what Firefox does.  Chrome instead seems to make the canvas
> background black.
>
> Which behavior do we want here?  What about other cases of filters being
> applied to the root element?  Do we want to effectively propagate these
> filters to the viewport, like CSS does with 'overflow' and backgrounds on
> the root element?


We ran into the same issue with 'mix-blend-mode' on the root element.
Tab updated CSS Color level 4 to include the following:

The default background of the root element must be transparent. The default
color of the canvas (the surface on which the document is painted) is
UA-dependent, but is recommended to be white, especially if the above
colorrules are used.


So, Firefox is following the spec and Chrome should change its behavior.
Received on Tuesday, 10 February 2015 22:09:09 UTC

This archive was generated by hypermail 2.3.1 : Monday, 22 June 2015 03:33:53 UTC