30710 - deleting a bookmark crashes on linux

Status: VERIFIED FIXED

(SeaMonkey :: Bookmarks & History, defect, P1)

Description

[Daniel (Leaf) Nunes]

open managing bookmarks window.
select a bookmark.
edit->delete bookmark
select `ok'

Result: app crash.

Comment 1

[Steve Lamm]

Marking for beta1 consideration.

Comment 2

[Cyril Bortolato]

Which build are you using? It WORKSFORME with 2000-03-06-13 on Linux.

Comment 3

[Mike Pinkerton (not reading bugmail)]

hrm, it seems to work now. i swear this didn't last time i tried.

Comment 4

[Claudius Gayle]

*** Bug 30922 has been marked as a duplicate of this bug. ***

Comment 5

[Claudius Gayle]

I reproduced this by typing the backspace key with the 2000030709 linux build.

Comment 6

[Daniel (Leaf) Nunes]

Yikes! this hasn't been reviewed by the pdt?

Comment 7

[Claudius Gayle]

cuz beta1 is a keyword, not a status comment.

Comment 8

[Phil Peterson]

PDT+

Comment 9

[Steve Lamm]

I reproduced this a couple times.
I managed to get a stack trace out of the debugger (below). I noticed something
else that was strange when I ran it in the debugger. If I hit "enter" to dismiss
the confirmation dialog, it would bring up another dialog instead of dismissing
the first one.

#0  0x40a8b38d in InMemoryAssertionEnumeratorImpl::HasMoreElements
(this=0x8a6a860, aResult=0xbfffd0f8) at nsInMemoryDataSource.cpp:360
#1  0x40a95285 in RDFContainerImpl::Renumber (this=0x832a678, aStartIndex=7,
aIncrement=-1) at nsRDFContainer.cpp:582
#2  0x40a94146 in RDFContainerImpl::RemoveElement (this=0x832a678,
aElement=0x83e62a8, aRenumber=1) at nsRDFContainer.cpp:270
#3  0x401a46f9 in XPTC_InvokeByIndex (that=0x832a678, methodIndex=9,
paramCount=2, params=0xbfffd2d4) at xptcinvoke_unixish_x86.cpp:159
#4  0x409b299d in nsXPCWrappedNativeClass::CallWrappedMethod (this=0x891c1b0,
cx=0x86c5288, wrapper=0x8898428, desc=0x8c0bdb4, callMode=CALL_METHOD, argc=2,
argv=0x8817d64, vp=0xbfffd454) at xpcwrappednativeclass.cpp:898
#5  0x409b4997 in WrappedNative_CallMethod (cx=0x86c5288, obj=0x88bec98, argc=2,
argv=0x8817d64, vp=0xbfffd454) at xpcwrappednativejsops.cpp:200
#6  0x400968d7 in js_Invoke (cx=0x86c5288, argc=2, flags=0) at jsinterp.c:665
#7  0x400a6c83 in js_Interpret (cx=0x86c5288, result=0xbfffde1c) at
jsinterp.c:2292
#8  0x40096935 in js_Invoke (cx=0x86c5288, argc=1, flags=0) at jsinterp.c:681
#9  0x400a6c83 in js_Interpret (cx=0x86c5288, result=0xbfffe814) at
jsinterp.c:2292
#10 0x40096935 in js_Invoke (cx=0x86c5288, argc=1, flags=2) at jsinterp.c:681
#11 0x40096c6c in js_InternalInvoke (cx=0x86c5288, obj=0x86291c0,
fval=140677576, flags=0, argc=1, argv=0xbfffea90, rval=0xbfffe984) at
jsinterp.c:754
#12 0x4006aac7 in JS_CallFunctionValue (cx=0x86c5288, obj=0x86291c0,
fval=140677576, argc=1, argv=0xbfffea90, rval=0xbfffe984) at jsapi.c:2790
#13 0x40321dc9 in nsJSContext::CallEventHandler (this=0x86c4ea8,
aTarget=0x86291c0, aHandler=0x86291c8, argc=1, argv=0xbfffea90,
aBoolResult=0xbfffe9e0) at nsJSEnvironment.cpp:562
#14 0x4035c6a9 in nsJSEventListener::HandleEvent (this=0x8806038,
aEvent=0x8b088c4) at nsJSEventListener.cpp:128
#15 0x40f14271 in nsEventListenerManager::HandleEventSubType (this=0x8805ff0,
aListenerStruct=0x879e1e0, aDOMEvent=0x8b088c4, aSubType=8, aPhaseFlags=7) at
nsEventListenerManager.cpp:698
#16 0x40f1617e in nsEventListenerManager::HandleEvent (this=0x8805ff0,
aPresContext=0x86e1180, aEvent=0xbfffefb8, aDOMEvent=0xbfffef58, aFlags=7,
aEventStatus=0xbfffeff8) at nsEventListenerManager.cpp:1457
#17 0x40abb465 in nsXULElement::HandleDOMEvent (this=0x8805f88,
aPresContext=0x86e1180, aEvent=0xbfffefb8, aDOMEvent=0xbfffef58, aFlags=1,
aEventStatus=0xbfffeff8) at nsXULElement.cpp:3191
#18 0x4116633e in nsMenuFrame::Execute (this=0x888c400) at nsMenuFrame.cpp:1204
#19 0x41162ebb in nsMenuFrame::HandleEvent (this=0x888c400,
aPresContext=0x86e1180, aEvent=0xbffff424, aEventStatus=0xbffff320) at
nsMenuFrame.cpp:296
#20 0x40f70e58 in PresShell::HandleEvent (this=0x86475a0, aView=0x8841220,
aEvent=0xbffff424, aEventStatus=0xbffff320) at nsPresShell.cpp:3025

Comment 10

[Robert John Churchill]

This smells a lot like bug # 21006.  Steve, you might want to confer with Chris
(Waterson) about this.

Comment 11

[don]

Hmmmmm, yeah this looks like it could be a dup of bug #21006.

Comment 12

[chris hofmann]

seems to work ok on win32 branch build from mar 15
can someone try mac?

Comment 13

[Claudius Gayle]

this is fine on the Mac 2000031518 beta branch build. It only ever was a problem on linux.

Comment 14

[Steve Lamm]

Yes, this looks like bug #21006 to me too. I am marking that as a blocker. I
will also mark it as beta1.

I will leave this bug open since it is another way to reproduce the bug.

Comment 15

[Steve Lamm]

I am trying out waterson's patch from but 21006.

Comment 16

[Steve Lamm]

Patch worked. At least, I did not crash in 5 runs and about 20 deletes. What I
am not sure is how safe waterson's patch is.

Comment 17

[Chris Waterson]

r=waterson. ;-)

Comment 18

[Jim Roskind]

The last update I had was that RJC had reviewed and improved Waterson's original
patch.  The new-improved patch is now being passed around for additional review.
 I'm hoping this can land by late Sunday night, and we can have it in our Monday
morning verification build.

Comment 19

[Robert John Churchill]

*** Bug 31457 has been marked as a duplicate of this bug. ***

Comment 20

[Chris Waterson]

This appears to have been fixed by 21006.

Comment 21

[Claudius Gayle]

VERIFIED with the 2000032006 linux branch build