August 2008 - Wine-devel - list.winehq.org

Re: Winscard: Correct the declaration of SCardReleaseStartedEvent to match MSDN
by Dmitry Timoshkov 27 Aug '08

27 Aug '08

"Mounir IDRASSI" <mounir.idrassi(a)idrix.net> wrote: > This patch corrects the declaration of the function > SCardReleaseStartedEvent of Winscard to match its definition in MSDN found > in the following link : > > http://msdn.microsoft.com/en-us/library/aa379799(VS.85).aspx > -void WINAPI SCardReleaseStartedEvent(void); > +void WINAPI SCardReleaseStartedEvent(HANDLE); This doesn't match an actual declaration used by PSDK. It's better to use PSDK for a reference, not an online documentation since the latter one is proved to have many typos. -- Dmitry.

2 1

Re: user32: Some changes in minimizing MDI windows (bug #1719)
by Dmitry Timoshkov 27 Aug '08

27 Aug '08

"Ilya Shpigor" <shpigor(a)etersoft.ru> wrote: These all changes need separate patches with test cases along with explanations why each of them is needed. > --- a/dlls/user32/icontitle.c > +++ b/dlls/user32/icontitle.c > @@ -217,6 +217,7 @@ LRESULT WINAPI IconTitleWndProc( HWND hWnd, UINT msg, > case WM_NCHITTEST: > return HTCAPTION; > case WM_NCMOUSEMOVE: > + case WM_NCLBUTTONDOWN: > case WM_NCLBUTTONDBLCLK: > return SendMessageW( owner, msg, wParam, lParam ); This one looks like an obvious one, but needs an explanation. > diff --git a/dlls/user32/mdi.c b/dlls/user32/mdi.c > index 8bffeb0..1967f0c 100644 > --- a/dlls/user32/mdi.c > +++ b/dlls/user32/mdi.c > @@ -1499,6 +1499,7 @@ LRESULT WINAPI DefMDIChildProcA( HWND hwnd, UINT message, > case WM_NEXTMENU: > case WM_SYSCHAR: > case WM_DESTROY: > + case WM_ICONERASEBKGND: > return DefMDIChildProcW( hwnd, message, wParam, lParam ); > } > return DefWindowProcA(hwnd, message, wParam, lParam); > @@ -1656,6 +1657,16 @@ LRESULT WINAPI DefMDIChildProcW( HWND hwnd, UINT message, > } > break; > > + case WM_ICONERASEBKGND: > + { > + RECT rect; > + HDC hdc = (HDC)wParam; > + HBRUSH hbr = CreateSolidBrush( GetSysColor(COLOR_BTNFACE) ); > + GetClipBox( hdc, &rect ); > + FillRect( hdc, &rect, hbr ); > + return 1; > + } Why the job done by DefWindowProc is not enough? > --- a/dlls/user32/winpos.c > +++ b/dlls/user32/winpos.c > @@ -823,7 +823,7 @@ static POINT WINPOS_FindIconPos( HWND hwnd, POINT pt ) > > hrgn = CreateRectRgn( 0, 0, 0, 0 ); > tmp = CreateRectRgn( 0, 0, 0, 0 ); > - for (child = GetWindow( parent, GW_HWNDFIRST ); child; child = GetWindow( child, GW_HWNDNEXT )) > + for (child = GetWindow( parent, GW_CHILD ); child; child = GetWindow( child, GW_HWNDNEXT )) > { > WND *childPtr; > if (child == hwnd) continue; This change looks wrong. > @@ -2033,6 +2033,15 @@ BOOL USER_SetWindowPos( WINDOWPOS * winpos ) > SetForegroundWindow( winpos->hwnd ); > } > > + /* If the window is iconic don't any > + * sizing are needed. bug #1719 > + */ > + if( IsIconic(winpos->hwnd) ) > + { > + winpos->flags |= SWP_NOCLIENTSIZE; > + winpos->flags &= ~SWP_STATECHANGED; > + } > + This requires a test case and an explanation why it's needed. -- Dmitry.

1 0

msi/automation tests on win9x
by Paul Vriens 27 Aug '08

27 Aug '08

Hi, Every time these tests run I get a message box on win9x stating that the service can't be created. Makes sense as we don't have services on win9x. Can we skip all tests if we are on win9x (check for OpenSCManager being not implemented)? -- Cheers, Paul.

2 1

Re: [2/2] user32/tests: Test case for "Fixed bug in loading .cur files"
by Paul Vriens 27 Aug '08

27 Aug '08

JIM CAMERON wrote: > The patch for cursor loading caused some existing > tests to succeed inside todo blocks. This patch > removes those todo and adds a regression test for CreateIconFromResource(). > > > ------------------------------------------------------------------------ > > Hi, That means the tests on Wine will 'fail' without this patch. You should add the removal of the todo_wine's to the first patch. -- Cheers, Paul.

1 0

Re: [ntdll/tests] Fix the SystemHandleInformation tests
by Alexandre Julliard 27 Aug '08

27 Aug '08

Paul Vriens <paul.vriens.wine(a)gmail.com> writes: > The last changes introduced an error on anything but Vista and skipped > the test for the number of handles on all platforms. This is breaking it on Vista again: info.c:411: Test failed: Unexpected ReturnLength : 32 info.c:424: Test failed: Expected STATUS_SUCCESS, got c0000004 -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: regedit: Fix concat handling for ANSI files (try2)
by Alexandre Julliard 27 Aug '08

27 Aug '08

Alexander Nicolaysen Sørnes <alex(a)thehandofagony.com> writes: > + if(c == EOF) > fprintf(stderr,"%s: ERROR - invalid continuation.\n", > getAppName()); > + > + *s = c; > + s++; It's better, but I don't think you want to store EOF in the result string... -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: taskmgr [try 6] [1/2]: Converted some functions and variables to Unicode
by Alexandre Julliard 27 Aug '08

27 Aug '08

Vladimir Pankratov <scriptkid(a)mail.ru> writes: > @@ -93,7 +100,7 @@ > ULONG Idx, Idx2; > HANDLE hProcess; > HANDLE hProcessToken; > - TCHAR szTemp[MAX_PATH]; > + WCHAR wszTemp[256]; > DWORD dwSize; > SYSTEM_PERFORMANCE_INFORMATION SysPerfInfo; > SYSTEM_TIME_INFORMATION SysTimeInfo; > @@ -290,22 +297,10 @@ > if (hProcess) { > if (OpenProcessToken(hProcess, TOKEN_QUERY|TOKEN_DUPLICATE|TOKEN_IMPERSONATE, &hProcessToken)) { > ImpersonateLoggedOnUser(hProcessToken); > - memset(szTemp, 0, sizeof(TCHAR[MAX_PATH])); > + memset(wszTemp, 0, sizeof(wszTemp)/sizeof(WCHAR)); > dwSize = MAX_PATH; You are introducing at least two bugs here. Please don't change things that don't need changing. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: rpcrt4: Fix format specifier warning in NdrAllocate when size_t is typedef'd to unsigned long.
by Alexandre Julliard 27 Aug '08

27 Aug '08

"Rob Shearman" <robertshearman(a)gmail.com> writes: > @@ -373,7 +373,8 @@ void * WINAPI NdrAllocate(MIDL_STUB_MESSAGE *pStubMsg, size_t len) > /* check for overflow */ > if (adjusted_len < len) > { > - ERR("overflow of adjusted_len %d, len %d\n", adjusted_len, len); > + ERR("overflow of adjusted_len %lu, len %lu\n", (long)adjusted_len, > + (long)len); A better fix would be to avoid size_t completely, we should use SIZE_T instead. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [3/3] mshtml: Add IHTMLBodyElement test
by Jacek Caban 27 Aug '08

27 Aug '08

Hi Alistair, Alistair Leslie-Hughes wrote: > Hi, > > Changelog: > mshtml: Add IHTMLBodyElement test > These tests should go to script.c. Also take a look at elem_type_infos table. It would help you to avoid a mistake in previous patch. Jacek

1 0

advapi32:service failing tests
by James Hawkins 27 Aug '08

27 Aug '08

Hi Paul, Tests you added in the following commit cause failures in NT4, 2000, and Vista. Can you have a look please? commit 0539b463e65b386f4aa4c4235bf83d7f1d003e48 Author: Paul Vriens <paul.vriens.wine(a)gmail.com> Date: Thu Aug 21 16:14:45 2008 +0200 advapi32/tests: Add tests for EnumServicesStatusA. http://test.winehq.org/data/5f727cb2afbfafc3b396cf4902e55ca95dbc69d5/ Thanks, James Hawkins

2 1