Closed Bug 233197 Opened 21 years ago Closed 21 years ago

[FIX]Warning about rePOSTing when actually not rePOSTing something

Categories

(Core :: Networking: HTTP, defect, P1)

defect

Tracking

()

VERIFIED FIXED
mozilla1.7alpha

People

(Reporter: mcsmurf, Assigned: bzbarsky)

Details

Attachments

(1 file)

In newest builds (i think this broke about 3 days ago, but bonsai doesn't tell something useful in this timeframe, maybe it's also some other Component) there this warning sometimes comes up after POSTing a form and waiting to load the new page: "The page you are trying to view contains POSTDATA. If you resend the data, any action the form carried out (such as a search or a online purchase) will be repeated. To resend the data, click OK. Otherwise, click Cancel." But i didn't click the Submit button on the form again, i just waited that it loads the resulting page. This happens on win2k with a current cvs trunk build. To reproduce: 1. Go to the test Bug 182019 (you won't spam anyone except me and that's ok *g*) 2. Type in Add CC: opi 3. Click Submit (Commit) 4. Watch the error coming up while loading the list of names that fit to opi
This worksforme with a Linux trunk build pulled at "Wed Feb 4 03:53:37 CST 2004" What are your cache settings? Can you narrow down the exact regression date?
Regression date, no, i just noticed this while working with Bugzilla and i don't have any builds available on my PC except this one cvs trunk build (i might test Mozilla 1.6 somewhat later, maybe that's of some use). Cache is enabled, I have a disc cache of 20MB, it compares the page every time it is out of date and Link Prefetching is enabled. I noticed this error alway comes up when the listbox with the 100 entries max. has been rendered (it seems to load the rest of the page in the background!). When i now click Cancel it stops rendering, when i click OK, it just immediatelly displays the rest of the page.
Mozilla 1.6 works fine.
Frank, the FTP server has daily builds going back a ways, so if you can try those, that would be much appreciated. Your cache settings are identical to mine...
Ok, tried some, broke between 2004-01-20-08 and 2004-02-03-17. Somehow nightlys for Windows are missing in this timeframe and i don't actually want to test installer builds...
About 40000 lines of code changed in that time period... none of the changes look like they should really cause this. There are SVG branch builds from the time period when we have no windows zip coverage.... perhaps you can try those? Nothing but svg should really differ... see eg ftp://ftp.mozilla.org/pub/mozilla.org/mozilla/nightly/2003-12-28-11-trunk/
Ok, broke between 2004-01-26-16 and 2004-01-28-16. There was no build from 27th (the build from the 27th was actually the same as the one from 26th).
Weird... I still can't reproduce this with a current Linux trunk build.... biesi, do you see this?
With a debug build i get when i press Cancel (this is the whole page load, not only the log after after clicking Cancel): Error loading URL http://bugzilla.mozilla.org/process_bug.cgi : 804b0002 ++WEBSHELL == 5 ++DOMWINDOW == 5 WARNING: getting z level of unregistered window, file e:/mozilla/tree6/mozilla/x pfe/appshell/src/nsWindowMediator.cpp, line 636 WARNING: getting z level of unregistered window, file e:/mozilla/tree6/mozilla/x pfe/appshell/src/nsWindowMediator.cpp, line 636 WARNING: nsTimeoutImpl::Release() proceeding without context., file e:/mozilla/t ree6/mozilla/dom/src/base/nsGlobalWindow.cpp, line 5336 --WEBSHELL == 4 --DOMWINDOW == 4 And this when clicking OK: Error loading URL http://bugzilla.mozilla.org/process_bug.cgi : 804b0002 ++WEBSHELL == 6 ++DOMWINDOW == 6 WARNING: getting z level of unregistered window, file e:/mozilla/tree6/mozilla/x pfe/appshell/src/nsWindowMediator.cpp, line 636 WARNING: getting z level of unregistered window, file e:/mozilla/tree6/mozilla/x pfe/appshell/src/nsWindowMediator.cpp, line 636 WARNING: nsTimeoutImpl::Release() proceeding without context., file e:/mozilla/t ree6/mozilla/dom/src/base/nsGlobalWindow.cpp, line 5336 --WEBSHELL == 5 Document http://bugzilla.mozilla.org/process_bug.cgi loaded successfully --DOMWINDOW == 5 maybe it helps...
last one...i got a stacktrace up to the point where the OK Cancel dialog appears, maybe this might also help: USER32! 77e4868b() nsXULWindow::ShowModal(nsXULWindow * const 0x047ff2d8) line 362 + 31 bytes nsWebShellWindow::ShowModal(nsWebShellWindow * const 0x047ff2d8) line 1106 nsContentTreeOwner::ShowAsModal(nsContentTreeOwner * const 0x0460b52c) line 449 nsWindowWatcher::OpenWindowJS(nsWindowWatcher * const 0x00f651ec, nsIDOMWindow * 0x00f9b7bc, const char * 0x012ed884 kPromptURL, const char * 0x012f343c, const char * 0x012f3418, int 1, unsigned int 1, long * 0x045f6740, nsIDOMWindow * * 0x0012f5ac) line 794 nsWindowWatcher::OpenWindow(nsWindowWatcher * const 0x00f651e8, nsIDOMWindow * 0x00f9b7bc, const char * 0x012ed884 kPromptURL, const char * 0x012f343c, const char * 0x012f3418, nsISupports * 0x03b779f8, nsIDOMWindow * * 0x0012f5ac) line 456 + 48 bytes nsPromptService::DoDialog(nsPromptService * const 0x02af12e4, nsIDOMWindow * 0x00f9b7bc, nsIDialogParamBlock * 0x03b779f8, const char * 0x012ed884 kPromptURL) line 631 + 77 bytes nsPromptService::Confirm(nsPromptService * const 0x02af12e0, nsIDOMWindow * 0x00f9b7bc, const unsigned short * 0x038fbbe0, const unsigned short * 0x045f6558, int * 0x0012f758) line 209 + 37 bytes nsPrompt::Confirm(nsPrompt * const 0x046142a0, const unsigned short * 0x00000000, const unsigned short * 0x045f6558, int * 0x0012f758) line 141 nsDocShell::LoadHistoryEntry(nsDocShell * const 0x03bc0c80, nsISHEntry * 0x04820e6c, unsigned int 67108866) line 6249 nsDocShell::Reload(nsDocShell * const 0x03bc0c90, unsigned int 1024) line 2737 + 38 bytes nsWebShell::ReloadDocument(nsWebShell * const 0x03bc0dd0, const char * 0x0360fde4 `string', int 7) line 383 + 27 bytes nsMyObserver::Notify(nsMyObserver * const 0x03b7bf18, const char * 0x0360fde4 `string', nsDetectionConfident eBestAnswer) line 68 + 32 bytes nsUniversalXPCOMDetector::Report(const char * 0x0360fde4 `string') line 355 nsUniversalDetector::DataEnd() line 285 nsUniversalXPCOMDetector::Done(nsUniversalXPCOMDetector * const 0x03705174) line 343 nsDetectionAdaptor::Finish(nsDetectionAdaptor * const 0x03b7bed0) line 169 + 26 bytes nsParser::OnStopRequest(nsParser * const 0x03b7bddc, nsIRequest * 0x03d66490, nsISupports * 0x00000000, unsigned int 0) line 2471 nsDocumentOpenInfo::OnStopRequest(nsDocumentOpenInfo * const 0x045d37a0, nsIRequest * 0x03d66490, nsISupports * 0x00000000, unsigned int 0) line 353 nsStreamListenerTee::OnStopRequest(nsStreamListenerTee * const 0x0490a498, nsIRequest * 0x03d66490, nsISupports * 0x00000000, unsigned int 0) line 66 nsHttpChannel::OnStopRequest(nsHttpChannel * const 0x03d66498, nsIRequest * 0x03cbef90, nsISupports * 0x00000000, unsigned int 0) line 3392 nsInputStreamPump::OnStateStop() line 499 nsInputStreamPump::OnInputStreamReady(nsInputStreamPump * const 0x03cbef94, nsIAsyncInputStream * 0x03cbed74) line 339 + 11 bytes nsInputStreamReadyEvent::EventHandler(PLEvent * 0x03b6135c) line 119 PL_HandleEvent(PLEvent * 0x03b6135c) line 671 + 10 bytes PL_ProcessPendingEvents(PLEventQueue * 0x00f14890) line 606 + 9 bytes _md_EventReceiverProc(HWND__ * 0x001b021c, unsigned int 49379, unsigned int 0, long 15812752) line 1412 + 9 bytes USER32! 77e2a2d0() USER32! 77e045e5() USER32! 77e0a816() nsAppShellService::Run(nsAppShellService * const 0x00f69660) line 484 main1(int 1, char * * 0x00262638, nsISupports * 0x00f435a8) line 1291 + 32 bytes main(int 1, char * * 0x00262638) line 1678 + 37 bytes mainCRTStartup() line 338 + 17 bytes KERNEL32! 77e987e7()
this works for me on linux... I can't test windows right now
Frank, what are your charset detection settings set to?
This is a regression from bug 202636 -- we shouldn't be autodetecting on POST pages.
Assignee: darin → bzbarsky
OS: Windows 2000 → All
Priority: -- → P1
Hardware: PC → All
Summary: Warning about rePOSTing when actually not rePOSTing something → [FIX]Warning about rePOSTing when actually not rePOSTing something
Target Milestone: --- → mozilla1.7alpha
Attached patch FixSplinter Review
Attachment #140808 - Flags: superreview?(jst)
Attachment #140808 - Flags: review?(jst)
Comment on attachment 140808 [details] [diff] [review] Fix r+sr=jst
Attachment #140808 - Flags: superreview?(jst)
Attachment #140808 - Flags: superreview+
Attachment #140808 - Flags: review?(jst)
Attachment #140808 - Flags: review+
Fixed.
Status: NEW → RESOLVED
Closed: 21 years ago
Resolution: --- → FIXED
yup V. with cvs trunk build :)
Status: RESOLVED → VERIFIED
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: