1910 - {inc} Box model vertical formatting problems

Status: VERIFIED FIXED

(Core :: Layout, defect, P3)

Description

[Braden]

Though a previous build of NGLayout nailed this test, it now has a few problems:

1. The floated images are being painted over.
2. The top margin is nolonger present as it should be. (It looks like it is being added to the bottom margin.)

Comment 1

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

The top-margin problem is a duplicate of 1583.

Comment 2

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

The painting problem is similar to bug 2007.

Comment 3

[leger]

Setting all current Open/Normal to M4.

Comment 4

[Paul MacQuiddy]

per leger, assigning QA contacts to all open bugs without QA contacts according
to list at http://bugzilla.mozilla.org/describecomponents.cgi?product=Browser

Comment 5

[kipp]

I've checked in a 90% fix for this - the page now lays out correctly (you have
to trigger a resize reflow to make it do the right thing). Incrementally,
however, it does the wrong thing so I've retitled the bug.

Comment 6

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

Actually, the text (but not the background -- this is fixed now anyway) overlaying the image is correct behavior, since the floats have zero width since their width is not given.

Comment 7

[Braden]

David:
I disagree with you about the correct behavior of the text here. It should not
overlay the images. From CSS2:

"10.3.6 Floating, replaced elements
If 'left', 'right', 'margin-left' or 'margin-right' are specified as 'auto',
their computed value is '0'. If 'width' is 'auto', its value is the element's
intrinsic width."

Kipp: Mozilla still isn't behaving correctly as far as the top and bottom
margins are concerned. This does not appear to be an incremental reflow
problem. Should a new bug be filed on this?

Comment 8

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

You're right.  I could have sworn I looked and the images were within a floated
DIV, rather than floated themselves.  I guess I didn't look.

Hmmm....  I thought this was fixed so width was zero only for non-replaced
content (counting tables as replaced, as they should be, since they have
separate width rules).  It must be going to zero even for replaced content.

This seems to have regressed.  The top margin still disappears on reflow, but
the images now are overwritten again (was that ever fixed?).

Comment 9

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

I think the behavior of the top margin is correct.  The top margin of #test
should collapse with the margin of BODY through the zero margin, padding, and
border of #testContainer and the zero padding of border of body.  This means
that the height of #testContainer doesn't start until the border top of #test.

See the CSS2 errata:
http://www.w3.org/Style/css2-updates/REC-CSS2-19980512-errata.html
for section 10.6.3, and think about where the height of #testContainer is.

You should put a padding on #testContainer to prevent the collapsing.  NGLayout
handles this correctly.

However, the page still shows incremental reflow problems with the replaced
element float widths.

Comment 10

[rickg]

*** Bug 8466 has been marked as a duplicate of this bug. ***

Comment 11

[kipp]

The initial reflow and resize reflow now agree.

Comment 12

[Chris Petersen]

Based on kipp's comments, marking as verified fixed with the Sept 16th build.

Comment 13

[Chris Petersen]

Based on kipp's comments, marking as verified fixed with the Sept 16th build.