315 Commits

Author	SHA1	Message	Date
roc+@cs.cmu.edu	5ecac60819	Backing out new textframe landing because of Windows bustage.	2007-06-13 15:31:31 -07:00
roc+@cs.cmu.edu	aabe78b6ca	Bug 367177. Turn on new textframe. Lord, have mercy upon us.	2007-06-13 13:20:19 -07:00
roc+@cs.cmu.edu	fba8d645e8	Not part of the build (barely). Don't skip updating per-char arrays for characters that are cluster continuations (in small-caps code).	2007-06-13 11:28:38 -07:00
roc+@cs.cmu.edu	5c7dda1060	Not part of the build (barely). Don't allow PeekOffsetWord to break before the first cluster. Avoids a fencepost error in ClusterIterator where going backwards we might try to look at one past the end of mWordBreaks.	2007-06-13 11:25:14 -07:00
roc+@cs.cmu.edu	ee1924e0c6	Not part of the build (barely). Fix intrinsic-min-widths properly for preformatted white-space... the previous checkin messed up the check for newline characters.	2007-06-13 03:58:44 -07:00
roc+@cs.cmu.edu	e799be426c	Not Part Of The Build (barely). Clean up function names and comments. Also fix an unreported regression where we failed to take preformatted newlines into account when computing text intrinsic-min-width.	2007-06-13 02:06:28 -07:00
roc+@cs.cmu.edu	2f18074ec6	Landing textframe updates: Use word-based textrun cache; rewritten PeekOffsetWord; adjust textframe content-offsets and content-lengths when CharacterDataChanged fires; search backwards for textrun scanning/creation start point, instead of all the way back to the start of the block; ensure that we pass the correct start iterator to PropertyProvider in TrimTrailingWhiteSpace.	2007-06-12 14:01:12 -07:00
dbaron@dbaron.org	9e2fef4c30	Implement correct intrinsic width calculation (for container) for elements with specified height properties and an intrinsic ratio. b=364066 r+sr=bzbarsky	2007-06-12 11:27:09 -07:00
smontagu@smontagu.org	2926465a3b	Clear textruns during bidi resolution. Not Part of the Build. Bug 382429, r+sr=roc	2007-06-02 23:53:10 -07:00
joshmoz@gmail.com	c4219bdb45	need to set plugin window x and y coords in CG mode. b=382765 r=sfraser sr=roc	2007-06-02 15:02:57 -07:00
roc+@cs.cmu.edu	fa9335bd42	Bug 382279. Disable ligatures when letter-spacing is in effect. Also, avoid adding space inside ligatures if the gfx backend doesn't respect our disabling of ligatures. r=pavlov	2007-05-30 22:01:56 -07:00
roc+@cs.cmu.edu	18223fc22d	Not part of the build. Change nsLineBreaker so that breaks at element boundaries are always governed by the white-space property on the element that's the nearest common ancestor of the DOM nodes on each side of the break point --- fixes bug 382289. Also, pass NBSP through to textruns so that whitespace trimming doesn't trim NBSP. Also, count trimmed spaces correctly when updating mClusterCount.	2007-05-30 15:27:27 -07:00
mats.palmgren@bredband.net	1cb22a625f	Check LL_GOTLINEBOX before using mLineBox. b=382396 r+sr=dbaron	2007-05-30 08:08:50 -07:00
smontagu@smontagu.org	d16cf8e31c	Do bidi resolution during GetPrefWidth() and GetMinWidth() as well as Reflow(). Bug 365130, r+sr=roc	2007-05-28 07:40:56 -07:00
roc+@cs.cmu.edu	3fac21ece0	Not part of the build. Don't trim whitespace if wrapping is disabled --- in min-width and pref-width calculations.	2007-05-27 19:38:21 -07:00
roc+@cs.cmu.edu	8b4068bd66	Not part of the build. Don't trim whitespace if wrapping is disabled. Don't compress a space that's preceded by an incompressible space.	2007-05-27 19:20:58 -07:00
roc+@cs.cmu.edu	b83f90d3ea	Not part Of The Build. Bug 381228. Fix handling of case where the text is not part of a block.	2007-05-23 20:06:37 -07:00
preed@mozilla.com	4750a2febd	Bug 381695: TextFrame feature makefile switch needs to pick up its setting from environment. r=bsmedberg	2007-05-23 19:13:04 -07:00
roc+@cs.cmu.edu	d88807555b	(Not Part Of The Build). Fix incorrect treatment of first character always as uppercase, even when it isn't, when building a small-caps textrun ...	2007-05-23 02:37:52 -07:00
roc+@cs.cmu.edu	b2cf85dacd	Bug 380625. (Not Part Of The Build). Fix cluster count for trimmed spaces.	2007-05-23 02:36:16 -07:00
roc+@cs.cmu.edu	2d4b4128c0	Not Part Of The Build (new textframe). IsInBounds uses incorrect logic and produces spurious warnings.	2007-05-22 21:08:38 -07:00
sharparrow1@yahoo.com	d4a50fcc87	Bug 381645: Remove nsBoxFrame::GetContentOf. r+sr=dbaron.	2007-05-22 20:48:43 -07:00
roc+@cs.cmu.edu	a56819194e	Not Part Of The Build. Update nsTextFrameThebes whitespace handling; allow line breaking only at the end of a run of whitespace.	2007-05-22 16:45:47 -07:00
sharparrow1@yahoo.com	888cd6828d	Bug 381621: Get rid of nsBoxFrame::AddRef/Release, plus a few misc cleanups. r+sr=dbaron.	2007-05-22 13:34:00 -07:00
sharparrow1@yahoo.com	e121bf78ce	Bug 377391: Parts of a rectangle with CSS outline remains, when changing rectangle size. r+sr=roc	2007-05-21 03:06:08 -07:00
sharparrow1@yahoo.com	197f518480	Bug 45631: text-indent: n% doesn't incrementally reflow correctly. r+sr=roc.	2007-05-21 02:55:27 -07:00
roc+@cs.cmu.edu	b7f3dc0143	Bug 381130. Small changes to layout to fix an inline reflow bug and add a new API to nsLineLayout to support tab stops. r+sr=dbaron	2007-05-20 20:44:26 -07:00
dbaron@dbaron.org	93042a4e45	Don't propagate line layout in reflow state when we're not in the line. b=320502 r+sr=roc	2007-05-17 23:04:04 -07:00
mats.palmgren@bredband.net	032cb5c51d	NPOTB. Add missing rendering context null-checks. b=380502 r+sr=roc	2007-05-17 03:52:20 -07:00
roc+@cs.cmu.edu	752cb1c711	Not Part Of The Build. 1) Round font underline/overline/strikeout offsets and sizes to device pixels to try to ensure consistent visual offsets and sizes even when the baseline point is not pixel-aligned. 2) snap the text baseline y coordinate to the nearest pixel, because the decoration lines will be snapped themselves and we need the gaps to remain consistent. 3) fix bug in selection-decoration x coordinate which was forgetting to include the frame x-offset.	2007-05-15 22:21:41 -07:00
roc+@cs.cmu.edu	00c14ee581	Not Part Of The Build. Fix underline/strikeout positioning in new text frame --- simple appunits/pixels confusion.	2007-05-15 16:41:41 -07:00
dbaron@dbaron.org	e9a2ce02d6	Fix vertical resizing for quirks mode percentage heights. b=380004 Patch by Daniel Holbert <dholbert@mozilla.com>. r+sr=dbaron	2007-05-15 14:52:44 -07:00
roc+@cs.cmu.edu	42a453e958	NPOTB, new-text-frame. Fix mouse events targeting after the end of a textframe.	2007-05-14 22:23:19 -07:00
roc+@cs.cmu.edu	c04ab10b10	Back out enabling of new text frame... its time has not yet come	2007-05-14 20:58:10 -07:00
roc+@cs.cmu.edu	69246cf0ee	Bug 380692. Change gfxTextRun API so the textrun copies text, if necessary, instead of the caller having to do it. r=vlad	2007-05-14 20:56:47 -07:00
Olli.Pettay@helsinki.fi	7ec10e025c	Bug 363089, Remove nsIDOMEventReceiver (this time passes mochitest), r+sr=jst	2007-05-14 02:11:38 -07:00
asqueella@gmail.com	1f0637ebc5	Bug 380094 - make mousethrough more generally usefulp=toshok@hungry.comr+sr=roc	2007-05-13 09:00:00 -07:00
neil@parkwaycc.co.uk	23a051773f	Follow-up cleanup b=375403 r+sr=roc	2007-05-13 03:06:57 -07:00
roc+@cs.cmu.edu	9a3b1f05c8	Not part of the build. Fix off-by-one error in selected-text painting in new text frame.	2007-05-11 15:54:51 -07:00
Olli.Pettay@helsinki.fi	f29b4e2b1b	backout Bug 363089	2007-05-11 06:34:05 -07:00
Olli.Pettay@helsinki.fi	91a5c67cd1	Bug 363089, Remove nsIDOMEventReceiver, r+sr=jst	2007-05-11 04:15:11 -07:00
dbaron@dbaron.org	6becb79be4	Remove now-unneeded checks for eStyleUnit_Null. b=379741 r+sr=bzbarsky	2007-05-10 23:02:31 -07:00
dbaron@dbaron.org	6375aad92f	Add eStyleUnit_None so we don't have to use eStyleUnit_Null as a real value for max-width and max-height. b=379741 r+sr=bzbarsky	2007-05-10 23:01:31 -07:00
neil@parkwaycc.co.uk	001e700eb6	Bug 375403 originally started off as a simple request to enable window translucency when windows or popups use opacity, alpha transparent background or non-zero border radius but mutated to include refactoring non zero side testing r+sr=roc	2007-05-10 08:46:42 -07:00
roc+@cs.cmu.edu	5dacf8d2c6	Fix const issue. Not Part Of The Build	2007-05-10 07:42:56 -07:00
bzbarsky@mit.edu	e77dda0c0b	Twiddle parentheses a little bit. Bug 311415 followup, r+sr=dbaron	2007-05-09 19:55:15 -07:00
roc+@cs.cmu.edu	55d1df583d	Not part of the build. Fix text lifetime issues for transformed textruns.	2007-05-09 17:58:47 -07:00
roc+@cs.cmu.edu	e360cfb42d	Not part of the build. Update new text frame for textrun API changes. Also implements a textrun cache for the new text frame	2007-05-09 15:04:56 -07:00
roc+@cs.cmu.edu	f8fbe830af	Bug 380209. Hooks for a textframe-specific textrun cache startup/shutdown. r+sr=bzbarsky	2007-05-09 15:02:29 -07:00
smontagu@smontagu.org	1589352d02	Uneven justification with new textframe. Bug 380062, r+sr=roc	2007-05-09 11:09:32 -07:00
smontagu@smontagu.org	7ce6d0b781	Clear mTextRun on a style change reflow. Bug 379969, r+sr=roc	2007-05-08 07:33:09 -07:00
bzbarsky@mit.edu	ab6e18ee19	Fix determination of the style context parent for the inner table frame of aroot table. Bug 379386, r+sr=roc	2007-05-07 20:59:12 -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
dbaron@dbaron.org	79c8ea4796	Remove special handling of empty continuations, which should no longer be needed. b=238473 Based on patch by Hideo Saito <saito@densan.co.jp>. r=dbaron sr=roc	2007-05-05 10:06:04 -07:00
dbaron@dbaron.org	2f7331b24a	Address bzbarsky's final review comments (comment 14) on bug 311415: consider ch units fixed as well.	2007-05-05 10:01:51 -07:00
bzbarsky@mit.edu	a73ee91520	Simplify code. Bug 367442 followup, jag's idea, r+sr=dbaron	2007-05-04 23:13:48 -07:00
bzbarsky@mit.edu	64c3e5df8a	Make computed style handle eStyleUnit_Chars. Bug 371043, r+sr=dbaron	2007-05-04 23:09:50 -07:00
bzbarsky@mit.edu	685a823789	Make computed style handle "normal" for line-height by computing an actualnumeric height. Bug 371041, r+sr=dbaron	2007-05-04 22:30:10 -07:00
dbaron@dbaron.org	7dc7c89d6c	Add support for -moz-intrinsic, -moz-min-intrinsic, -moz-shrink-wrap, and -moz-fill for width, min-width, and max-width. b=311415 r+sr=bzbarsky	2007-05-03 16:11:00 -07:00
bzbarsky@mit.edu	4272f75194	Fix assert to be more correct. Bug 379383, r+sr=dbaron	2007-05-02 13:32:17 -07:00
roc+@cs.cmu.edu	61473971b1	Bug 372703. Position child views whenever we place an abs-pos child; even if the child hasn't moved, this frame itself might have, so view offsets might have changed. r=eli,sr=dbaron	2007-05-01 17:07:51 -07:00
bzbarsky@mit.edu	32d3d359f4	Introduce a GetPrimaryShell() API on nsIDocument. Convert existing callers ofGetShellAt(0) to using this API. Bug 378780. API introduction part by me,r+sr=jst. Mass-changes done by taras using squash, r+sr=me	2007-05-01 15:24:20 -07:00
bzbarsky@mit.edu	e83c55e645	Make InitialReflow just post a reflow event instead of doing a synchronousreflow. Bug 378975, r=dbaron and Eli Friedman, sr=dbaron	2007-05-01 10:15:08 -07:00
mats.palmgren@bredband.net	dfbd9d77a5	Bye bye gtk1 and xlib (part 1 of 2). b=326152 r+sr=roc	2007-04-30 16:52:44 -07:00
bzbarsky@mit.edu	45ccef2087	remove debugging code again.	2007-04-29 21:36:03 -07:00
bzbarsky@mit.edu	013df0d678	Put back this test code for one run so I can get data from Tp/Tp2	2007-04-29 20:11:04 -07:00
asqueella@gmail.com	20e9450a7d	bug 375291 - update layout/generic/punct_marks.ccmap to unicode 5.0p=arno. <arno.@no-log.org>r=smontagu, sr=dbaron	2007-04-27 09:42:22 -07:00
asqueella@gmail.com	18c7325a7b	Bug 376034 - Possible incorrect coordinate transform in nsImageFramep=Jeremy Lea <reg@freebsd.org>r+sr=roc	2007-04-27 07:40:17 -07:00
bzbarsky@mit.edu	e0ed942e8a	Back out the test code.	2007-04-26 19:06:50 -07:00
bzbarsky@mit.edu	a5e93e2e05	Sprinkle null-checks in an attempt to "fix" the orange so I can get answers.	2007-04-26 17:47:34 -07:00
bzbarsky@mit.edu	789ffd0e73	Adding some test code for a single tinderbox run so I can see what's going onwith Tp2 and such. Will back out after one cycle.	2007-04-26 15:46:05 -07:00
bzbarsky@mit.edu	f1c4c68543	Don't do incremental reflow for now, because tables can't deal. Followup forbug 378480, r+sr=roc	2007-04-25 22:33:34 -07:00
bzbarsky@mit.edu	24c328d268	Fix the vertical scrollbar guessing in nsHTMLScrollFrame to do better for aninitial reflow if we're doing that initial reflow a ways into the documentload. Bug 378480, r+sr=roc	2007-04-25 19:29:21 -07:00
bzbarsky@mit.edu	c77758b510	Resolve the right (parentless) style context for a table as a document root,and fix the assertion when adjusting parent style context frames to not triggerin non-scrollable documents. Bug 378693, r+sr=roc	2007-04-25 19:26:25 -07:00
benjamin@smedbergs.us	2e25a321f8	Bug 376636 - Building with gcc 4.3 and -pendatic fails due to extra semicolons, patch by Art Haas <ahaas@airmail.net>, rs=me	2007-04-23 07:21:53 -07:00
sayrer@gmail.com	d75d2627f8	bug 378383. nsReflowStatus not always initialized in nsBlockFrame::ReflowInlineFrame. r+sr=bzbarsky	2007-04-22 13:57:54 -07:00
bzbarsky@mit.edu	5a4fccb862	Fix assertion to be more correct. No bug.	2007-04-20 20:57:40 -07:00
roc+@cs.cmu.edu	2eec67fef4	Bug 368863. Reparent floats properly when placeholders move between inline continuations. r+sr=dbaron	2007-04-20 17:42:58 -07:00
sharparrow1@yahoo.com	b9768958d9	Bug 378129: Consistently pass the right frame to InvalidateInternal. r+sr=roc.	2007-04-20 16:51:12 -07:00
sayrer@gmail.com	0ce8af1698	Bug 368994. move mochitests near the code they test. round 7. r=bzbarsky	2007-04-16 08:18:50 -07:00
bzbarsky@mit.edu	c27dc765b6	The parent style context frame determination for an inner table should use theinner table's pseudo, not the outer table's. Bug 377603, r+sr=dbaron	2007-04-15 22:55:42 -07:00
bzbarsky@mit.edu	2dca5abb0a	Improve detection of blocks with first-letter style. Bug 372550, r+sr=dbaron	2007-04-15 17:42:54 -07:00
bzbarsky@mit.edu	e6d41ebe8b	Change the way we determine the style context parent frame to skip css anonboxes unless we're determining the style context parent for something that isitself a css anon box (and is not a non-element frame). Fixes bug 323656(which is where the patch is), bug 85872, bug 280610. As far as I can tell,also fixes bug 317876, bug 372376, bug 374297. r+sr=dbaron	2007-04-15 16:14:26 -07:00
uriber@gmail.com	b813fa7847	Test for null parent content, rather than null parent frame, to avoid possible crash in FindBlockFrameOrBR(). bug=370174 r=sharparrow1 sr=roc	2007-04-14 09:34:12 -07:00
roc+@cs.cmu.edu	056f048065	Not part of the build. Fix new textframe issues with mixed-direction text and textrun reconstruction.	2007-04-13 15:57:54 -07:00
Olli.Pettay@helsinki.fi	3dcecf2a6e	Bug 374090, Crash [@ nsGfxScrollFrameInner::GetScrollPortSize] with noframes and setting binding on root element, r+sr=roc	2007-04-13 07:07:40 -07:00
enndeakin@sympatico.ca	8b6667e253	Bug 375681, use drag event names from what-wg spec, also implement new drag and dragend events, r=smaug,sr=roc	2007-04-11 21:37:39 -07:00
roc+@cs.cmu.edu	b90ebbbc6c	Bug 370439 (sort of). GetAbsPosClipRect needs to take the new frame size as a parameter, because the current frame size may be invalid when we are called by FinishAndStoreOverflow. r+sr=dbaron	2007-04-10 22:04:09 -07:00
Olli.Pettay@helsinki.fi	cb7222adf2	Additional patch for bug 369150, r+sr=bz	2007-04-10 12:08:21 -07:00
bzbarsky@mit.edu	9bdd8b028c	Make sure to update view positions if we move our block around verticallyduring reflow. Bug 365294, r+sr=dbaron	2007-04-05 15:47:38 -07:00
roc+@cs.cmu.edu	f1007b83a5	Bug 368860. Don't perform out-of-bounds access if we have a bogus empty previous-text-frame. r+sr=dbaron	2007-04-03 19:28:27 -07:00
roc+@cs.cmu.edu	9ebf4ec76a	Not part of the build --- update GetPresContext to PresContext in a few not-built places.	2007-04-02 20:56:55 -07:00
bzbarsky@mit.edu	e0ad31bc86	Make non-blockframe kids of xul not have to end up with a zero height. Bug374038, r+sr=dbaron	2007-04-01 08:15:17 -07:00
bzbarsky@mit.edu	2fd17c94d8	Rename nsIFrame::GetPresContext to nsIFrame::PresContext. Bug 376042, patch byTaras Glek <tglek@mozilla.com>, rs=roc.	2007-03-30 14:11:41 -07:00
sharparrow1@yahoo.com	bc0f1805e5	Bug 372062: Auto values for right and bottom in CSS clip rect broken. r+sr=roc.	2007-03-28 18:31:26 -07:00
roc+@cs.cmu.edu	6599844c53	Not part of the build (well, gfxSkipChars is built but not used): fix bugs in gfxSkipChars, and new textframe fixes for soft hyphens, justification, and caret movement (PeekOffset*) among other things	2007-03-28 16:54:32 -07:00
sharparrow1@yahoo.com	b087c08cc6	Bug 373474: Scrollbar buttons don't get invalidated (i.e. for notification box or resizing). r+sr=roc.	2007-03-27 23:24:40 -07:00
sharparrow1@yahoo.com	1cf7602a3e	Bug 373433: views inside absolutely positioned blocks sometimes mispositioned. r+sr=roc.	2007-03-27 23:14:03 -07:00
joshmoz@gmail.com	e836dc9c3f	get rid of old backbuffer API and related code. b=371392 r/sr=roc	2007-03-26 23:21:55 -07:00
Olli.Pettay@helsinki.fi	756bfcf6f0	Bug 369168, r+sr=roc	2007-03-26 22:30:14 -07:00
sharparrow1@yahoo.com	6fdcc6c041	Bug 372037: CSS clip property should clip the overflow rect as well as the paint rect; r+sr=roc.	2007-03-26 20:58:17 -07:00
joshmoz@gmail.com	094f428519	Implement CoreGraphics NPAPI Drawing Model. NPAPI plugins now have the option to render using CoreGraphics instead of Quickdraw. b=344427 r=sfraser sr=jst	2007-03-26 18:07:57 -07:00
sharparrow1@yahoo.com	bdebedc073	Bug 375273: minor code cleanup in nsContainerFrame. r+sr=roc	2007-03-25 18:12:49 -07:00
asqueella@gmail.com	69e6a9a9b7	Bug 374486 - "WARNING: Nonempty list left over?" in visual debugging mode ... p=Sylvain Pasche <sylvain.pasche@gmail.com>r+sr=roc	2007-03-24 22:30:59 -07:00
roc+@cs.cmu.edu	29a2c959bb	No bug. A variety of fixes to fix various reftest failures. Not Part Of The Build.	2007-03-24 04:07:36 -07:00
mats.palmgren@bredband.net	42d130d286	Fix "selection colors are sometimes wrong". b=374765 r+sr=pavlov	2007-03-24 02:21:35 -07:00
Olli.Pettay@helsinki.fi	1e619f723b	Bug 375093, Crash [@ nsTypedSelection::ScrollIntoView] when right-clicking paste in input which then gets destroyed oninput, r+sr=dbaron	2007-03-23 10:32:34 -07:00
Olli.Pettay@helsinki.fi	be5c439142	backout Bug 369168	2007-03-23 03:03:25 -07:00
Olli.Pettay@helsinki.fi	956eca1093	Bug 374937, Crashes [@ nsTypedSelection::GetFrameSelection], r+sr=roc	2007-03-22 17:06:38 -07:00
Olli.Pettay@helsinki.fi	5efe36d569	Bug 369168, r+sr=roc	2007-03-22 17:01:36 -07:00
Olli.Pettay@helsinki.fi	f9d1cc6908	Bug 369150, r+sr=roc	2007-03-22 16:57:55 -07:00
roc+@cs.cmu.edu	374a8cc077	Bug 333659. Major new-textframe update. You can really browse the Web with it now, although major bugs remain.	2007-03-22 16:13:12 -07:00
roc+@cs.cmu.edu	859e758dce	No bug. Comment-only change to note a potential optimization opportunity.	2007-03-22 16:10:10 -07:00
joshmoz@gmail.com	500c06e1d0	fix crash, suspend native events when instantiating plugins. patch by Steven Michaud. b=345627 r=josh sr=roc	2007-03-22 16:04:51 -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