311935 - DIV with 'overflow' has margin-left added to left float's position

Status: VERIFIED INVALID

(Core :: Layout: Block and Inline, defect)

Description

[Samuel Sidler (old account; do not CC)]

Splitting from bug 289577.

This is a regression occurring between 2004-09-04 and 2004-09-05. The bonsai
link below gives the checkins for that window of time.

Testcase in next comment.

Comment 1

[Samuel Sidler (old account; do not CC)]

Attachment: Testcase

Comment 2

[Samuel Sidler (old account; do not CC)]

Regression window: 

http://bonsai.mozilla.org/cvsquery.cgi?treeid=default&module=all&branch=HEAD&branchtype=match&dir=&file=&filetype=match&who=&whotype=match&sortby=Date&hours=2&date=explicit&mindate=2004-09-04+04%3A00%3A00&maxdate=2004-09-05+09%3A00%3A00&cvsroot=%2Fcvsroot

Comment 3

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

As I said in the other bug, I'm reasonably confident even without testing that
your alleged regression window is simply an artefact of the testcase's use of
'overflow-x' rather than 'overflow'.

Also, bug 289577 comment 10 (dbaron) says:

The overflow property actually should sometimes affect layout (especially in the
presence of floats), because a DIV with 'overflow' non-visible establishes a new
block formatting context.  That said, the testcase above does look wrong.

and bug 289577 comment 12 (bzbarsky) says:

The layout of Simon's testcase (and Samuel's too) is affected by CSS 2.1,
section 9.5, which says:

   The margin box of a table, a block-level replaced element, or an element in
   the normal flow that establishes a new block formatting context (such as an
   element with 'overflow' other than 'visible') must not overlap any floats in
   the same block formatting context as the element itself. If necessary,
   implementations should clear the said element by placing it below any
   preceding floats, but may place it adjacent to such floats if there is
   sufficient space.

The only possible bug on that testcase is that there is not "sufficient space"
and so we should be clearing, not shifting to the right.

Comment 4

[Samuel Sidler (old account; do not CC)]

(In reply to comment #3)
> As I said in the other bug, I'm reasonably confident even without testing that
> your alleged regression window is simply an artefact of the testcase's use of
> 'overflow-x' rather than 'overflow'.

Changing 'overflow-x' to 'overflow' yields the same results in the testcase so
this doesn't appear to be exclusive to overflow-x.

Comment 5

[Boris Zbarsky [:bzbarsky]]

> Changing 'overflow-x' to 'overflow' yields the same results in the testcase

Yes, but does it yield the same "regression window"?

The only "improvement" here would be to clear things (move the float down past
that div).

Comment 6

[Scott MacGregor]

this doesn't seem like a critical stop the press type bug. David/bz, please
re-nominate and add some commentary if you disagree.

Comment 7

[Hixie (not reading bugmail)]

Like dbaron said, the only bug here is that if you make your window small enough
that the overflow:hidden box would go off the side of the screen, it should wrap
to below the float.

Comment 8

[Boris Zbarsky [:bzbarsky]]

Which is bug 290146.