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

appdb test results queue count wrong?
by Dan Kegel 23 Jan '08

23 Jan '08

When an app maintainer is logged in to the appdb, and there are test results queued for approval, the test results queue length is wrongly reported as zero at the moment...

3 3

Re: Alexandre Julliard : winex11.drv: Fix handling of strange window sizes in CreateWindow, with tests.
by Lei Zhang 23 Jan '08

23 Jan '08

On Jan 22, 2008 4:01 AM, Alexandre Julliard <julliard(a)winehq.org> wrote: > Module: wine > Branch: master > Commit: 390ae9cc2845fe002bc2e68c14d104d927af68b1 > URL: http://source.winehq.org/git/wine.git/?a=commit;h=390ae9cc2845fe002bc2e68c1… > > Author: Alexandre Julliard <julliard(a)winehq.org> > Date: Tue Jan 22 10:16:19 2008 +0100 > > winex11.drv: Fix handling of strange window sizes in CreateWindow, with tests. > > --- > > dlls/user32/tests/win.c | 79 +++++++++++++++++++++++++++++++++++++++++++++ > dlls/winex11.drv/window.c | 42 +++++------------------- > 2 files changed, 88 insertions(+), 33 deletions(-) > > diff --git a/dlls/user32/tests/win.c b/dlls/user32/tests/win.c > index 49fe766..d0a9673 100644 > --- a/dlls/user32/tests/win.c > +++ b/dlls/user32/tests/win.c > @@ -3733,6 +3733,43 @@ static LRESULT CALLBACK minmax_wnd_proc(HWND hwnd, UINT msg, WPARAM wp, LPARAM l > return 1; > } > > +static int expected_cx, expected_cy; > +static RECT expected_rect; > + > +static LRESULT CALLBACK winsizes_wnd_proc(HWND hwnd, UINT msg, WPARAM wp, LPARAM lp) > +{ > + switch(msg) > + { > + case WM_GETMINMAXINFO: > + { > + RECT rect; > + GetWindowRect( hwnd, &rect ); > + ok( !rect.left && !rect.top && !rect.right && !rect.bottom, > + "wrong rect %d,%d-%d,%d\n", rect.left, rect.top, rect.right, rect.bottom ); > + return DefWindowProc(hwnd, msg, wp, lp); > + } > + case WM_NCCREATE: > + case WM_CREATE: > + { > + CREATESTRUCTA *cs = (CREATESTRUCTA *)lp; > + RECT rect; > + GetWindowRect( hwnd, &rect ); > + trace( "hwnd %p msg %x size %dx%d rect %d,%d-%d,%d\n", > + hwnd, msg, cs->cx, cs->cy, rect.left, rect.top, rect.right, rect.bottom ); > + ok( cs->cx == expected_cx, "wrong x size %d/%d\n", cs->cx, expected_cx ); > + ok( cs->cy == expected_cy, "wrong y size %d/%d\n", cs->cy, expected_cy ); > + ok( rect.right - rect.left == expected_rect.right - expected_rect.left && > + rect.bottom - rect.top == expected_rect.bottom - expected_rect.top, > + "wrong rect %d,%d-%d,%d / %d,%d-%d,%d\n", > + rect.left, rect.top, rect.right, rect.bottom, > + expected_rect.left, expected_rect.top, expected_rect.right, expected_rect.bottom ); > + return DefWindowProc(hwnd, msg, wp, lp); > + } > + default: > + return DefWindowProc(hwnd, msg, wp, lp); > + } > +} > + > static void test_CreateWindow(void) > { > WNDCLASS cls; > @@ -3986,11 +4023,53 @@ static void test_CreateWindow(void) > ok(EqualRect(&rc, &rc_minmax), "rects don't match: (%d,%d-%d,%d) and (%d,%d-%d,%d)\n", > rc.left, rc.top, rc.right, rc.bottom, > rc_minmax.left, rc_minmax.top, rc_minmax.right, rc_minmax.bottom); > + DestroyWindow(hwnd); > + > + cls.lpfnWndProc = winsizes_wnd_proc; > + cls.lpszClassName = "Sizes_WndClass"; > + RegisterClass(&cls); > + > + expected_cx = expected_cy = 200000; > + SetRect( &expected_rect, 0, 0, 200000, 200000 ); > + hwnd = CreateWindowExA(0, "Sizes_WndClass", NULL, WS_CHILD, 300000, 300000, 200000, 200000, parent, 0, 0, NULL); > + ok( hwnd != 0, "creation failed err %u\n", GetLastError()); > + GetClientRect( hwnd, &rc ); > + ok( rc.right == 200000, "invalid rect right %u\n", rc.right ); > + ok( rc.bottom == 200000, "invalid rect bottom %u\n", rc.bottom ); > + DestroyWindow(hwnd); > + > + expected_cx = expected_cy = -10; > + SetRect( &expected_rect, 0, 0, 0, 0 ); > + hwnd = CreateWindowExA(0, "Sizes_WndClass", NULL, WS_CHILD, -20, -20, -10, -10, parent, 0, 0, NULL); > + ok( hwnd != 0, "creation failed err %u\n", GetLastError()); > + GetClientRect( hwnd, &rc ); > + ok( rc.right == 0, "invalid rect right %u\n", rc.right ); > + ok( rc.bottom == 0, "invalid rect bottom %u\n", rc.bottom ); > + DestroyWindow(hwnd); > > + expected_cx = expected_cy = -200000; > + SetRect( &expected_rect, 0, 0, 0, 0 ); > + hwnd = CreateWindowExA(0, "Sizes_WndClass", NULL, WS_CHILD, -300000, -300000, -200000, -200000, parent, 0, 0, NULL); > + ok( hwnd != 0, "creation failed err %u\n", GetLastError()); > + GetClientRect( hwnd, &rc ); > + ok( rc.right == 0, "invalid rect right %u\n", rc.right ); > + ok( rc.bottom == 0, "invalid rect bottom %u\n", rc.bottom ); > DestroyWindow(hwnd); > + > + /* top level window */ > + expected_cx = expected_cy = 200000; > + SetRect( &expected_rect, 0, 0, GetSystemMetrics(SM_CXMAXTRACK), GetSystemMetrics(SM_CYMAXTRACK) ); > + hwnd = CreateWindowExA(0, "Sizes_WndClass", NULL, WS_OVERLAPPEDWINDOW, 300000, 300000, 200000, 200000, 0, 0, 0, NULL); > + ok( hwnd != 0, "creation failed err %u\n", GetLastError()); > + GetClientRect( hwnd, &rc ); > + ok( rc.right <= expected_cx, "invalid rect right %u\n", rc.right ); > + ok( rc.bottom <= expected_cy, "invalid rect bottom %u\n", rc.bottom ); > + DestroyWindow(hwnd); > + > DestroyWindow(parent); > > UnregisterClass("MinMax_WndClass", GetModuleHandle(0)); > + UnregisterClass("Sizes_WndClass", GetModuleHandle(0)); > > #undef expect_menu > #undef expect_style > diff --git a/dlls/winex11.drv/window.c b/dlls/winex11.drv/window.c > index 84209b7..3f65602 100644 > --- a/dlls/winex11.drv/window.c > +++ b/dlls/winex11.drv/window.c > @@ -1296,37 +1296,13 @@ BOOL X11DRV_CreateWindow( HWND hwnd, CREATESTRUCTA *cs, BOOL unicode ) > CBT_CREATEWNDA cbtc; > CREATESTRUCTA cbcs; > BOOL ret = FALSE; > + INT cx = cs->cx, cy = cs->cy; > > if (!(data = alloc_win_data( display, hwnd ))) return FALSE; > > - if (cs->cx > 65535) > - { > - ERR( "invalid window width %d\n", cs->cx ); > - cs->cx = 65535; > - } > - if (cs->cy > 65535) > - { > - ERR( "invalid window height %d\n", cs->cy ); > - cs->cy = 65535; > - } > - if (cs->cx < 0) > - { > - ERR( "invalid window width %d\n", cs->cx ); > - cs->cx = 0; > - } > - if (cs->cy < 0) > - { > - ERR( "invalid window height %d\n", cs->cy ); > - cs->cy = 0; > - } > - > if (hwnd == GetDesktopWindow()) get_desktop_xwin( display, data ); > else > { > - /* initialize the dimensions before sending WM_GETMINMAXINFO */ > - SetRect( &rect, cs->x, cs->y, cs->x + cs->cx, cs->y + cs->cy ); > - X11DRV_SetWindowPos( hwnd, 0, &rect, &rect, SWP_NOZORDER | SWP_NOACTIVATE, NULL ); > - > /* create an X window if it's a top level window */ > if (GetAncestor( hwnd, GA_PARENT ) == GetDesktopWindow()) > { > @@ -1360,16 +1336,16 @@ BOOL X11DRV_CreateWindow( HWND hwnd, CREATESTRUCTA *cs, BOOL unicode ) > POINT maxSize, maxPos, minTrack, maxTrack; > > WINPOS_GetMinMaxInfo( hwnd, &maxSize, &maxPos, &minTrack, &maxTrack); > - if (maxTrack.x < cs->cx) cs->cx = maxTrack.x; > - if (maxTrack.y < cs->cy) cs->cy = maxTrack.y; > - if (cs->cx < 0) cs->cx = 0; > - if (cs->cy < 0) cs->cy = 0; > - > - SetRect( &rect, cs->x, cs->y, cs->x + cs->cx, cs->y + cs->cy ); > - if (!X11DRV_SetWindowPos( hwnd, 0, &rect, &rect, SWP_NOZORDER | SWP_NOACTIVATE, NULL )) > - return FALSE; > + if (maxTrack.x < cx) cx = maxTrack.x; > + if (maxTrack.y < cy) cy = maxTrack.y; > } > > + if (cx < 0) cx = 0; > + if (cy < 0) cy = 0; > + SetRect( &rect, cs->x, cs->y, cs->x + cx, cs->y + cy ); > + if (!X11DRV_SetWindowPos( hwnd, 0, &rect, &rect, SWP_NOZORDER | SWP_NOACTIVATE, NULL )) > + return FALSE; > + > /* send WM_NCCREATE */ > TRACE( "hwnd %p cs %d,%d %dx%d\n", hwnd, cs->x, cs->y, cs->cx, cs->cy ); > if (unicode) > > > > I've having trouble on my dual screen (twinview) setup with this patch. With a fresh ~/.wine directory, I ran notepad or regedit. The app starts on the left monitor. When I move the window to the right monitor, the window is no longer being redrawn.

3 2

Re: gdi32: Fix spec file to export the right symbols for GetGlyphOutline
by Dmitry Timoshkov 23 Jan '08

23 Jan '08

"Kai Blin" <kai.blin(a)gmail.com> wrote: > --- a/dlls/gdi32/gdi32.spec > +++ b/dlls/gdi32/gdi32.spec > @@ -282,7 +282,6 @@ > @ stdcall GetFontUnicodeRanges(ptr ptr) > @ stdcall GetGlyphIndicesA(long ptr long ptr long) > @ stdcall GetGlyphIndicesW(long ptr long ptr long) > -@ stub GetGlyphOutline > @ stdcall GetGlyphOutlineA(long long long ptr long ptr ptr) > @ stdcall GetGlyphOutlineW(long long long ptr long ptr ptr) > @ stub GetGlyphOutlineWow This patch is not correct. gdi32 does export GetGlyphOutline at least in XP. It just points to GetGlyphOutlineA. -- Dmitry.

2 1

Problem with winelib
by Nagarjuna Bandi 23 Jan '08

23 Jan '08

Dear Wine Developers Team, I am using the wine for porting windows applications under linux. This is a very powerful tool for doing this job thank you very much for providing us a such tool. Here i am getting problem with winelib when i trying to build the notepad application with the given guidelines in winelib user guide i had made the changes to Makefile you are specifying but i got another error msg like [nagarjunab(a)ibi212 notepad2]$ vim Makefile [nagarjunab(a)ibi212 notepad2]$ make wrc -I. -forsrc.res rsrc.rc winegcc -mwindows -mno-cygwin -o notepad2.exe.so dialog.o main.o rsrc.res -lodbc32 -lole32 -loleaut32 -lwinspool -luuid dialog.o(.text+0x1a5): In function `DIALOG_StringMsgBox': : undefined reference to `wnsprintfW' collect2: ld returned 1 exit status winegcc: gcc failed. make: *** [notepad2.exe.so] Error 2 This is the error i am getting so please let me know what is the problem and what i have to do for avoiding this problem. I am expecting a fast response from you. Thanks in advance. --------------------------------- Regards, Nagarjuna Bandi , Infobrain Technologies Pvt. Ltd. | My Home Sarovar Plaza | Hyderabad | India. --------------------------------- Explore your hobbies and interests. Click here to begin.

4 4

Re: [dbghelp 4/10] Rewrite much of the symbol lookup method to work with Optimized PDB files as well
by Eric Pouech 22 Jan '08

22 Jan '08

Jason Green a écrit : > ------------------------------------------------------------------------ > > From 14d91ebd5974c8fc02f8b83d53e8eff0df7ad76d Mon Sep 17 00:00:00 2001 > From: Jason Green <jason(a)jave01.(none)> > Date: Thu, 17 Jan 2008 16:41:11 -0500 > Subject: [PATCH] Rewrite much of the symbol lookup method to work with Optimized PDB files as well. > Also, improves a number of TRACEs, and wraps strings in the debugstr() functions. > what's the rationale of optimized pdb ? the patch is already big enough not to mix code evolution with stuffing new trace:s all over the place A+ -- Eric Pouech "The problem with designing something completely foolproof is to underestimate the ingenuity of a complete idiot." (Douglas Adams)

2 3

Suggestion to the list maintainer
by kubrick＠fgv6.net 22 Jan '08

22 Jan '08

Hello. I have many subscriptions to different mailing list with my email-address and I'd appreciate a lot to be able to identify a message from this mailing list quickly by reading it's subject. Could you please add a subject prefix like [WINE-DEVEL] in the mailman preferences ? It's very easy to do, so maybe there is a good reason why you haven't done it yet... Thanks. François.

8 15

Re: [WineHQ][WWN] WWN 340
by Dmitry Timoshkov 22 Jan '08

22 Jan '08

"Zachary Goldberg" <zgs(a)seas.upenn.edu> wrote: > +<section > + title="Theming WINE" > + subject="WINE" Could you please use a commonly accepted capitalization of the project name - Wine? -- Dmitry.

1 0

Re: WineD3D: don't use texture_rectangle for paletted textures
by Stefan Dösinger 22 Jan '08

22 Jan '08

Am Montag, 21. Januar 2008 22:33:17 schrieb Roderick Colenbrander: > Hi, > > Don't use ARB_texture_rectangle in combination with EXT_paletted_texture. > This should get StarCraft running again on Geforce1-FX cards in case of > DirectDrawRenderer opengl with RenderTargetLockMode readtex. (bug 10697) You'll have to modify IWineD3DDeviceImpl_CreateTexture as well

1 0

Re: Fix detection error in configure.ac
by Alexandre Julliard 22 Jan '08

22 Jan '08

Thomas Zimmermann <kuhundbaer(a)web.de> writes: > this patch fixes a detection error in configure.ac. > > 'Xrandr' and 'Xinerama' detection fails if GCC is set to Ansi mode. > The reason is the use of 'typeof' in the test. GCC does not support > this keyword when it is in Ansi mode. The patch solves that problem by > using __typeof__ instead. I tested the patch on my machine. Since the corresponding code won't build either, it's actually a feature that the configure check fails. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: gdi32: Make GetICMProfile behave more like native. Rewrite ansi version as a wrapper and move color management functions to their own file.
by Detlef Riekenberg 22 Jan '08

22 Jan '08

On Mo, 2008-01-21 at 16:33 +0100, Hans Leidekker wrote: - gdi32: Make GetICMProfile behave more like native. Rewrite ansi version as a wrapper and move color management functions to their own file. The large subject is an indicator, that the Patch can be splitted. IMHO, moving the stubs to a different file can be a seperate patch. > +BOOL WINAPI GetICMProfileA(HDC hdc, LPDWORD size, LPSTR filename) > + if (!hdc || !size || !filename) return FALSE; This is not the same as documented in MSDN, and > +BOOL WINAPI GetICMProfileW(HDC hdc, LPDWORD size, LPWSTR filename) > + > + if (!hdc || !size) return FALSE; ANSI and UNICODE are different. Are you really sure? (Please insert a big warning in the comment then) Returning FALSE without a SetLastError() looks not correct, compared to SetLastError(ERROR_INSUFFICIENT_BUFFER) below. Sorry, i'm unable to test your code now. I try to do it tomorrow. -- By by ... Detlef

3 2