August 2010 - Wine-devel - list.winehq.org

Re: user32/dialog.c: GetNextDlgTabItem returns hwndDlg if it does not contain controls.
by Alexandre Julliard 19 Aug '10

19 Aug '10

"Mikhail Maroukhine" <mikolg(a)yandex.ru> writes: > @@ -1695,7 +1696,19 @@ HWND WINAPI GetNextDlgTabItem( HWND hwndDlg, HWND hwndCtrl, > */ > if (!hwndCtrl && fPrevious) return 0; > > - return DIALOG_GetNextTabItem(hwndDlg,hwndDlg,hwndCtrl,fPrevious); > + nextItem = DIALOG_GetNextTabItem(hwndDlg,hwndDlg,hwndCtrl,fPrevious); > + > + /* if next tab item not found return the first child even if it does not have TABSTOP > + Checked on XP > + */ > + if (!nextItem && !fPrevious) nextItem = GetWindow(hwndDlg,GW_CHILD); > + > + /* if there is no controls on hwndDlg return hwndDlg > + Checked on XP > + */ > + if (!nextItem && !fPrevious) nextItem = hwndDlg; > + > + return nextItem; You should test the fPrevious case too, it's probably similar. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: oleau32: Use correct sysking in QueryPathOfRegTypeLib.
by Alexandre Julliard 19 Aug '10

19 Aug '10

Jacek Caban <jacek(a)codeweavers.com> writes: > --- > dlls/oleaut32/typelib.c | 2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) It breaks the 64-bit tests: ../../../../wine/tools/runtest -q -P wine -M oleaut32.dll -T ../../.. -p oleaut32_test.exe.so ../../../../wine/dlls/oleaut32/tests/typelib.c && touch typelib.ok typelib.c:784: Test failed: QueryPathOfRegTypeLib(3.0) returned 8002801d typelib.c:784: Test failed: QueryPathOfRegTypeLib(3.1) returned 8002801d typelib.c:784: Test failed: QueryPathOfRegTypeLib(3.22) returned 8002801d typelib.c:784: Test failed: QueryPathOfRegTypeLib(3.37) returned 8002801d typelib.c:784: Test failed: QueryPathOfRegTypeLib(3.40) returned 8002801d typelib.c:784: Test failed: QueryPathOfRegTypeLib(65535.65535) returned 8002801d make: *** [typelib.ok] Error 6 -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH 5/5] ddraw: Allow creating back buffer for DirectX 1 interfaces. (try 5, resend)
by Stefan Dösinger 19 Aug '10

19 Aug '10

Am 18.08.2010 um 22:01 schrieb Oldřich Jedlička: > + if(DDSD->ddsCaps.dwCaps & DDSCAPS_FRONTBUFFER) { > + WARN("Application tried to create an explicit front buffer\n"); > LeaveCriticalSection(&ddraw_cs); > return DDERR_INVALIDCAPS; > } This test seems redundant now. The ddraw1 thunk removes the flag, and all other thunks check for the frontbuffer flag on their own.

1 0

have all MCI tests in winmm/tests/ or distributed across mci*/tests/?
by Joerg-Cyril.Hoehle＠t-systems.com 19 Aug '10

19 Aug '10

Hi, although my patch creates mcicda/tests/ I'd like to query whether it would make more sense to have all mci tests grouped in winmm/tests/. Pro: + Future fixes to winmm/mci.c would affect winmm/tests/*.c only. E.g. when I'll make "stop all" work I'd have to touch mcicda/tests/mcicda.c as well to eliminate the todo_wine there. Same when I'll patch mciparse to accept [TT:]MM:SS:FF format. + When Maarten will write mciqtz tests, perhaps resources could be reused to execute mciavi tests as well (the file creation code need not be duplicated). Similar tests could be written for: "open foo.avi" (mciavi) and "open foo.avi type mpegvideo" (mciqtz) or "open foo.wav" (micwave) and "open foo.wav type mpegvideo" although there's some difference in behaviour. + The functions test_notification could be in one file only, not repeated. I can rewrite my current mcicda/tests/mcicda.c patch to live in winmm/tests/. What do you think? Jörg Höhle

2 1

Re: [1/2] user32: add a way to poll for messages on extra handles
by Alexandre Julliard 19 Aug '10

19 Aug '10

Damjan Jovanovic <damjan.jov(a)gmail.com> writes: > diff --git a/dlls/user32/driver.c b/dlls/user32/driver.c > index f2f9892..05c83d4 100644 > --- a/dlls/user32/driver.c > +++ b/dlls/user32/driver.c > @@ -126,6 +126,7 @@ static const USER_DRIVER *load_driver(void) > GET_USER_FUNC(WindowMessage); > GET_USER_FUNC(WindowPosChanging); > GET_USER_FUNC(WindowPosChanged); > + GET_USER_FUNC(GetExtraMessageHandles); That's ugly, it doesn't correspond to anything in the Windows API. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH 2/2] ntdll/tests: Add tests for RtlIpv4AddressToString / RtlIpv4AddressToStringEx [try 4]
by Alexandre Julliard 19 Aug '10

19 Aug '10

Detlef Riekenberg <wine.dev(a)web.de> writes: > + ok( (res == (void *)~0) || > + broken(res == (NULL + len)), /* XP and w2003 */ You can't do that. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH] comctl32/tests: DateTimePicker and MonthCalendar controls only support Gregorian calendar.
by Nikolay Sivov 19 Aug '10

19 Aug '10

On Thu, Aug 19, 2010 at 9:50 AM, Austin Lund <austin.lund(a)gmail.com> wrote: > http://www.microsoft.com/middleeast/msdn/ArabicCalendar.aspx Hi, Austin. This information is outdated probably. There's a message to set a calendar for MonthCal at least, see: http://msdn.microsoft.com/en-us/library/bb760995%28VS.85%29.aspx Not sure about DateTimePicker, it uses Monthcal internally so a calendar could still be switched. Anyway this won't work on pre-Vista and without a manifest for v6, but you could enable tests using V6 loading helpers that we have already and MCM_SETCALID message, so tests will run on Arabic or some Asian boxes.

2 1

Re: [PATCH 5/5] ddraw: Allow creating back buffer for DirectX 1 interfaces. (try 5, resend)
by testbot＠testbot.winehq.org 19 Aug '10

19 Aug '10

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=4557 Your paranoid android. === W2KPROSP4 (32 bit visual) === visual.c:1619: Test failed: Got color 000a246a, expected 00ff0040 or near visual.c:1625: Test failed: Got color 006b8bbd, expected 00ff0080 or near === W98SE (32 bit dsurface) === No test summary line found

1 0

Re: [PATCH 1/5] ddraw/tests: Add test for bad size of surface caps in CreateSurface. (try 5, resend)
by testbot＠testbot.winehq.org 19 Aug '10

19 Aug '10

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=4553 Your paranoid android. === W98SE (32 bit dsurface) === No test summary line found

1 0

Re: [PATCH 2/2] gameux/tests: verifying registry values created by GameExplorer (try 2)
by testbot＠testbot.winehq.org 19 Aug '10

19 Aug '10

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=4552 Your paranoid android. === W98SE (32 bit gameexplorer) === Failure running script in VM: Exceeded timeout limit of 315 sec === W2KPROSP4 (32 bit gameexplorer) === Timeout === WXPPROSP3 (32 bit gameexplorer) === Timeout

1 0