Robert O'Callahan
|
cddb22d87c
|
Bug 243519. Rework root element frame construction and the CanvasFrame so that the CanvasFrame is an abs-pos container and the root element frame can be positioned. r=fantasai,sr=dbaron
|
2008-09-08 20:13:17 +12:00 |
|
Robert O'Callahan
|
ea7596e60c
|
Backing out changesets 75919d3eb3d0 and 14ce7619e9c1 due to test failures
|
2008-09-07 00:35:20 +12:00 |
|
Robert O'Callahan
|
f7339a0460
|
Bug 243519. Rework root element frame construction and the CanvasFrame so that the CanvasFrame is an abs-pos container and the root element frame can be positioned. r=fantasai,sr=dbaron
|
2008-09-06 21:35:05 +12:00 |
|
dbaron@dbaron.org
|
4f711db346
|
Don't let the pres context's visible area get permanently set to an unconstrained height when we call ResizeReflow with an unconstrained height during SizeToContent. b=371508 r+sr=roc a=blocking1.9+
|
2008-03-12 15:04:45 -07:00 |
|
fantasai.cvs@inkedblade.net
|
678b52fb1e
|
fix use of DEBUG-only data member, bug=154892 r+sr+a=roc
|
2007-10-02 01:29:23 -07:00 |
|
fantasai.cvs@inkedblade.net
|
2456fd7a33
|
paginate absolutely-positioned elements, b=154892 r+sr=roc a=roc
|
2007-10-01 22:57:45 -07:00 |
|
jwalden@mit.edu
|
96470d6e9b
|
Bug 386801 - nsAbsoluteContainingBlock doesn't need to store a child list name; this shaves four bytes off nsViewportFrame, nsBlockFrame, and nsPositionedInlineFrame. r+sr+a=dbaron
|
2007-08-02 15:44:36 -07:00 |
|
bzbarsky@mit.edu
|
21e0d65a42
|
Introduce a setter for nsHTMLReflowState::mComputedHeight so that we're sure to
update the resize flags properly. Bug 388084, r+sr+a=dbaron
|
2007-08-02 11:08:05 -07:00 |
|
roc+@cs.cmu.edu
|
61373b8418
|
Bug 379349. Add support for 'overflow containers' --- special frame continuations that do not map any content of their own, but serve only as containers for laying out children that overflowed their parents' content height. patch by fantasai, r+sr=eli,roc
|
2007-07-25 21:03:29 -07:00 |
|
bzbarsky@mit.edu
|
3cb4bc9240
|
Change the FrameNeedsReflow API to pass the dirty flags to be added directly tothe method, instead of setting them before calling the method. That way we canavoid reflowing the ancestor of a reflow root which is not itself dirty but hasdirty children. This also makes it harder to set dirty bits inconsistentlywith the FrameNeedsReflow call. Bug 378784, r+sr=dbaron, pending rbs' reviewon the mathml parts.
|
2007-05-06 12:16:51 -07:00 |
|
hg@mozilla.com
|
465265d0d4
|
Free the (distributed) Lizard! Automatic merge from CVS: Module mozilla: tag HG_REPO_INITIAL_IMPORT at 22 Mar 2007 10:30 PDT,
|
2007-03-22 10:30:00 -07:00 |
|