Last Comment Bug 455338 - allow all frame classes (any elements) to be absolute positioning containing blocks when position:relative
: allow all frame classes (any elements) to be absolute positioning containing ...
Status: RESOLVED FIXED
:
Product: Core
Classification: Components
Component: Layout: R & A Pos (show other bugs)
: Trunk
: All All
: -- normal with 4 votes (vote)
: mozilla10
Assigned To: :Ehsan Akhgari (busy, don't ask for review please)
:
Mentors:
Depends on: 10209
Blocks: 63895 437722 450418 454749 508335 640443
  Show dependency treegraph
 
Reported: 2008-09-15 09:36 PDT by David Baron :dbaron: ⌚️UTC-7 (review requests must explain patch)
Modified: 2013-08-20 10:27 PDT (History)
16 users (show)
roc: blocking1.9.2-
roc: wanted1.9.2-
See Also:
Crash Signature:
(edit)
QA Whiteboard:
Iteration: ---
Points: ---
Has Regression Range: ---
Has STR: ---
-


Attachments

Description David Baron :dbaron: ⌚️UTC-7 (review requests must explain patch) 2008-09-15 09:36:18 PDT
I thought we had a bug on this, but I can't find one.

The CSS spec says that any element with position:relative can be a containing block for absolutely positioned elements.  We don't currently support this, since only block and inline frames currently support being an absolute positioning containing block.

We should figure out a way to allow all frames to be absolute positioning containing blocks.
Comment 1 Emil Ivanov 2010-11-01 15:24:53 PDT
I guess this does not have a chance for Firefox 4, but what about the next release?
Comment 2 David Baron :dbaron: ⌚️UTC-7 (review requests must explain patch) 2010-11-12 09:55:53 PST
I think the best path to fixing this is perhaps something like:

  (1) add a codepath for dynamic recomputation of overflow areas without reflow (this is also really useful for a bunch of performance optimizations)

  (2) store all absolutely positioned frames in a frame property (probably always on the first continuation), and in DidReflow of *any* continuation of a frame with such a property (probably also needs a frame state bit to be transferred to all continuations), record (in the reflow state?) that we need to reflow the absolute frames when we're done reflowing the deeper of (a) the nearest common ancestor of all the continuations of the frame or (b) the reflow root.
Comment 3 Brad Lassey [:blassey] (use needinfo?) 2010-12-06 10:32:51 PST
(In reply to comment #0)
> I thought we had a bug on this, but I can't find one.
is this related to bug 63895?
Comment 4 David Baron :dbaron: ⌚️UTC-7 (review requests must explain patch) 2011-05-08 08:36:52 PDT
*** Bug 655591 has been marked as a duplicate of this bug. ***
Comment 5 David Baron :dbaron: ⌚️UTC-7 (review requests must explain patch) 2011-05-08 08:37:45 PDT
Ehsan is working on this in bug 10209.
Comment 6 :Ehsan Akhgari (busy, don't ask for review please) 2011-09-29 14:54:40 PDT
The infrastructure for this has landed in bug 10209.  I'll look into using this infrastructure for the remaining types in the future.

Note You need to log in before you can comment on or make changes to this bug.