RE: [css3-transitions] under what transition-duration/delay conditions should there be a transition?



> 
> [L. David Baron:]
> >
> > In http://lists.w3.org/Archives/Public/www-style/2012Mar/0013.html

> > we agreed to accept my proposal in
> > http://lists.w3.org/Archives/Public/www-style/2012Feb/1083.html for
> > https://www.w3.org/Bugs/Public/show_bug.cgi?id=15838 :  that is, we
> > agreed that when transition-duration and transition-delay are both 0s
> > that there should be no transition (and thus no events).
> >
> > However, we didn't agree whether that was the only case.  I realized
> > that there are a set of other cases where there probably shouldn't be
> > a
> > transition:  in particular, when transition delay is as negative or
> > more negative than transition duration is positive.  In these cases
> > the transition end event would have a firing time at or before the
> > time the transition started, which suggests to me that maybe it
> > shouldn't fire at all.
> >
> > Thus, I've written:
> >   # When the computed value of a property changes, implementations
> >   # must start transitions based on the relevant item (see the
> >   # definition of ‘transition-property’) in the computed value of
> >   # ‘transition-property’. Corresponding to this item there are
> >   # values of ‘transition-duration’ and ‘transition-delay’ (see the
> >   # rules on matching lists). Define the combined duration of the
> >   # transition as the sum of max(‘transition-duration’, ‘0s’) and
> >   # ‘transition-delay’. When the combined duration is greater than
> >   # ‘0s’, then a transition starts based on the values of
> >   # ‘transition-duration’, ‘transition-delay’, and
> >   # ‘transition-timing-function’; in other cases transitions do not
> >   # occur.
> >
> > Does this seem reasonable to others?
> >
> My apologies for catching this so late; I agree with the resolution and
> had only one minor editorial question: is the "max(‘transition-duration’,
> ‘0s’)"
> expression needed in this prose given that transition-duration explicitly
> maps negative <time> values to 0s? Or is this just an editorial reminder
> to the reader?
It looks like I also missed another email on this specific clamping [1].
This does not seem resolved in either spec; I've opened a bug against 
Animations to track this [2].

[1] http://lists.w3.org/Archives/Public/www-style/2012Mar/0591.html
[2] https://www.w3.org/Bugs/Public/show_bug.cgi?id=16806

Received on Friday, 20 April 2012 00:50:24 UTC