Re: [CSS21] Collapsing Margins

Alex Mogilevsky wrote:
> Have you changed the test recently?

> It seems the red moved since this thread started,

Yes, the test was originally invalid and I fixed it.

> The expected behavior is as follows
> 
> * "outer" has min-height of 21px * "inner" is 20px tall, therefore
> "outer" would be at most 20px tall if not for min-height, which means
> "inner" bottom margin doesn't collapse with "outer" bottom margin *
> Since "inner" bottom margin doesn't collapse with the parent, it
> contributes to the height of the parent * "outer" height therefore
> must be 100px * "green" div then is positioned at 100px

I have to agree with Bruno about this, since the 3rd step in 10.7 does 
return 21px as the height.  Since 'min-height' is used as the computed 
value of 'height' in that that step, the result is the same as if 
'height' itself had been set to 21px.  I've added 2 more tests to 
demonstrate this using 'height' instead of 'min-height'

http://lachy.id.au/dev/css/tests/adhoc/collapsing-margins-04.html
http://lachy.id.au/dev/css/tests/adhoc/collapsing-margins-05.html

> Firefox doesn't include "inner" bottom margin into "outer" height. 
> Opera and Safari are 1 pixel off. Not sure why. IE8 has correct
> layout.

Opera and Safari are off by 1px due to the same bug they show in 
collapsing-margins-01.html.

-- 
Lachlan Hunt - Opera Software
http://lachy.id.au/
http://www.opera.com/