Re: [css3-transitions] In transition-property: all, <property>, is <property> a duplicate ?

On Oct 22, 2010, at 4:20 PM, Sylvain Galineau wrote:

>> From: L. David Baron [mailto:dbaron@dbaron.org]
>> Subject: Re: [css3-transitions] In transition-property: all, <property>,
>> is <property> a duplicate ?
>> 
>> On Friday 2010-10-22 22:42 +0000, Sylvain Galineau wrote:
>>> Firefox and Opera certainly seem correct per the syntax. My only
>> concern
>>> is whether this is what authors expect, specifically in the first
>> case where
>>> one might expect a default duration for all but overridden for width.
>>> 
>>> WebKit's implementation also behaves as one would expect per the
>> duplicate
>>> rule so authors don't have to remember that all is a special case.
>> Lastly,
>>> this does allow them to set a default duration for all animatable
>> property
>>> together with a few chosen exceptions. If all is a stand-alone value,
>> there
>>> doesn't seem to be a practical way to do that.
>> 
>> Yeah, changing the spec to allow 'all' to be special within a list
>> makes sense to me.
>> 
>> I'm not sure if it makes sense for 'none', though, but I guess
>> there's no harm, and it makes things more consistent.  (Flipping a
>> property to 'none' could save authors from having to remove an entry
>> from parallel lists for the other properties.)
> 
> I agree there is no harm in making it consistent; and if the consistency 
> might even be useful, even better.
> 

I agree. Seems like the better syntax would be:

	[ none | all | IDENT ] [ none | all | IDENT ]*

Does that look right? If so, then the rule would be that a 'none' in the list would turn off transitions on all properties. The corresponding duration would be ignored. An 'all' would turn transitions on for all properties and would set their duration to the corresponding one. Specific property names later in the list would override these. That seems like the most logical rule to me.

-----
~Chris
cmarrin@apple.com

Received on Wednesday, 27 October 2010 20:17:39 UTC