On Wed, Feb 24, 2010 at 1:00 PM, Philippe Wittenbergh <[email protected]> wrote: > > On Feb 24, 2010, at 7:15 PM, Bruno Fassino wrote: > >> There are probably other cases with different "interpretations". For >> example I see here >> http://www.brunildo.org/test/Op_top_auto.html >> that Gecko and Webkit differ. The a.p. box has a margin-top. When >> calculating the top:auto position Webkit makes the a.p. margin-top to >> collapse with a previous margin-bottom, Gecko does not. >> Again, I don't think we can "strictly" say that one is correct, and >> the other not. > > Hmm. funky. > We have 8.3.1 Collapsing margins that says: Margins of absolutely positioned > boxes do not collapse (not even with their in-flow children) > > WebKit is apparently ignoring this. Or partly ignoring this. > I'm not sure if this is how Gecko processes the information/requirements from > the stylesheet: > 1. place the element as if it were static --> top margin collapses > 2. element becomes absolute positioned --> taken out of the flow. > (2nd pass) > 3. we need to recompute the top-margin now to take into account the > constrains of 8.3.1 > > </slightly confused> > or is it the spec ?
In that case I "prefer" the WebKit behavior (assuming the intention of the spec is to "preserve" the static position). In the "static position" there is collapsing, and it determines the position of the top margin edge of the box. When the box "becomes" a.p. (out of the flow) there is no need to re-invoke 8.3.1, the value of "top" is already determined. To explain WebKit behavior there no need to say that it makes the a.p. top margin to collapse with anything. Of course (sorry if I repeat myself here :-)), the spec does not seem to mandate anything with that sentence: "But rather than actually calculating the dimensions of that hypothetical box, user agents are free to make a guess at its probable position." I cannot even say to fully understand it (in the "rather than" part). But that's probably a question for www-style... Regards, Bruno -- Bruno Fassino http://www.brunildo.org/test ______________________________________________________________________ css-discuss [[email protected]] http://www.css-discuss.org/mailman/listinfo/css-d List wiki/FAQ -- http://css-discuss.incutio.com/ List policies -- http://css-discuss.org/policies.html Supported by evolt.org -- http://www.evolt.org/help_support_evolt/
