Re: [css-background] Add the *-x/y longhands to appropriate properties?

On Tue, Apr 22, 2014 at 5:36 PM, Brad Kemper <brad.kemper@gmail.com> wrote:
>> On Apr 21, 2014, at 6:45 PM, "Tab Atkins Jr." <jackalmage@gmail.com> wrote:
>>> On Mon, Apr 21, 2014 at 6:41 PM, Brad Kemper <brad.kemper@gmail.com> wrote:
>>> On Apr 21, 2014, at 10:44 AM, "Tab Atkins Jr." <jackalmage@gmail.com> wrote:
>>>>> OK, so I guess 'bg-pos-y:start' would get mapped to 'bg-pos-y:top' and to 'bg-pos-block:start' once you knew the mode was horizontal-tb. Right?
>>>>
>>>> Well, it would just get mapped to "bg-pos-y: top;".  No need to map it
>>>> into the other property, since they're aliases.
>>>
>>> Not sure if you are sitting hairs between "mapping" vs. "aliasing",
>>
>> Splitting hairs between computed values and aliasing.
>>
>>> but my point was that it is only an alias once you know the mode is horizontal-tb, no? Otherwise, if(!horizontal) y != block.
>>
>> "bg-pos-y: start" isn't a shorthand for anything else, but it does
>> turn into "bg-pos-y: top" (or bottom) depending on w-m/direction.
>>
>> bg-pos-y is then also an alias for either bg-pos-inline or
>> bg-pos-block, depending on w-m/direction.
>
> I think I understand you correctly: an alias is like a shorthand that's no shorter, and happens in the cascaded value time (in a second cascade after the initial cascade to determine writing mode). But for "bg-pos-y: start" to get turned into "bg-pos-y: top", that's just a normalization of the value that happens at computed value time, and by then the wm is well known anyway.

Yes, exactly.  (An alias is actually *literally* a shorthand, in terms
of mechanics, but you get the point.)

~TJ