Open Bug 349760 Opened 18 years ago Updated 2 years ago

[reflow branch] Pref width calculation off with float followed by a block

Categories

(Core :: Layout, defect)

x86
Linux
defect

Tracking

()

People

(Reporter: bzbarsky, Unassigned)

References

(Blocks 1 open bug, )

Details

(Whiteboard: [float pref width])

Attachments

(2 files)

The testcase is from http://bugzilla.opendarwin.org/show_bug.cgi?id=10451

We should really try to put the little image to the right of the float, not below it...
se by not clearing the float list when doing the Break() before a block line.  We could get closer to Opera's behavior if we also didn't clear after a block and did something smart with block lines containing a single block with "clear" set on it (for example, only clear out part of the float list?  Or something?).

Looking into what IE does on those testcases.
It looks like IE renders the testcases identically to trunk seamonkey, for what it's worth...

David, what do you think of trying to do something per comment 3?
So I'd previously been under the impression that Safari's behavior with intrinsic widths around floats was generally more Web-compatible than Opera's, and I'm inclined towards compatibility with Safari rather than Opera.  I'm also inclined towards producing smaller pref widths, although there are designs that are broken both ways.
For what it's worth, I tried doing comment 3 and getting sane consistent behavior was pretty hard... as in, I didn't get there.  ;)

I agree we should do the simple (and Safari-compatible) thing for now.  So does roc.
Whiteboard: [float pref width]
Product: Core → Core Graveyard
Component: Layout: Misc Code → Layout
Product: Core Graveyard → Core
Severity: normal → S3
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: