54512 - assertion every time a site spawns a pop-up window

Status: VERIFIED WORKSFORME

(Core Graveyard :: Embedding: APIs, defect, P3)

Description

[buster]

using debug bits from 9/27/00, WinNT
using viewer, every time a site spawns a pop-up window (typically for an ad), we
get the following assertion:

###!!! ASSERTION: NS_ENSURE_TRUE(prompter) failed: 'prompter', file
S:\mozilla\dom\src\base\nsGlobalWindow.cpp, line 742

GlobalWindowImpl::GetPrompter() line 742 + 38 bytes
nsCookieService::SetCookieString() line 119 + 47 bytes
nsHTMLDocument::SetCookie() line 1989 + 69 bytes
SetHTMLDocumentProperty() line 511 + 22 bytes
js_SetProperty() line 2349 + 195 bytes
js_Interpret() line 2480 + 1578 bytes
js_Execute() line 992 + 13 bytes
JS_EvaluateUCScriptForPrincipals() line 3146 + 27 bytes
nsJSContext::EvaluateString() line 583 + 68 bytes
HTMLContentSink::EvaluateScript() line 4610
HTMLContentSink::ProcessSCRIPTTag() line 4959
HTMLContentSink::AddLeaf() line 3133 + 12 bytes
CNavDTD::AddLeaf() line 3621 + 22 bytes
CNavDTD::HandleScriptToken() line 2073 + 12 bytes
CNavDTD::OpenContainer() line 3290 + 12 bytes
CNavDTD::HandleDefaultStartToken() line 1141 + 20 bytes
CNavDTD::HandleStartToken() line 1569 + 22 bytes
CNavDTD::HandleToken() line 745 + 12 bytes
CNavDTD::BuildModel() line 485 + 20 bytes
nsParser::BuildModel() line 2010 + 34 bytes
nsParser::ResumeParse() line 1891 + 11 bytes
nsParser::EnableParser() line 1528 + 15 bytes
HTMLContentSink::ResumeParsing() line 4543 + 19 bytes
HTMLContentSink::OnStreamComplete() line 4740 + 11 bytes
nsStreamLoader::OnStopRequest() line 121 + 78 bytes
nsHTTPFinalListener::OnStopRequest() line 1158 + 42 bytes
InterceptStreamListener::OnStopRequest() line 1199
nsHTTPChannel::ResponseCompleted() line 1877 + 42 bytes
nsHTTPServerListener::OnStopRequest() line 729
nsOnStopRequestEvent::HandleEvent() line 302
nsStreamListenerEvent::HandlePLEvent() line 97 + 12 bytes
PL_HandleEvent() line 575 + 10 bytes
PL_ProcessPendingEvents() line 508 + 9 bytes
_md_EventReceiverProc() line 1044 + 9 bytes
USER32! 77e71820()
00c82b10()

Comment 1

[buster]

This assertion seems to often, but not always, accompany the other:

###!!! ASSERTION: NS_ENSURE_TRUE(mBrowserWindow->mStatus) failed:
'mBrowserWindow->mStatus', file
S:\mozilla\webshell\tests\viewer\nsWebBrowserChrome.cpp, line 85

I think these assertions are probably too strong. A null prompter should
probably be legal.

nsDebug::Assertion() line 256 + 13 bytes
nsDebug::WarnIfFalse() line 358 + 21 bytes
nsWebBrowserChrome::SetStatus() line 85 + 35 bytes
nsWebShell::OnOverLink() line 858
nsGenericElement::TriggerLink() line 2425
nsGenericHTMLElement::HandleDOMEventForAnchors() line 1202 + 40 bytes
nsHTMLAreaElement::HandleDOMEvent() line 235
nsEventStateManager::GenerateMouseEnterExit() line 1519
nsEventStateManager::PreHandleEvent() line 306
PresShell::HandleEventInternal() line 4249 + 43 bytes
PresShell::HandleEvent() line 4190 + 25 bytes
nsView::HandleEvent() line 379
nsView::HandleEvent() line 352
nsView::HandleEvent() line 352
nsViewManager2::DispatchEvent() line 1429
HandleEvent() line 68
nsWindow::DispatchEvent() line 681 + 10 bytes
nsWindow::DispatchWindowEvent() line 702
nsWindow::DispatchMouseEvent() line 3890 + 21 bytes
ChildWindow::DispatchMouseEvent() line 4100
nsWindow::ProcessMessage() line 2937 + 24 bytes
nsWindow::WindowProc() line 950 + 27 bytes

Comment 2

[Ann Adamcik]

We're seeing something similar with Solaris and Gnome:

#0  nsHTTPResponse::GetContentLength (this=0x0,
o_ContentLength=0xffbeec5c) at nsHTTPResponse.cpp:115
#1  0xfd3da1f4 in nsHTTPServerListener::OnDataAvailable (this=0x728cf8,
channel=0x765854, context=0x67b318, 
    i_pStream=0x72af30, i_SourceOffset=0, i_Length=2167) at
nsHTTPResponseListener.cpp:557
#2  0xfd3447f8 in nsOnDataAvailableEvent::HandleEvent (this=0x791ca8) at
nsAsyncStreamListener.cpp:400
#3  0xfd3433a0 in nsStreamListenerEvent::HandlePLEvent (aEvent=0x742940)
at nsAsyncStreamListener.cpp:97
#4  0xff0a9344 in PL_HandleEvent (self=0x742940) at plevent.c:575
#5  0xff0a9070 in PL_ProcessPendingEvents (self=0xaf750) at
plevent.c:508
#6  0xff0abb1c in nsEventQueueImpl::ProcessPendingEvents (this=0xa7420)
at nsEventQueue.cpp:356
#7  0xfce5dbb4 in event_processor_callback (data=0xa7420, source=5,
condition=GDK_INPUT_READ) at nsAppShell.cpp:158
#8  0xfce5d604 in our_gdk_io_invoke (source=0x1ef7a0, condition=G_IO_IN,
data=0x1bb188) at nsAppShell.cpp:58
#9  0xfcdc5a10 in g_io_unix_dispatch (source_data=0x1e2ee8,
current_time=0xffbef208, user_data=0x1bb188) at giounix.c:135
#10 0xfcdc76ec in g_main_dispatch (current_time=0xffbef208) at
gmain.c:656
#11 0xfcdc7f74 in g_main_iterate (block=-52507388, dispatch=1) at
gmain.c:874
#12 0xfcdc8188 in g_main_run (loop=0x16f8a8) at gmain.c:932
#13 0xfccc9b54 in gtk_main () at gtkmain.c:476
#14 0xfce5e544 in nsAppShell::Run (this=0x82b78) at nsAppShell.cpp:335
#15 0xfdaddf40 in nsAppShellService::Run (this=0xe45f8) at
nsAppShellService.cpp:407
#16 0x29a44 in main1 (argc=1, argv=0xffbef69c, nativeApp=0x0) at
nsAppRunner.cpp:958
#17 0x2a468 in main (argc=1, argv=0xffbef69c) at nsAppRunner.cpp:1139

Comment 3

[buster]

Ann: I think I'd submit that as a separate bug.  The stacks do not look at all
similar to me.

Comment 4

[Judson Valeski]

over to dan. window/prompter retrieval problem.

Comment 5

[Paul Wyskoczka]

Updating QA Contact

Comment 6

[Judson Valeski]

is this still a problem?

Comment 7

[ian holsman]

no.
The inclusion of the Iprompt interface fixed this.
If you can close it please.

Comment 8

[Judson Valeski]

gone.

Comment 9

[Michael Dunn]

Correction: Changing QA contact for the Embed API bugs to David Epstein.

Comment 10

[David Epstein]

no assertion for popups. Verifying.