Commit Graph

1739 Commits

Author SHA1 Message Date
dbaron%fas.harvard.edu
b07eae2b6f 1) Make PRE line layout backwards compatible in quirks mode (related to bug 26998).
2) Don't use strings when I can use atoms.
3) Fix bug 28472 by correcting error in earlier checkin.
r=buster@netscape.com
2000-03-16 01:14:57 +00:00
troy%netscape.com
4f7d1e3f2a b=24343 r=karnaze@netscape.com Changes so we properly lay out after changing
the font using the pref UI. Block and table code changed to handle
eReflowReason_StyleChange
2000-03-15 15:16:03 +00:00
troy%netscape.com
8a75ec0c9b b=24506 r=buster@netscape.com When calculating the width of a block frame
and it's an unconstrained reflow, then make sure to include the right
margin
2000-03-14 04:03:33 +00:00
rickg%netscape.com
ffe78e83e5 removed calls to deprecated string methods 2000-03-12 09:14:14 +00:00
troy%netscape.com
3b124abc37 b=5588. Changed it so we lazily reparent the child frames on the overflow list
when it's our initial reflow
2000-03-12 03:14:11 +00:00
troy%netscape.com
1bc0e305c9 b=5588. Added new function ReparentFrameViewList() that is much faster
then ReparentFrameView() when you have a list of child frames
2000-03-12 03:09:36 +00:00
troy%netscape.com
0c7b650a86 b=5588. Performance improvement. Avoids walking the child list
twice when inserting frames at the head of the list
2000-03-12 03:07:52 +00:00
troy%netscape.com
6464eafc57 Now line boxes are allocated out of the pres shell arena. This reduces the
overhead of memory allocation, and it gets us recycling which speeds up
resize reflows
2000-03-12 03:00:51 +00:00
troy%netscape.com
dfd7864b4c b=30391 r=troy@netscape.com Patch from rbs@maths.uq.edu.au to add parens when
necessary to PR_ABS macro
2000-03-11 16:50:52 +00:00
norris%netscape.com
060e388a6b Files:
caps/idl/nsICertificatePrincipal.idl
	caps/idl/nsIPrincipal.idl
	caps/src/nsBasePrincipal.cpp
Implement the ability to manipulate multiple capabilties simultaneously.
r=mstoltz@netscape.com

Files:
	caps/src/nsCodebasePrincipal.cpp
Codebase equality should be based upon origin, not full path.
r=mstoltz@netscape.com

Files:
	caps/src/nsScriptSecurityManager.cpp
Change URI checking to deny based upon scheme rather than allow based upon
scheme for greater flexibility.
r=mstoltz@netscape.com

Files:
	dom/public/nsDOMPropEnums.h
	dom/public/nsDOMPropNames.h
	dom/src/base/nsGlobalWindow.cpp
	modules/libpref/src/init/all.js
Fix bug 20469 Seeing JS functions and global variables from arbitrary host
r=vidur@netscape.com

Files:
	dom/src/base/nsJSUtils.cpp
	dom/src/base/nsJSUtils.h
	dom/src/base/nsJSEnvironment.cpp
	dom/tools/JSStubGen.cpp
	layout/base/src/nsDocument.cpp
	layout/html/content/src/nsGenericHTMLElement.cpp
Improve performance by removing NS_WITH_SERVICE call for every DOM access.
Propagate XPCOM failure codes out properly.
r=vidur@netscape.com

Files:
	layout/html/document/src/nsFrameFrame.cpp
Fix 27387 Circumventing Same Origin security policy using setAttribute
r=vidur@netscape.com
2000-03-11 06:32:42 +00:00
tbogard%aol.net
7b6b143608 Use the nsIWebBrowserChrome interface to set the status rather than the old nsIBrowserWindow interface. 2000-03-11 00:41:57 +00:00
tbogard%aol.net
bf5156a460 Changed nsHTMLFrameInnerFrame to hold on to a subShell that is an nsIBaseWindow instead of a webShell. This paves the way for holding different types of objects. Also gets rid of that stupid nsIWebShell interface. :) When creating the docShell use the new docShell interfaces rather than the old webShell ones. 2000-03-11 00:41:04 +00:00
joki%netscape.com
b822db9d33 Fix bug 29515, prevent excess processing inside GetFrameForPoint. a:rickg, r:dbaron@fas.harvard.edu 2000-03-09 15:31:44 +00:00
evaughan%netscape.com
bfe3296fa7 Fix for bug# 30548
-r hyatt, -a jar
2000-03-09 01:37:23 +00:00
warren%netscape.com
c0497e31be Bug 21556: Making linux be thread-safe. Making tons of classes implement threadsafe AddRef/Release. a=jar 2000-03-05 21:26:01 +00:00
buster%netscape.com
8076fee6b0 fixed bug 27161 by adding interfacs onto nsIPresShell to get/set the "subshell" for an nsIContent
this part of the fix is temporary, but sufficient for beta.  it calls the pres shell to do the initial association.
after beta, this will be done by the frame construction code.
r=travis
a=jar
2000-03-03 23:16:43 +00:00
tbogard%aol.net
4eec4fb506 nsIPref is no longer passed around through inits on webshell, content viewers and presContext. In places where it is needed, it can be retrieved from the service manager. I removed a number of the getters and setters for prefs as the prefs will be the same from the service manager so we don't need getters. r=waterson a=rickg 2000-03-02 07:13:02 +00:00
evaughan%netscape.com
cf0d42fc25 bugs fixes for: 28460, 27242, 29462
-r hyatt, -a jar
2000-03-02 03:01:30 +00:00
av%netscape.com
0271ec9277 Fix for 27486, r=stanley.ho@eng.sun.com, a=rickg 2000-03-02 01:57:14 +00:00
pollmann%netscape.com
982af48162 Bug 28670: Move scroll methods from webshell to docshell r=travis a=rickg 2000-02-29 06:51:48 +00:00
av%netscape.com
035f4468fc Fix for 27486, r=rickg, a=rickg 2000-02-26 00:14:16 +00:00
evaughan%netscape.com
9853879e89 Fixes for bugs: 28460, 27573, 27242, 24721
-r waterson -a jar
2000-02-25 04:18:34 +00:00
av%netscape.com
12bfe22dc4 Fixing 27486, r=beard 2000-02-25 01:02:53 +00:00
norris%netscape.com
d64387736b Fix 28612 META Refresh allowed in Mail/News
r=mstoltz,a=jar
Fix 28658 File upload vulnerability
r=vidur,a=jar
2000-02-23 22:34:40 +00:00
buster%netscape.com
6ad008c507 fixed bug 25510 and 28084
r=troy
a=jar
2000-02-22 22:06:19 +00:00
buster%netscape.com
f3317d4cfe added an assertion that will catch regressions associated with bug 25510
r=troy
a=jar
2000-02-22 22:05:04 +00:00
dbaron%fas.harvard.edu
2fc933f4a5 Refix bug 5821, which I broke with a mistake in my earlier change. r=buster@netscape.com a=rickg@netscape.com 2000-02-19 03:42:30 +00:00
pollmann%netscape.com
1d66570c0c Bug 14827: Adding a comment, removing unused code r=harishd a=rickg 2000-02-19 02:32:38 +00:00
jdunn%netscape.com
080142580c Fix problem with NS_IF_ADDREF on member variables inside of const
member functions.  This includes removing all references to
NS_DEF_PTR and the including of nsIPtr.h

r=troy@netscape.com,dougt@netscape.com,scc@netscape.com
# 28114
a=chofmann@netscape.com
2000-02-17 23:19:18 +00:00
buster%netscape.com
72482e5f58 bug 27927
r=troy
2000-02-16 04:11:37 +00:00
av%netscape.com
31d1739ac9 Added check for null pointer to avoid possible crashes, could be reason for 27558 2000-02-16 01:49:42 +00:00
buster%netscape.com
a536a762d8 bug 22853
absolute containing block was not checking frame bit, so sometimes contained block would not get an initial reflow
2000-02-16 01:05:38 +00:00
buster%netscape.com
5665393ec2 continuing work on bug 25510
r=troy
2000-02-16 00:50:06 +00:00
pinkerton%netscape.com
15c85ef5ee removed obsolete (and ifdef'd out) d&d test code. r=saari. 2000-02-15 07:13:21 +00:00
dbaron%fas.harvard.edu
e0c71d1e53 Fix bug 15428 as suggested by VYV03354@nifty.ne.jp . Stop using 0 as the initial values for min and max of inline box contents so that lines that do not (vertically) contain their own baseline are sized correctly. Also, correctly size BR frames using their line-height (for strict-mode only).
Fix bugs 24186, 26996, and 26998 by reworking the quirks-mode line-height handling so that inline elements that do not have text as children (or meet a few other conditions that show the author wants real CSS support) will not increase the size of the line, but will be as close to their correct size as possible without enlarging the line.  This fix includes emulation of the strange Nav4/IE quirk that a block's line-height is enforced on the first line of LI elements and the last line of LI, DD, and DT elements.  Many of the changes are removing an earlier fix for 24186.

r=buster
2000-02-15 04:26:44 +00:00
buster%netscape.com
66696f0a23 bug 26840 r=mjudge 2000-02-14 01:54:20 +00:00
buster%netscape.com
9c7df3bb31 bug 25510 r=troy
bug 26289 for roc+moz@cs.cmu.edu, r=buster
2000-02-14 01:52:22 +00:00
troy%netscape.com
e548878b47 b=26184 r=karnaze@netscape.com Changed it so that for a percentage width based
floater if the containing block width is unconstrained we use NS_UNCONSTRAINEDSIZE
for the floater's computed width. We were using 0 and this was causing table column
widths to be incorrect
2000-02-13 17:55:18 +00:00
troy%netscape.com
26a77156ec b=27043 r=buster@netscape.com Inner table cell specifies NS_BLOCK_MARGIN_ROOT
flag now
2000-02-11 15:31:41 +00:00
nisheeth%netscape.com
99f85886b8 r=Troy. We were specifying the incorrect child as a reflow dirty target in some cases. 2000-02-11 07:11:43 +00:00
vidur%netscape.com
d4b67b3219 Fix for bug 9140. Area elements now have event fired for them. r=joki 2000-02-11 01:24:59 +00:00
evaughan%netscape.com
c1c4f7681f 1) Fixed freeze because of sync drawing in forms
2) Fixed fieldset blocker

-r rods, granrose
2000-02-10 21:36:28 +00:00
mjudge%netscape.com
90364fa4e5 fix for memory leak bug 24332 2000-02-10 07:55:34 +00:00
mjudge%netscape.com
b3444d6801 fix for bug 18528. fixing tabs controls in selection guessing. 2000-02-10 07:48:17 +00:00
tbogard%aol.net
4c46df4275 Made sure we set the tree item before adding it to the parent. Then we also must notify the tree owner of the addition of the content shell after it has been added. This the bugs having to do with treeOwner showing up (27087, 2719, 25951). r=hyatt 2000-02-10 07:04:09 +00:00
mjudge%netscape.com
b32999a1de big checkin. fixing problems blocking Cmanske (also a reviewer). Checking in preliminary for gen content.(nothing damaged yet) lots of bug numbers not going to look them up, the tree just opened. 2000-02-10 04:35:51 +00:00
evaughan%netscape.com
b361107ebe 1) implememted box reflow coelescing.
2) implemented gfx scrollbars for list boxes
3) fixed progess meter to be an animated gif
4) fixed bugs 23521, 24721, 19114, 20546, 24385, 24457, 23156, 20226, 22543

-r hyatt, troy, rod
2000-02-09 22:02:40 +00:00
troy%netscape.com
a1ed8d02de b=26727 r=jfrancis@netscape.com Changed MarkLineDirty() to always mark the
line dirty even if it also marks the preceding line dirty
2000-02-07 15:36:08 +00:00
dbaron%fas.harvard.edu
bf2b8a41ed Event Target Debugging (b=18175), a debugging feature to help with fixing bugs where events are going to the wrong elements. It shows (like Visual Debugging) which frame is receiving events, and one can therefore move the mouse over the page to see what frame gets events. All the code is #ifdef NS_DEBUG. However, it is not yet hooked up to the viewer menu (to be checked in later).
r=joki
2000-02-03 02:49:58 +00:00
troy%netscape.com
792c9c1053 r=vidur@netscape.com Performance. Added NS_FRAME_HAS_VIEW frame state flag
so we can quickly tell whether the frame has an associated view
2000-02-03 00:51:37 +00:00