W3C home > Mailing lists > Public > www-style@w3.org > June 2012

Re: [css3-flexbox] Some references to "flex basis" should be replaced with "hypothetical main size"

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Tue, 5 Jun 2012 14:48:21 -0700
Message-ID: <CAAWBYDDCQYS--MRyfmY9hrranbAZCaN-pyhBkPh4WQfiWtpPWQ@mail.gmail.com>
To: Daniel Holbert <dholbert@mozilla.com>
Cc: www-style@w3.org
On Tue, Jun 5, 2012 at 2:38 PM, Daniel Holbert <dholbert@mozilla.com> wrote:
> Just noticed one issue in the flexbox spec:
>
> Section 9.7. "Resolving Flexible Lengths" tells us to do math with the
> flex basis, even though our flex basis may be a keyword like "auto" or
> "fit-content", which we can't do math with.
>
> In particular, there are three mentions of "flex basis", quoted below:
>   # If the flexibility is positive
>   [...]
>   #   Set the item's main size to its flex basis plus a fraction
>   #   of the free space proportional to the ratio.
> and...
>   # If the flexibility is negative
>   #   For every item on the line, multiply its flex shrink ratio
>   #   by its flex basis, [...]  Set the item's main size to its
>   #   flex basis minus a fraction of the free space [...]
>
> http://dev.w3.org/csswg/css3-flexbox/#resolve-flexible-lengths
>
> These mathematical operations are meaningless when the flex basis is a
> keyword like e.g. "auto", "fit-content"  (or really when it's any
> indefinite size),
>
> I think this chunk of spec really wants to refer to the "hypothetical
> main size", which is a definite length value that's derived from the
> (possibly-keyword-valued) flex basis in Section 9.2.
>
> Correct?

Correct, and fixed.

~TJ
Received on Tuesday, 5 June 2012 21:49:11 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:55 GMT