Re: [CSS2.1] Clarifying 8.3.1 Collapsing Margins

On 09/15/2010 10:52 PM, L. David Baron wrote:
> On Wednesday 2010-09-15 01:34 -0700, fantasai wrote:
>>> I think this also needs to say that 'min-height' is nonzero, but
>>> that's a fix to a transitivity bug existing in the current spec.
>>
>> I'd like to address that as a separate issue. I'd like to keep this
>> one purely editorial.
>
> I'm a bit comfortable deferring that (what you filed as Issue 211),
> because right now the situation is that the spec has an internal
> contradiction, whereas with your new wording it's explicitly defined
> to the wrong behavior, because of the new text:
>     | A collapsed margin is considered adjoining to another margin if
>     | any of its component margins is adjoining to that margin.
>
> If you wanted to keep it identical, you'd need to change:  ...

The text you're suggesting to change is non-normative in the new
version. So that wouldn't actually make a difference. Which bit
in the normative text needs updating?

> Plus, the case where we want to change the behavior is actually
> broader than the case I make undefined above.  (Remove "and the
> parent box's top margin collapses with the last child's bottom
> margin" to get the full case.)

I've made that case more restrictive by using min-height: 0 as a
requirement in the "if" clause.

~fantasai

Received on Wednesday, 22 September 2010 13:41:43 UTC