W3C home > Mailing lists > Public > www-style@w3.org > June 2013

Re: transition using matrix to/from zero values can not be decomposed

From: Rik Cabanier <cabanier@gmail.com>
Date: Sat, 22 Jun 2013 19:58:56 -0700
Message-ID: <CAGN7qDAJ4DWit50CuG6gFwReGH2Lo4OQ34jntSSZ29m6okHwyA@mail.gmail.com>
To: Shane Stephens <shans@google.com>
Cc: Jon Ronnenberg <jon.ronnenberg@gmail.com>, www-style list <www-style@w3.org>
yes,
in general it's much better to use the other transform function than a
matrix. Matrix interpolation was added for completeness but you shouldn't
really be using it.

The current matrix decomposition logic also only matches Chrome at the
moment. Both Safari and Firefox will use different interpolation logic so
your animation might look different.

On Sat, Jun 22, 2013 at 7:32 PM, Shane Stephens <shans@google.com> wrote:

>
> On Jun 23, 2013 11:48 AM, "Jon Ronnenberg" <jon.ronnenberg@gmail.com>
> wrote:
> >
> > According to Decomposing the Matrix, transitions from matrix(1, 0, 0, 0,
> 0, 0) to matrix(1, 0, 0, 1, 0, 0) is not possible. This simple transform is
> saying an element is zero units high and will become it's full height.
> Imagine a slide-down animation.
> >
> > Due to this, browser vendors can not "fix" this without violating the
> specs and web authors have to work around this limitation by using
> max-height transitions with guessed maximum height or revert to use
> javascript.
>
> That may be true with regards to browser vendors, but web authors can use
> the much simpler work around of transitioning scaleY.
>
> Cheers,
>     - Shane
>
> > I'm not sure if this is a documentation bug or a spec omission. But I
> sincerely hope that this is not the intended behavior.
> >
> > It should be said that all browsers that supports transform transitions
> can animate from matrix(1, 0, 0, 0.0001, 0, 0) to matrix(1, 0, 0, 1, 0, 0).
> Webkit does however leave a 1px artifact instead of hiding the element
> completely.
> >
> > Cheers, Jon
>
Received on Sunday, 23 June 2013 02:59:23 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:39:12 UTC