RE: [css3-backgrounds] Example XV inconsistent with prose of section 3.6

What does
 Background-position: left 17px center;
serialize out as?

"left 17px 50%"?

That seems to collide with
"If three or four values are given, then each <percentage> or<length> represents an offset and must be preceded by a keyword, which specifies from which edge the offset is given."

Or are you saying center sometimes becomes 50% when serialized out, and sometimes doesn't?

Confused again,
- Brian


> -----Original Message-----
> From: fantasai [mailto:fantasai.lists@inkedblade.net]
> Sent: Tuesday, September 14, 2010 12:09 AM
> To: Tab Atkins Jr.
> Cc: Sylvain Galineau; Brian Manthos; www-style@w3.org
> Subject: Re: [css3-backgrounds] Example XV inconsistent with prose of
> section 3.6
> 
> On 09/13/2010 06:22 PM, Tab Atkins Jr. wrote:
> > On Mon, Sep 13, 2010 at 6:15 PM, Sylvain Galineau wrote:
> >>>>> 'center' and '50%' are equivalent in the context of a background-
> position.
> >>>>
> >>>> For rendering yes, for OM no...
> >>>
> >>> Ah, right.  Yeah, then that example is incorrect.
> >>>
> >> Why ? This module and its examples are not about the OM so I'm not
> >> clear on why we can deem this example incorrect on OM grounds. But
> >> beyond that why is 50% not equivalent to center in the OM ?
> >
> > Well, it's not specified in any significant way, but it would be
> > expected that saying "center" would make the current OM return
> > "center", and saying "50%" would make the current OM return "50%".
> 
> IIRC, Anne was going to define OM and serialization bits consistent with the
> Computed Values line wherever that gave enough detail. In this particular
> instance, there is enough detail: keywords compute to percentage values.
>    http://www.w3.org/TR/CSS21/colors.html#propdef-background-position

>    http://www.w3.org/TR/css3-background/#the-background-position

> 
> > The Values API that Anne's been floating would differentiate between
> > the two as well, claiming the type of the value to be "keyword" and
> > "percentage" respectively.  (Like it allows you to differentiate
> > between "red" and "#f00".)
> 
> Agreed with Sylvain on this one.
> 
> ~fantasai

Received on Tuesday, 14 September 2010 09:18:31 UTC