12945 - vertical auto margins don't work on absolutely positioned elements

Status: VERIFIED INVALID

(Core :: Layout, defect, P3)

Description

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

DESCRIPTION:  Vertical auto margins don't do as they should for absolutely
positioned elements.  I don't think you're following many of the rules in CSS2
10.6.4.

STEPS TO REPRODUCE:  Load attached test case.

ACTUAL RESULTS:  Three elements are at top of page.

EXPECTED RESULTS:  Three elements should be as described (top, center, bottom)
vertically.  They currently are in the correct horizontal positions.

DOES NOT WORK CORRECTLY ON:
 * Linux, viewer, 1999-08-30-01-M10

Comment 1

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Attachment: testcase described in bug

Comment 2

[troy]

I'm slightly offended by the quip "I don't think you're following many of the
rules in CSS2 10.6.4." :-)

There are six rules. In order for your statement to be correct I would guess
that I would have to follow less than two of the rules.

I think I support at least four rules, possibly five. I'm pretty sure I don't
follow rule six, though, because you pointed that out once before

I'm not going to actually accept the bug at this time...

Comment 3

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

OK, sorry for being so harsh.

I suspect what's happening is that you're not following rules 4 or 5 (expansion
of auto values).  However, this makes it impossible to test rules 2, 3, and 6,
so that means (by the logic of standards compliance testing) that you're not
following those either (since I can't find any evidence that you are, and the
burden of proof is on the UA to show partial support when missing support makes
features hard to test).  You do follow rule 1 though.

This type of logic isn't too nice to the competition either, you know. :-)

I think the other bug I filed is on rules 2 and 5 (height not expanding), and
this one is on rule 4 (margins not expanding).

Comment 4

[troy]

FWI the height expanding bug is a regression, and although it would appear to
imply lack of following the rules that really isn't the case.

I'll fix that problem first and then we can reevaluate the other issues

Comment 5

[Hixie (not reading bugmail)]

See bug 13568, which is also related to problems in section 10.6.4.

Comment 6

[troy]

This test really isn't valid. At least not with the way the CSS2 spec is
currently defined.

The issue is that you expect the third positioned element to be in the lower
right corner, yet the containing block is the HTML element (also the ICB) and it
has a 'height' of 'auto'.

Comment 7

[Hixie (not reading bugmail)]

Migrating from {css2} to css2 keyword. The {css1}, {css2}, {css3} and {css-moz}
radars should now be considered deprecated in favour of keywords.
I am *really* sorry about the spam...

Comment 8

[troy]

*** Bug 25920 has been marked as a duplicate of this bug. ***

Comment 9

[troy]

I'm marking this INVALID because we now implement the new positioning rules that 
will be in the upcoming errata

The rules are sufficiently different that the old rules are no longer meaningful

Comment 10

[Christine Hoffman]

Will marked this verified invalid, pending official changes to spec