Re: [css2.1] Propagation of CSS direction property to the initial containing block

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:33 GMT

On 10/26/2010 11:42 AM, David Hyatt wrote:
> The CSS 2.1 spec states in section 10.1.1:
>
> "The 'direction' property of the initial containing block is the same as
> for the root element."
>
> This is incomplete and doesn't really capture what browsers do. For
> example, IE propagates the direction from the body to the ICB, and it
> even ignores the directionality specified on the root element (replacing
> it with the body's direction).
>
> I'd like to propose that the spec mention that you can propagate from
> the body.
>
> The behavior I just implemented in WebKit [1] propagates from the body
> to the root element if no explicit direction is set on the root element.
> It then propagates the direction from the root element to the ICB. This
> behavior is consistent with how backgrounds propagate (and is also
> similar to how overflow applies to the viewport as well).

Is this necessary? If it's not a significant web-compat issue, then
I would rather not do this. Propagation from <BODY> is a hack to
handle old content, not a practice we should be using going forward.

 > Note that whatever we decide should apply to writing-mode also.

Disagree, for reason above.

~fantasai