Граф коммитов

60 Коммитов

Автор SHA1 Сообщение Дата
warren%netscape.com cd56c0575b Bug 47207. Backing out logging/PRINTF changes until we can fix stopwatch.h, introduce double parens, etc. 2000-10-28 22:17:53 +00:00
warren%netscape.com 9a6d92a433 Bug 47207. Changing printf to PRINTF to use new logging facility. r=valeski,sr=waterson 2000-10-27 22:43:51 +00:00
attinasi%netscape.com 1fd5ba1fd2 Added default implementation for frames to allow a frame to specify which frame provides its parent style context. Also reworked ReResolveStyleContext and VerifyStyleTree to allow the style context tree to diverge from the frame tree. b=45210 r=karnaze 2000-09-12 22:47:09 +00:00
pierre%netscape.com 84f381ec78 Fix blocker bug #52039 (crash when resolving pseudo-style for anonymous content). 2000-09-11 03:58:22 +00:00
jst%netscape.com 7aa7ed7f6d Fixing nsbeta3+ bug 28277. Add a callback to nsHTMLImageElement so that we know when a image created in JavaScript has loaded (or fails to load), this is needed to support onload and onerror event fireing and also to support the '.complete' property on images create in JavaScript. r=nisheeth@netscape.com 2000-09-01 07:05:08 +00:00
pollmann%netscape.com 3f1d49a015 Bug 32770: Save scroll position in session history r=nisheeth 2000-06-15 00:35:46 +00:00
attinasi%netscape.com d264987b1c Removed erroneous assertion: b=38768 r=dcone 2000-06-07 23:05:30 +00:00
buster%netscape.com e1de241d59 bug 38665
r=rickg a=rickg
fixed an optimized-only crash that looks like a race condition
where we send a message containing a pointer to a frame back to
the frame manager after the frames have been deleted.
2000-06-05 03:37:33 +00:00
attinasi%netscape.com 009460e9a7 Added a new method GetCanvasFrame which walks down from the root frame looking for the frame that represents the canvas. b=40217 r=kmcclusk a=karnaze 2000-05-31 22:33:12 +00:00
attinasi%netscape.com dd07df1f66 Fixed a problem with the method AttributeAffectsStyle: it was erroneously ignoring the fact that HTML attributes can affect style even if there are no CSS rules with thos attributes in the selectors. Changed to restrict the selector evaluation to XML elements. r=shaver, a=karnaze, b=39642 2000-05-25 23:02:31 +00:00
shaver%mozilla.org 26b7a4c5b3 make sure that style attributes get the reresolve dance 2000-05-16 22:05:47 +00:00
shaver%mozilla.org 2da0ab457d Significantly reduce the amount of time we blow on style reresolution by
tracking the attributes which appear in CSS selectors, and only firing
a complete reresolution when we change one of those.  In the irrelevant-attr
case, we just let the frame update itself and get on with our lives right
away.  Fixed 38378, r=brendan, booyah.

Also, when setting an attribute to the same value that it currently holds,
don't think about it especially hard, and _don't_ go anywhere near the
style code.  Fixed 21879, r=brendan, boo_yah_.
2000-05-16 02:43:33 +00:00
mkaply%us.ibm.com 23620fb329 # 34082
r= troy
OS/2 bring up continues - PR_CALLBACKs for VisualAge
2000-05-02 23:07:11 +00:00
attinasi%netscape.com 255aa49d63 Fixed PseudoStyle resolution for primary style context in ReResolveStyleContext: was passing child-content when it should be passing parent-content. r=pierre 2000-04-23 03:49:19 +00:00
nisheeth%netscape.com 7b7f9d1021 Fix for bug 30958: Reflow commands are now processed asynchronously all the time. The semantics of the API for queuing and processing reflow commands has changed. Now, AppendReflowCommand() initiates an asynchronous reflow event and all reflow command processing happens when that event fires. It is no longer possible for consumers outside the presentation shell to call ProcessReflowCommands() synchronously. Because of this change, reflow locks are no longer needed and have been removed. 2000-04-17 07:13:57 +00:00
rickg%netscape.com 1d1db214c3 add null ptr check to stop crash seen in talkback 2000-03-28 01:08:36 +00:00
slamm%netscape.com f9a4756c9a Change "struct" to "class ... { public:" to quiet a warning. 2000-03-24 21:50:40 +00:00
attinasi%netscape.com 91f9e59318 ReResolveStyleContext now checks for the case where a background image is no longer used, in which case it stops the load for the frame. r=buster,pierre b=27042 2000-02-16 02:49:01 +00:00
dveditz%netscape.com 84688e0640 Attempting to fix nsCOMPtr .get() related bustage on Linux 2000-02-15 02:55:23 +00:00
beard%netscape.com b5072ce22c fixes nsIAtom leaks, bug # 25761, r=waterson 2000-02-15 01:17:18 +00:00
pp%ludusdesign.com 58065dda04 Converting ::GetIID() into NS_GET_IID(). Bug #20232. r=scc, r=mozbot 2000-02-02 22:24:56 +00:00
rods%netscape.com c56fe2a751 when an option contained a comment node and and an attr was unset case a
WipeContainingBlock to be called. This was because the undisplayed node had a psuedo
style that wasn't being properly reolved to this code thought the comment was now visible
instead of being hidden.
b=none r=pierre,attinasi
2000-01-28 15:03:00 +00:00
nisheeth%netscape.com 9c6617b147 r=vidur. Add support for a new pref "layout.reflow.async.afterDocLoad" that enables async reflow in the pres shell only after the document is done loading. Cleaned up ExitReflowLock() so that it uses preferences to decide whether to do async reflow, not a parameter. 2000-01-25 16:00:44 +00:00
troy%netscape.com 8a78d916a5 b=21657 a=chofmann@netscape.com Made "overflow list" a full fledged
child list and changed ReparentFrameView() to check it.
2000-01-22 01:16:50 +00:00
hyatt%netscape.com e740cdb714 Fixes for bugs 17460 and 21656. r=travis 2000-01-14 09:28:54 +00:00
nisheeth%netscape.com 9d77060d0e a=donm, r=vidur.
- Added support for two prefs:
a) a boolean pref "layout.reflow.async" that turns asynchronous reflow on/off.  The default is off.
b) an integer pref "layout.reflow.timeslice" that specifies a reflow timeslice, in microseconds.  When asynchronous reflow is on, the reflow command processing loop in ProcessReflowCommands() tries to return to the event loop if the time spent processing reflow commands exceeds this timeslice.  The default value is 1 second.
1999-12-23 02:02:33 +00:00
pierre%netscape.com 6cf30a82f9 20042 "crash on www.netscape.com". Back out my change from v1.30 to fix a crash on InfoWorld which was caused by invalid html code and can no longer be reproduced with their current presentation. 1999-12-02 09:19:16 +00:00
dougt%netscape.com f66be6c30c fix for 18078. r=mstoltz,warren 1999-11-30 00:32:43 +00:00
nisheeth%netscape.com 767ba5b269 Fix for bug 20194. r=beard. Thank to Patrick Beard for finding this leak in CaptureFrameStateFor(). 1999-11-30 00:17:40 +00:00
tbogard%aol.net af1ce8d410 Changed nsIPresContext& to nsIPresContext*. Changed nsEventStatus& to nsEventStatus*. Now more compatible with XPIDL. a=vidur r=scc 1999-11-24 06:03:41 +00:00
pierre%netscape.com 36d3ba78e4 18285: crash on infoworld. When we append a change to the ChangeList, make sure we return its hint, especially if it's a HINT_REFLOW because the reflow may delete some frames that are referred to elsewhere in the ChangeList. r=troy. 1999-11-23 01:08:20 +00:00
dmose%mozilla.org 5312eacf8c updated license boilerplate to xPL 1.1, a=chofmann@netscape.com,r=endico@mozilla.org 1999-11-06 03:40:37 +00:00
troy%netscape.com 1585a751fb Moved a bunch of stuff out to a nsIFrameDebug interface 1999-11-01 22:12:45 +00:00
troy%netscape.com 87afecabc0 Changed SetView/GetView to take an additional argument which is the
pres context
1999-10-26 04:44:41 +00:00
nisheeth%netscape.com 04ee464ef6 r=troy. Siblings of frames get traversed iteratively during capturing/restoring of state. Earlier they were traversed recursively which was causing the call stack to grow too large. 1999-10-23 01:43:49 +00:00
troy%netscape.com 0959b411de r=kipp@netscape.com
Added API to get/set properties on frames. Also changed nsContainerFrame
to use a property to implement the frame overflow list. This saves 4 bytes
per container frames
1999-10-21 05:11:43 +00:00
troy%netscape.com 8ed2cc1d26 b=16781, r=vidur@netscape.com
Re-ordered destructor to delete maps after destroying the frame hierarchy
1999-10-19 19:06:37 +00:00
troy%netscape.com abf8a1ff80 Changes to frame property API (code's not being used yet) 1999-10-19 14:41:55 +00:00
troy%netscape.com c81762bdd1 Change to code that's not currently being used. Before it gets used it
will get code reviewed
1999-10-19 04:52:33 +00:00
troy%netscape.com 153e50eb7f r=kipp@netscape.com Moved ownership of root frame (and frame hierarchy) over to
the frame manager
1999-10-18 22:20:37 +00:00
troy%netscape.com d1f8db9147 b=16615. Two line change. Added check for case of root node being a leaf.
The rest of the changes are NOT yet in use. They willbe reviewed by Kipp before
they're turned on
1999-10-18 19:02:09 +00:00
troy%netscape.com bc981fba0d Added some new code that's not in use yet 1999-10-16 00:55:48 +00:00
peterl%netscape.com 2429e3f7e5 Added code to detect generated content that depends on attributes and reflow when the attribute changes
Fix for Bug 15153
r=pierre
Also added namespace ID to the AttributeChanged method
r=troy
1999-10-15 23:14:44 +00:00
kipp%netscape.com 65f6ba5343 Cleanup moz-decl-counter usage and fix NS_LOG_ADDREF usage 1999-10-08 20:41:19 +00:00
peterl%netscape.com 69c028cae8 fixed leak of undisplayed content map nodes (which leak style contexts)
r=kipp
1999-10-04 23:33:23 +00:00
peterl%netscape.com 88203d51ef Don't access undisplayed content map if it isn't there... 1999-10-02 19:45:18 +00:00
peterl%netscape.com 01c3dcd2fe added multimap to track content that didn't get a frame
Fix for bug 9013 (and any dynamic change from display:none) r=troy
1999-10-02 04:25:29 +00:00
peterl%netscape.com bcb317606a remap style contexts that don't get swapped when computing style changes 1999-09-29 03:32:19 +00:00
peterl%netscape.com fbfc908c89 remove warnings
fix bug 10007 - when style change needs to regenerate frame do not set new context into frame
approved chofman
1999-09-25 05:00:41 +00:00
peterl%netscape.com 2093ce913d made style changes handle floaters and positioned children
added style tree verify
moved re-resolve style logic from frames
1999-09-21 07:53:14 +00:00