wine-devel

wine-devel@winehq.org

10 participants
84459 discussions

Re: prevent dereferencing null
by Dimitrie O. Paun 14 Jan '03

14 Jan '03

On January 9, 2003 07:08 pm, Bill Medland wrote: > + if (!tpinfo->chanbuf) { > + ERR("tpinfo has no Rpc Channel Buffer\n"); > + return 0; > + } Is this expected behaviour? If so, there's no need for the ERR msg. If not, there's no need for the test, we need to fix the root cause... -- Dimi.

3 4

patch for "GetFileTitleW: illegal call to GetFileTitleA" 2ed
by liu spider 14 Jan '03

14 Jan '03

Thanks to Dmitry Timoshkov. This patch is the second edition, adopted his suggestions. ChangeLog: -dlls/commdlg/filetitle.c implemented GetFileTitleW GetFileTitleA now is a call to GetFileTitleW (The patch seems to be ignored without any comments. Could anyone give some clues on how to improve this patch? thanks.) liuspider __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com Index: dlls/commdlg/filetitle.c =================================================================== RCS file: /home/wine/wine/dlls/commdlg/filetitle.c,v retrieving revision 1.12 diff -u -r1.12 filetitle.c --- dlls/commdlg/filetitle.c 31 May 2002 23:25:45 -0000 1.12 +++ dlls/commdlg/filetitle.c 12 Jan 2003 07:09:56 -0000 @@ -3,6 +3,7 @@ * * Copyright 1994 Martin Ayotte * Copyright 1996 Albrecht Kleine + * Copyright 2003 liuspider * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public @@ -24,10 +25,10 @@ #include "winbase.h" #include "winnls.h" #include "commdlg.h" +#include "wine/unicode.h" +#include "winternl.h" #include "wine/debug.h" -#include "heap.h" /* Has to go */ - WINE_DEFAULT_DEBUG_CHANNEL(commdlg); #include "cdlg.h" @@ -38,19 +39,39 @@ */ short WINAPI GetFileTitleA(LPCSTR lpFile, LPSTR lpTitle, UINT cbBuf) { - int i, len; + int ret; + UNICODE_STRING strWFile, strWTitle; + RtlCreateUnicodeStringFromAsciiz(&strWFile, lpFile); + RtlCreateUnicodeStringFromAsciiz(&strWTitle,lpTitle); + ret = GetFileTitleW(strWFile.Buffer, strWTitle.Buffer, cbBuf * sizeof(WCHAR)); + if (!ret){ + WideCharToMultiByte( CP_ACP, 0, strWTitle.Buffer, -1, lpTitle, cbBuf, NULL, NULL ); + } + RtlFreeUnicodeString( &strWFile ); + RtlFreeUnicodeString( &strWTitle ); + return ret; +} + +/*********************************************************************** + * GetFileTitleW (COMDLG32.@) + * + */ +short WINAPI GetFileTitleW(LPCWSTR lpFile, LPWSTR lpTitle, UINT cbBuf) +{ + int i, len; + WCHAR brkpoint[] = {'*','[',']'}; TRACE("(%p %p %d); \n", lpFile, lpTitle, cbBuf); if(lpFile == NULL || lpTitle == NULL) return -1; - len = strlen(lpFile); + len = strlenW(lpFile); if (len == 0) return -1; - if(strpbrk(lpFile, "*[]")) + if(strpbrkW(lpFile, brkpoint)) return -1; len--; @@ -70,34 +91,14 @@ if(i == -1) i++; - TRACE("---> '%s' \n", &lpFile[i]); + TRACE("---> '%s' \n", debugstr_w(&lpFile[i])); - len = strlen(lpFile+i)+1; + len = strlenW(lpFile+i)+1; if(cbBuf < len) return len; - strncpy(lpTitle, &lpFile[i], len); + strncpyW(lpTitle, &lpFile[i], len); return 0; -} - - -/*********************************************************************** - * GetFileTitleW (COMDLG32.@) - * - */ -short WINAPI GetFileTitleW(LPCWSTR lpFile, LPWSTR lpTitle, UINT cbBuf) -{ - LPSTR file = HEAP_strdupWtoA(GetProcessHeap(), 0, lpFile); /* Has to go */ - LPSTR title = HeapAlloc(GetProcessHeap(), 0, cbBuf); - short ret; - - ret = GetFileTitleA(file, title, cbBuf); - - if (cbBuf > 0 && !MultiByteToWideChar( CP_ACP, 0, title, -1, lpTitle, cbBuf )) - lpTitle[cbBuf-1] = 0; - HeapFree(GetProcessHeap(), 0, file); - HeapFree(GetProcessHeap(), 0, title); - return ret; }

1 0

Re: tests for GetPrinterDriverDirectoryA
by Francois Gouget 14 Jan '03

14 Jan '03

On Mon, 13 Jan 2003, Stefan Leichter wrote: > Am Montag 13 Januar 2003 18:42 schrieben Sie: > > On Sun, 12 Jan 2003, Stefan Leichter wrote: > > > Am Sonntag 12 Januar 2003 02:43 schrieben Sie: [...] > > > > However it fails on Wine: > > > > info.c:25: Test failed: pcbNeeded set to 18 instead of 19 > > > > info.c:30: Test failed: pcbNeeded set to 18 instead of 19 > > > > info.c:34: Test failed: expected result == 0, got 18 > > > > info.c:35: Test failed: pcbNeeded set to 18 instead of 19 > > > > info.c:48: Test failed: expected result != 0, got 0 > > > > info.c:71: Test failed: expected result != 0, got 0 > > after this patch wine will fix the errors listed above Yep, that works. -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ Cahn's Axiom: When all else fails, read the instructions.

1 0

Re: Help with /?? stuff
by Alberto Massari 13 Jan '03

13 Jan '03

At 20.24 10/01/2003 +0200, Shachar Shemesh wrote: >Alberto Massari wrote: > >[...] >> >>BTW, do you need any help on this task? > >"Need" is a strong term. I can complete the task on my own, but your help >will be most welcome (in other words, if there is another aread of Wine >that needs your attention and noone seems to be working on, that would be >a better use of everyone's time). If you want to help with wineboot, I >think the best place to start is with the "win.ini" processing, and then >to port Andy's function to run whole "run" keys to the new prog. > >If you do decide to help with wineboot, please email me in private when >starting to work on each activity, to prevent "competition" on writing >stuff in. Unfortunately I'm currently fully booked on trying to fix the bugs that prevent ole32 and shell32 from working with IE6; but installing IE6 needs a working wineboot, so, whenever I succeed in the first task, I will start working on wineboot. Alberto

3 11

what does __leave do?
by Greg Turner 13 Jan '03

13 Jan '03

Could someone run the following code under a microsoft compiler for me, and let me know the result? main() { int x = 0; __try { x = 1; __leave; } __finally { x = 2; } printf("x was %d\n", x); } thanks, -- gmt "If everyone is thinking alike then somebody isn't thinking." --George S. Patton

2 2

Sysparams registry tests
by Francois Gouget 13 Jan '03

13 Jan '03

Here are the failing user32/sysparams registry tests on Win95, Win98 and NT4. sysparams.c:676:testing SPI_{GET,SET}MENUDROPALIGNMENT sysparams.c:161: Test failed: Wrong value in registry: subKey=Software\Microsoft\Windows NT\CurrentVersion\Windows, valName=MenuDropAlignment, testValue=1, value= sysparams.c:161: Test failed: Wrong value in registry: subKey=Software\Microsoft\Windows NT\CurrentVersion\Windows, valName=MenuDropAlignment, testValue=0, value= -> on Win9x this is in Control Panel\desktop sysparams.c:711:testing SPI_{GET,SET}DOUBLECLKWIDTH sysparams.c:161: Test failed: Wrong value in registry: subKey=Control Panel\Mouse, valName=DoubleClickWidth, testValue=0, value= sysparams.c:161: Test failed: Wrong value in registry: subKey=Control Panel\Mouse, valName=DoubleClickWidth, testValue=10000, value= -> on Win9x this is in Control Panel\desktop sysparams.c:741:testing SPI_{GET,SET}DOUBLECLKHEIGHT sysparams.c:161: Test failed: Wrong value in registry: subKey=Control Panel\Mouse, valName=DoubleClickHeight, testValue=0, value= sysparams.c:161: Test failed: Wrong value in registry: subKey=Control Panel\Mouse, valName=DoubleClickHeight, testValue=10000, value= -> on Win9x this is in Control Panel\desktop sysparams.c:644:testing SPI_{GET,SET}ICONTITLEWRAP sysparams.c:161: Test failed: Wrong value in registry: subKey=Control Panel\Desktop\WindowMetrics, valName=IconTitleWrap, testValue=1, value= sysparams.c:161: Test failed: Wrong value in registry: subKey=Control Panel\Desktop\WindowMetrics, valName=IconTitleWrap, testValue=0, value= -> on NT4 this is in Control Panel\Desktop Notice how this is not even a clean separation between NT OSes and Win9x ones: Win9x aligns with XP for SPI_SETICONTITLEWRAP but not NT4, while for the other settings it's the opposite. Anyway, is it worth keeping these tests? That is making sure that SPI_xxx stores the setting in the registry key we expect? Do we know of applications that depend on these specific settings? If not then I will just submit a patch to remove these checks. Or maybe Wine should store these settings in all known locations? Note that I already removed these tests for SPI_ICONHORIZONTALSPACING and SPI_ICONVERTICALSPACING because for these Windows does not even always store the same value: sometimes it's the value we provide but most of the time that value seems to be multiplied by -15 (resulting in the negative number!). And the -15 changes to anothe value if you change the icon size. -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ tcA thgirypoC muinelliM latigiD eht detaloiv tsuj evah uoY

1 0

Re: user32.SendMessageCallback
by Alexandre Julliard 13 Jan '03

13 Jan '03

"Robert Shearman" <R.J.Shearman(a)warwick.ac.uk> writes: > ChangeLog: > - Fix SendMessageCallback so that it works in all cases now You can't do it that way, the callback has to be called in the context of the thread that sent the message. -- Alexandre Julliard julliard(a)winehq.com

2 1

Re: tests for GetPrinterDriverDirectoryA
by Francois Gouget 13 Jan '03

13 Jan '03

On Tue, 7 Jan 2003, Stefan Leichter wrote: > Hello, > > here are some tests for GetPrinterDriverDirectoryA. They should pass on NT4 > and hopefully win95 too. Thanks to Francois Gouget for testing. I see this test has not been committed yet... I tested it and it works on Win95, Win98, NT4 and XP. It's a much better score than some other tests... However it fails on Wine: info.c:25: Test failed: pcbNeeded set to 18 instead of 19 info.c:30: Test failed: pcbNeeded set to 18 instead of 19 info.c:34: Test failed: expected result == 0, got 18 info.c:35: Test failed: pcbNeeded set to 18 instead of 19 info.c:48: Test failed: expected result != 0, got 0 info.c:71: Test failed: expected result != 0, got 0 The easy solution is to just wrap the relevant tests in a 'todo_wine'. The better one is to fix Wine... Or if the fix is too long to do, wrap with todo_wine now so the test can be committed now, and then fix Wine and remove the relevant todo_wine statements after each fix. -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ f u kn rd ts, ur wy 2 gky 4 ur wn gd.

2 2

Re: Re: The version info patches + d3d question
by fenix＠club-internet.fr 13 Jan '03

13 Jan '03

----Message d'origine---- >Date: Mon, 13 Jan 2003 16:04:10 +0100 >De: Lionel Ulmer <lionel.ulmer(a)free.fr> >A: fenix(a)club-internet.fr >Copie à: farmboy1(a)subdimension.com, wine-devel(a)winehq.com >Sujet: Re: The version info patches + d3d question > >> >the problem is they dont run even with the patches because they cant get >> >version info from d3dim.dll and probably other d3d dlls. >> >since wine doesnt have these dlls how can wine give these programs the needed >> >info or is it planned to split ddraw in a ddraw and a d3d component? >> >> actually, we are speaking about d3d architecture, but it will be possible to implement d3dim.dll (minors d3d calls) into near future using the "new" codebase. >> And d3d component (for version 3..7) is contained into ddraw.dll > >I don't even know WHAT d3dim.dll is :-) What does it contain for API ? d3dim.dll - D3DFree - D3DMalloc - D3DRealloc - Direct3DCreateDevice - Direct3DCreateTexture - Direct3DGetSWRastZPixFmts - Direct3DCreate - Direct3D_HALCleanUp - FlushD3DDevices2 - FlushD3DDevices - PaletteAssociateNotify - PaletteUpdateNotify - SurfaceFlipNotify d3dim700.dll - D3DFree - D3DMalloc - D3DRealloc - Direct3DCreateDevice - CreateTexture - D3DBreakVBLock - D3DTextureUpdate - DestroyTexture - Direct3DCreate - Direct3D_HALCleanUp - FlushD3DDevices - GetLOD - GetPriority - PaletteAssociateNotify - PaletteUpdateNotify - SetLOD - SetPriority - SurfaceFlipNotify >I know about D3DRM.DLL (that won't be supported) for the 'retained mode' API >but I thought that all 'immediate mode' API was in ddraw.dll and not in a >separate library. yes, you are right. d3dim is only some utilities functions and interfaces for d3d immediate mode. (with d3d8 most of the code of d3dim and d3dxof was moved into d3dx8) >Why not first test with an empty DLL with just the version ressources in it >and see if it's enough ? maybe with only emtpy stubs now :) > Lionel Raphael

1 0

Re: The version info patches + d3d question
by fenix＠club-internet.fr 13 Jan '03

13 Jan '03

----Message d'origine---- >De: Enrico Horn <farmboy1(a)subdimension.com> >A: wine-devel(a)winehq.com >Sujet: The version info patches + d3d question >Date: Mon, 13 Jan 2003 14:47:31 +0100 > >Hi, Hi, >The patches add version info to directx libraries. >some games(for instance lucasarts games) wont run without finding a valid >version of the directx files. thanks, maybe you can do the same for d3d8 :) >the problem is they dont run even with the patches because they cant get >version info from d3dim.dll and probably other d3d dlls. >since wine doesnt have these dlls how can wine give these programs the needed >info or is it planned to split ddraw in a ddraw and a d3d component? actually, we are speaking about d3d architecture, but it will be possible to implement d3dim.dll (minors d3d calls) into near future using the "new" codebase. And d3d component (for version 3..7) is contained into ddraw.dll >Have fun >Enrico Regards, Raphael

2 1

The version info patches + d3d question
by Enrico Horn 13 Jan '03

13 Jan '03

Hi, The patches add version info to directx libraries. some games(for instance lucasarts games) wont run without finding a valid version of the directx files. the problem is they dont run even with the patches because they cant get version info from d3dim.dll and probably other d3d dlls. since wine doesnt have these dlls how can wine give these programs the needed info or is it planned to split ddraw in a ddraw and a d3d component? Have fun Enrico

1 0

Direct3D/OpenGL compilation error
by Francois Gouget 13 Jan '03

13 Jan '03

One of the machines I compile on from time to time is an old RedHat 6.2 which is nice because it still has glibc 2.1.3 so the binaries it produces work on all Linux distributions. Anyway, the latest Wine does not compile anymore on that machine. I get: gcc -c -I. -I. -I../../include -I../../include -I/usr/X11R6/include -g -O2 -Wall -gstabs+ -fPIC -D__WINESRC__ -D_REENTRANT -o device.o device.c device.c: In function `SOURCEx_RGB_EXT': device.c:859: `GL_SOURCE2_RGB_EXT' undeclared (first use in this function) device.c:859: (Each undeclared identifier is reported only once device.c:859: for each function it appears in.) device.c:860: `GL_SOURCE0_RGB_EXT' undeclared (first use in this function) device.c:861: `GL_SOURCE1_RGB_EXT' undeclared (first use in this function) device.c:869: warning: control reaches end of non-void function device.c: In function `OPERANDx_RGB_EXT': device.c:872: `GL_OPERAND2_RGB_EXT' undeclared (first use in this function) device.c:873: `GL_OPERAND0_RGB_EXT' undeclared (first use in this function) device.c:874: `GL_OPERAND1_RGB_EXT' undeclared (first use in this function) device.c:882: warning: control reaches end of non-void function device.c: In function `SOURCEx_ALPHA_EXT': device.c:885: `GL_SOURCE2_ALPHA_EXT' undeclared (first use in this function) device.c:886: `GL_SOURCE0_ALPHA_EXT' undeclared (first use in this function) device.c:887: `GL_SOURCE1_ALPHA_EXT' undeclared (first use in this function) device.c:895: warning: control reaches end of non-void function device.c: In function `OPERANDx_ALPHA_EXT': device.c:898: `GL_OPERAND2_ALPHA_EXT' undeclared (first use in this function) device.c:899: `GL_OPERAND0_ALPHA_EXT' undeclared (first use in this function) device.c:900: `GL_OPERAND1_ALPHA_EXT' undeclared (first use in this function) device.c:908: warning: control reaches end of non-void function device.c: In function `IDirect3DDevice8Impl_SetTextureStageState': device.c:3312: `GL_PREVIOUS_EXT' undeclared (first use in this function) device.c:3314: `GL_PRIMARY_COLOR_EXT' undeclared (first use in this function) device.c:3318: `GL_CONSTANT_EXT' undeclared (first use in this function) device.c:3350: `GL_COMBINE_ALPHA_EXT' undeclared (first use in this function) device.c:3350: `GL_COMBINE_RGB_EXT' undeclared (first use in this function) device.c:3409: `GL_RGB_SCALE_EXT' undeclared (first use in this function) device.c:3423: `GL_ADD_SIGNED_EXT' undeclared (first use in this function) So is it just that the OpenGL headers on that machine are hopelessly out of date (mesa 3.3) or is there some configure test that can/should be done? Note that if I compile with --without-opengl Wine compiles ok so it's not so bad. -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ f u kn rd ts, ur wy 2 gky 4 ur wn gd.

1 0

Developers current e-mail addresses listed in authors.ent
by Tom Wickline 13 Jan '03

13 Jan '03

I am getting ready to update the WINE Packagers Guide See : http://www.winehq.com/Docs/wine-pkg/ And the first thing that I noticed was Andreas & Marcus e-mail address were incorrect. So I then looked at : http://cvs.winehq.com/cvsweb/~checkout~/wine/documentation/authors.ent?rev=… And found that alot of other's were different from my Address Book. I have listed them below Current is what is now there current address in the *authors.ent* file. And below is the addresses that is listed in my address book. There are also a couple DOA addresses listed in this file and there at the bottom. I dont want to change this file and not have people's e-mail correct! So can everyone look over the file and let me know if your e-mail needs to be changed ?? Also a couple people below have multiable addresses can you let me know what address you want listed. Tom Current = ovek(a)winehq.com Ove Kaaven <ovek(a)arcticnet.no Ove Kaaven <ovehk(a)ping.uio.no ovek <ovek(a)transgaming.com --- Current = Marcus.Meissner(a)caldera.de Marcus Meissner <marcus(a)jet.franken.de Marcus Meissner <mm(a)lst.de ---- Current = amohr(a)codeweavers.com Andreas Mohr <andi(a)rhlx01.fht-esslingen.de --- Current = Speeddymon(a)yahoo.com Dustin Navea <Speeddymon(a)sbcglobal.net --- Current = dimi(a)cs.toronto.edu "Dimitrie O. Paun" <dpaun(a)rogers.com --- Current = juergen.schmied(a)metronet.de juergen schmied <juergen.schmied(a)debitel.net DOA addresses Should I leave them or remove them ?? douglas-ridgway "ridgway(a)winehq.com"> john-sheets "jsheets(a)codeweavers.com">

2 2

Re: Bug Fix With Delay load function in Shell32 (version2)
by Dimitrie O. Paun 12 Jan '03

12 Jan '03

On January 12, 2003 12:08 pm, Robert Shearman wrote: > Index: shellole.c Please make your patches relative to wine's root. > -HINSTANCE hShellOle32 = 0; > +HINSTANCE g_hShellOle32 = 0; /* global handle to ole32.dll */ Also, please do not introduce ugly naming conventions like the above, especially when the style the DLL is written in does not call for it. In general, we have none of these conventions in Wine, and it's preferable to just avoid them. We have a bunch of others though. :) -- Dimi.

1 0

Re: FreeBSD Compilation Fix
by Dimitrie O. Paun 12 Jan '03

12 Jan '03

On January 11, 2003 09:00 pm, Jared Goldberg wrote: > Index: dlls/dsound/tests/dsound.c > =================================================================== > RCS file: /home/wine/wine/dlls/dsound/tests/dsound.c,v > retrieving revision 1.3 > diff -r1.3 dsound.c > 20a21,23 > > >#if defined(__FreeBSD__) > >#include <stdlib.h> > >#else > > 21a25 > > >#endif Uniffied diffs please: cvs diff -u or diff -u See: http://www.winehq.org/development/#patches In this case, I suggest you just include stdlib.h without any ifdefs. It's a standard header, it should exist on all systems. -- Dimi.

1 0

Re: Shell Ordinals
by j.schmied＠t-online.de 12 Jan '03

12 Jan '03

Hi! We have a problem here... win9x and winnt have a very different set of ordinal functions exported. We have to decide what we support. At the moment most ordinals functions are win9x style. We should never mix both cases since then both explorer won't run. I think we don't have any chance to get the win9x explorer running. A long time ago I did a lot of explorer debugging and the nt4 explorer was nearly running since it's not as much involved into the booting of the system as the win9x one. So I would suggest to change the ordinals into the nt ones. It would cause some 9x only functions to be removed. The change of shell32.spec should be done in one go. PS: Since I recently started to do some shell32 work again: Who is doing what at the moment? Bye Juergen > > This includes a few ordinals used by explorer and desk.cpl > There seems to be a conflict between SHLocalAlloc/Free and the ordinals I > have made stubs for (on Win2k SP3 shell32.dll), but I have no way of knowing > which ordinal (if any) SHLocalAlloc/Free should be. > > ChangeLog: > - Implement some PropertySheetExtArray ordinals > - Add some stubs used by explorer > > Rob > -- juergen.schmied(a)debitel.net FAX über T-Online Fax&Fon: 01805 060 334 316 35 (12Cent/Min)

2 1

Re: conformance test for ShellExecute
by Francois Gouget 12 Jan '03

12 Jan '03

On Sun, 29 Dec 2002, Dan Kegel wrote: > Well, the puny little beginnings of one, anyway. > (I'll improve this later. Still, it already > catches a difference between Windows and Wine, > so it's worth committing as is.) I tested it on Win95, Win98, NT4 and XP and it worked on all platforms. However, since it currently fails in Wine should should enclose the test in a 'todo_wine' and resubmit it. -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ Computers are like airconditioners They stop working properly if you open WINDOWS

1 0

patch for "GetFileTitleW: illegal call to GetFileTitleA" 2ed
by liu spider 12 Jan '03

12 Jan '03

Thanks to Dmitry Timoshkov. This patch is the second edition, adopted his suggestions. ChangeLog: -dlls/commdlg/filetitle.c implemented GetFileTitleW make GetFileTitleA a wrapper of GetFileTitleW liuspider __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com

1 0

Re: wine/programs/wineboot wineboot.c Makefile.in
by Shachar Shemesh 11 Jan '03

11 Jan '03

Alexandre Julliard wrote: >ChangeSet ID: 6997 >CVSROOT: /opt/cvs-commit >Module name: wine >Changes by: julliard(a)wine.codeweavers.com 2003/01/11 15:02:50 > >Modified files: > programs/wineboot: wineboot.c Makefile.in > >Log message: > Shachar Shemesh <wine-patches(a)sun.consumer.org.il> > Implemented the "pending rename" registry processing. > >Patch: http://cvs.winehq.com/patch.py?id=6997 > >Old revision New revision Changes Path > 1.2 1.3 +158 -3 wine/programs/wineboot/wineboot.c > 1.1 1.2 +1 -1 wine/programs/wineboot/Makefile.in > > > > Hi Alexander, Two points about the differences between the version I submitted and the version that made it in. 1. If you are going to reject the printing of "Windows is finishing to install your programs..." closing message, at least also remove the starting message (lines 105 and 106 of version 1.3 of the file). 2. The reason I added the cast on the free line was to remove a compilation warning. If you made sure the warning doesn't happen by changing somewhere else in the code, that's, of course, fine. Otherwise there is a warning when compiling (I am currently doing a clean compile to make sure whether that happens on my system). Shachar

3 5

Re: MciWave: some small cleanups
by Francois Gouget 11 Jan '03

11 Jan '03

On Sat, 11 Jan 2003, Eric Pouech wrote: > see changelog for the details [...] > Name: mciwave > ChangeLog: > License: X11 Actually the changelog is a bit short... -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ Linux: Because rebooting is for adding new hardware

2 1

Re: [D3D8 - 1] resend of Jason Texture work + many fixes
by Raphaël Junqueira 11 Jan '03

11 Jan '03

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Saturday 11 January 2003 01:27, Raphaël Junqueira wrote: > Hi again, I forgot to say with this patch now: - - i have run successfully the NVIDIA Effects Browser 3.0 (http://developer.nvidia.com/view.asp?IO=NVEffectsBrowser_3) with many effects as Aniso (http://developer.nvidia.com/view.asp?IO=Anisotropic_Lighting) or membrane_lighting (http://developer.nvidia.com/view.asp?IO=membrane_lighting) in pure vertex shader software emulation - - Warcraf3 Demo begin to run - - SplinterCell Demo now complain about stencil support anyone have other Dx8 games demos for testing (simple if possible) ? > Thanks > Raphael -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) iD8DBQE+H2hKp7NA3AmQTU4RAvalAJ9v746sFf+RiEGzC/Vuv1NBnprAsQCaAwdQ WiIc/yV8HbZe0iw3CwJoFuE= =He8Q -----END PGP SIGNATURE-----

2 2

C++ and winedbg: testers wanted
by Eric Pouech 11 Jan '03

11 Jan '03

as some folks have already noticed, loading ELF modules with C++ compilation units gave a zillion of traces while parsing debug info in winedbg this patch should silence this type of output it won't however add the C++ ABI to the debugger: heavy C++ debugging will require gdb (and also DwarfII debug info; stabs is so broken in this respect) on the C++ front, what should be possible is: - every class should be seen as a regular structures - non virtual methods will be available as functions (of course, this will be seen as a regular parameter, and functions will be seen as undecorated names... this may recall to some of us the old good days of cfront ;-) anyway, as gcc has some discrepencies it its stabs output in its version, I'd like to get some feedback of people loading C++ info into winedbg I already tested gcc 3.2 and 2.96, but results from 2.95, 3.0 and 3.3 would be appreciated TIA A+ -- Eric Pouech Name: wd_stabs ChangeLog: - added experimental parsing for C++ code (but winedbg doesn't support C++ ABI) - now loading symbols for included files License: X11 GenDate: 2003/01/11 17:14:34 UTC ModifiedFiles: programs/winedbg/stabs.c AddedFiles: =================================================================== RCS file: /home/cvs/cvsroot/wine/wine/programs/winedbg/stabs.c,v retrieving revision 1.3 diff -u -u -r1.3 stabs.c --- programs/winedbg/stabs.c 29 Sep 2002 18:35:31 -0000 1.3 +++ programs/winedbg/stabs.c 11 Jan 2003 17:14:30 -0000 @@ -31,6 +31,8 @@ * available (hopefully) from http:\\sources.redhat.com\gdb\onlinedocs */ +#define WITH_CPP_STABS_PARSING 1 + #include "config.h" #include <sys/types.h> @@ -269,13 +271,13 @@ if (**x=='(') { (*x)++; /* '(' */ - filenr=strtol(*x,x,10); /* <int> */ + filenr=strtol(*x,x,10); /* <int> */ (*x)++; /* ',' */ - subnr=strtol(*x,x,10); /* <int> */ + subnr=strtol(*x,x,10); /* <int> */ (*x)++; /* ')' */ } else { filenr = 0; - subnr = strtol(*x,x,10); /* <int> */ + subnr = strtol(*x,x,10); /* <int> */ } return DEBUG_FileSubNr2StabEnum(filenr,subnr); } @@ -343,14 +345,58 @@ if (*ptd->ptr++ != ';') return -1; /* ';' */ return 0; } +#ifdef WITH_CPP_STABS_PARSING +static inline int DEBUG_PTS_ReadMethodInfo(struct ParseTypedefData* ptd, + struct datatype* dt) +{ + char* tmp; + + if (*ptd->ptr++ == ':') + { + if (!(tmp = strchr(ptd->ptr, ';'))) return -1; + ptd->ptr = tmp + 1; + + } + if (!(*ptd->ptr >= '0' && *ptd->ptr <= '9')) return -1; + ptd->ptr++; + if (!(ptd->ptr[0] >= 'A' && *ptd->ptr <= 'D')) return -1; + ptd->ptr++; + if (*ptd->ptr++ != '.') return -1; + if (*ptd->ptr != ';') + { + while (*ptd->ptr) + { + /* some GCC versions (2.96 for example) really spit out wrongly defined + * stabs... in some cases, the methods are not correctly handled + * so just swallow the garbage until either .; (end of method) + * or ;; (end of struct) are found + */ + if (ptd->ptr[0] == '.' && ptd->ptr[1] == ';') + { + ptd->ptr++; + break; + } + if (ptd->ptr[0] == ';' && ptd->ptr[1] == ';') + break; + ptd->ptr++; + } + } + + return 0; +} +#endif -static inline int DEBUG_PTS_ReadAggregate(struct ParseTypedefData* ptd, struct datatype* sdt) +static inline int DEBUG_PTS_ReadAggregate(struct ParseTypedefData* ptd, + struct datatype* sdt) { int sz, ofs; char* last; struct datatype* adt; int idx; int doadd; +#ifdef WITH_CPP_STABS_PARSING + BOOL inMethod; +#endif sz = strtol(ptd->ptr, &last, 10); if (last == ptd->ptr) return -1; @@ -366,29 +412,67 @@ while (*ptd->ptr != ';') { /* agg_name : type ',' <int:offset> ',' <int:size> */ idx = ptd->idx; + if (DEBUG_PTS_ReadID(ptd) == -1) return -1; /* Ref. TSDF R2.130 Section 7.4. When the field name is a method name * it is followed by two colons rather than one. */ - if (*ptd->ptr == ':') ptd->ptr++; + if (*ptd->ptr == ':') + { + ptd->ptr++; +#ifdef WITH_CPP_STABS_PARSING + { + char* psemic = strchr(ptd->ptr, ';'); + char* psharp = strchr(ptd->ptr, '#'); + if (!psharp || psharp > psemic) + { + struct datatype* dt = NULL; + /* FIXME: hack... this shall be a ctor */ + if (DEBUG_PTS_ReadTypedef(ptd, NULL, &dt) == -1) return -1; + if (DEBUG_PTS_ReadMethodInfo(ptd, dt) == -1) return -1; + if (*ptd->ptr++ != ';') return -1; + ptd->idx = idx; + continue; + } + } + inMethod = TRUE; + } + else + { + inMethod = FALSE; +#endif + } + if (DEBUG_PTS_ReadTypedef(ptd, NULL, &adt) == -1) return -1; if (!adt) return -1; - if (*ptd->ptr++ != ',') return -1; - if (DEBUG_PTS_ReadNum(ptd, &ofs) == -1) return -1; - if (*ptd->ptr++ != ',') return -1; - if (DEBUG_PTS_ReadNum(ptd, &sz) == -1) return -1; - if (*ptd->ptr++ != ';') return -1; + if (*ptd->ptr != ',') + { +#ifdef WITH_CPP_STABS_PARSING + if (!inMethod) +#endif + return -1; + ptd->ptr++; + } + else + { + ptd->ptr++; + if (DEBUG_PTS_ReadNum(ptd, &ofs) == -1) return -1; + if (*ptd->ptr++ != ',') return -1; + if (DEBUG_PTS_ReadNum(ptd, &sz) == -1) return -1; + if (*ptd->ptr++ != ';') return -1; - if (doadd) DEBUG_AddStructElement(sdt, ptd->buf + idx, adt, ofs, sz); + if (doadd) DEBUG_AddStructElement(sdt, ptd->buf + idx, adt, ofs, sz); + } ptd->idx = idx; } ptd->ptr++; /* ';' */ return 0; } -static inline int DEBUG_PTS_ReadEnum(struct ParseTypedefData* ptd, struct datatype* edt) +static inline int DEBUG_PTS_ReadEnum(struct ParseTypedefData* ptd, + struct datatype* edt) { int ofs; int idx; @@ -405,7 +489,8 @@ return 0; } -static inline int DEBUG_PTS_ReadArray(struct ParseTypedefData* ptd, struct datatype* adt) +static inline int DEBUG_PTS_ReadArray(struct ParseTypedefData* ptd, + struct datatype* adt) { int lo, hi; struct datatype* rdt; @@ -463,14 +548,17 @@ /* then the real definitions */ switch (*ptd->ptr++) { case '*': +#ifdef WITH_CPP_STABS_PARSING + case '&': +#endif new_dt = DEBUG_NewDataType(DT_POINTER, NULL); if (DEBUG_PTS_ReadTypedef(ptd, NULL, &ref_dt) == -1) return -1; DEBUG_SetPointerType(new_dt, ref_dt); break; - case 'k': /* 'const' modifier */ - case 'B': /* 'volatile' modifier */ - /* just kinda ignore the modifier, I guess -gmt */ - if (DEBUG_PTS_ReadTypedef(ptd, NULL, &new_dt) == -1) return -1; + case 'k': /* 'const' modifier */ + case 'B': /* 'volatile' modifier */ + /* just kinda ignore the modifier, I guess -gmt */ + if (DEBUG_PTS_ReadTypedef(ptd, NULL, &new_dt) == -1) return -1; break; case '(': ptd->ptr--; @@ -588,7 +676,8 @@ case 31: basic = DT_BASIC_LONGLONGINT; break; case 32: basic = DT_BASIC_ULONGLONGINT; break; default: - DEBUG_Printf(DBG_CHN_MESG, "Unsupported integral type (%d/%d)\n", lo, sz); + DEBUG_Printf(DBG_CHN_MESG, + "Unsupported integral type (%d/%d)\n", lo, sz); return -1; } if (!(new_dt = DEBUG_GetBasicType(basic))) { @@ -598,6 +687,35 @@ if (*ptd->ptr++ != ';') return -1; } break; +#ifdef WITH_CPP_STABS_PARSING + case '#': + new_dt = DEBUG_NewDataType(DT_FUNC, NULL); + if (*ptd->ptr == '#') + { + ptd->ptr++; + if (DEBUG_PTS_ReadTypedef(ptd, NULL, &ref_dt) == -1) return -1; + DEBUG_SetPointerType(new_dt, ref_dt); + if (*ptd->ptr++ != ';') return -1; + + } + else + { + struct datatype* cls_dt; + struct datatype* pmt_dt; + + if (DEBUG_PTS_ReadTypedef(ptd, NULL, &cls_dt) == -1) return -1; + if (*ptd->ptr++ != ',') return -1; + if (DEBUG_PTS_ReadTypedef(ptd, NULL, &ref_dt) == -1) return -1; + DEBUG_SetPointerType(new_dt, ref_dt); + while (*ptd->ptr++ == ',') + { + if (DEBUG_PTS_ReadTypedef(ptd, NULL, &pmt_dt) == -1) return -1; + } + } + if (DEBUG_PTS_ReadMethodInfo(ptd, new_dt) == -1) return -1; + + break; +#endif default: DEBUG_Printf(DBG_CHN_MESG, "Unknown type '%c'\n", ptd->ptr[-1]); return -1; @@ -647,11 +765,15 @@ if ((ptd.ptr = strchr(ptr, ':'))) { ptd.ptr++; if (*ptd.ptr != '(') ptd.ptr++; +#ifdef WITH_CPP_STABS_PARSING + /* most of type definitions take one char, except Tt */ + if (*ptd.ptr != '(') ptd.ptr++; +#endif ret = DEBUG_PTS_ReadTypedef(&ptd, typename, &dt); } if (ret == -1 || *ptd.ptr) { - DEBUG_Printf(DBG_CHN_MESG, "failure on %s at %s\n", ptr, ptd.ptr); + DEBUG_Printf(DBG_CHN_MESG, "Failure on %s at %s\n", ptr, ptd.ptr); return FALSE; } @@ -729,7 +851,7 @@ strtabinc = 0; stabbuff[0] = '\0'; - for(i=0; i < nstab; i++, stab_ptr++ ) + for (i = 0; i < nstab; i++, stab_ptr++) { ptr = strs + (unsigned int) stab_ptr->n_un.n_name; if( ptr[strlen(ptr) - 1] == '\\' ) @@ -944,7 +1066,7 @@ } #endif - if( *ptr == '\0' ) + if( *ptr == '\0' ) /* end of N_SO file */ { /* * Nuke old path. @@ -1004,7 +1127,7 @@ stabbuff[0] = '\0'; #if 0 - DEBUG_Printf(DBG_CHN_MESG, "%d %x %s\n", stab_ptr->n_type, + DEBUG_Printf(DBG_CHN_MESG, "0x%02x %x %s\n", stab_ptr->n_type, (unsigned int) stab_ptr->n_value, strs + (unsigned int) stab_ptr->n_un.n_name); #endif @@ -1044,7 +1167,7 @@ nsym = symtab->sh_size / sizeof(*symp); strp = (char *) (addr + strtab->sh_offset); - for(i=0; i < nsym; i++, symp++) + for (i = 0; i < nsym; i++, symp++) { /* * Ignore certain types of entries which really aren't of that much @@ -1165,7 +1288,7 @@ } if (stabsect == -1 || stabstrsect == -1) { - DEBUG_Printf(DBG_CHN_WARN, "no .stab section\n"); + DEBUG_Printf(DBG_CHN_WARN, "No .stab section\n"); goto leave; } @@ -1181,7 +1304,7 @@ dil = DIL_LOADED; } else { dil = DIL_ERROR; - DEBUG_Printf(DBG_CHN_WARN, "bad stabs\n"); + DEBUG_Printf(DBG_CHN_WARN, "Couldn't read correctly read stabs\n"); goto leave; } @@ -1350,7 +1473,9 @@ if (filename == NULL) return DIL_ERROR; if (DEBUG_FindModuleByName(filename, DMT_ELF)) return DIL_LOADED; +#ifndef WITH_CPP_STABS_PARSING if (strstr (filename, "libstdc++")) return DIL_ERROR; /* We know we can't do it */ +#endif dil = DEBUG_ProcessElfFile(filename, load_offset, dyn_addr); /* if relative pathname, try some absolute base dirs */ @@ -1383,6 +1508,7 @@ for (lm_addr = (u_long)dbg_hdr->r_map; lm_addr; lm_addr = (u_long)lm.l_next) { if (!DEBUG_READ_MEM_VERBOSE((void*)lm_addr, &lm, sizeof(lm))) return FALSE; + if (lm.l_addr != 0 && DEBUG_READ_MEM_VERBOSE((void*)lm.l_addr, &ehdr, sizeof(ehdr)) && ehdr.e_type == ET_DYN && /* only look at dynamic modules */

1 0

Re: patch for "GetFileTitleW: illegal call to GetFileTitleA"
by Dmitry Timoshkov 11 Jan '03

11 Jan '03

"liu spider" <liuspider(a)yahoo.com> wrote: > this is a patch to get rid of W->A calls: > GetFileTitleW: illegal call to GetFileTitleA > + ret = GetFileTitleW(lpwstrFile, lpwstrTitle, cbBuf); > + HeapFree(GetProcessHeap(), 0, lpwstrTitle); > + HeapFree(GetProcessHeap(), 0, lpwstrFile); > + return ret; > +} It would be a good idea to translate the result from unicode to ANSI. One more suggestion: RtlCreateUnicodeStringFromAsciiz is supposed to reduce trouble of conversion ASCIIZ strings to unicode. -- Dmitry.

2 1

Some ddraw / GDI palette problem
by Lionel Ulmer 11 Jan '03

11 Jan '03

Hi all, In 'old' D3D games, a lot of textures are created paletted... And each of these texture has its own DDraw palette object. And thus each has some GDI memory allocated. This means that after some time, we run out of GDI Heap space due to all these palettes created there : trace:ddraw:Main_DirectDraw_CreatePalette (0x4038ff48)->(00000044,0x406b1340,0x406b1978,(nil)) trace:palette:CreatePalette entries=256 err:local:LOCAL_GetBlock not enough space in GDI heap 01b7 for 1052 bytes Now that problem is fixed in one game (it was due to some bad reference counting is some D3D code fixed in patch 79), but if ever it happens in yet another game, is it possible to increase this space ? Lionel -- Lionel Ulmer - http://www.bbrox.org/

2 1

Re: Help with /?? stuff
by John K. Hohm 11 Jan '03

11 Jan '03

> > Alberto Massari wrote: > > I guess Wine is converting the NULL into \?? to be able to store it as > > a string. Shachar Shemesh wrote: > You guessed wrong. The leading \?? is there under both Wine and Windows > 2000, and in both cases there is ALSO a NULL. As neither Wine nor > Windows 2000 then accepts that as a valid path, I'll just remove it. I > was just wondering whether anyone knows why Windows 2000 puts that extra > stuff in (it's obvious why wine does that - to be compatible). I'm not sure about \??, but see this excerpt from MSDN Re CreateFile: lpFileName [in] Pointer to a null-terminated string that specifies the name of the object to create or open. In the ANSI version of this function, the name is limited to MAX_PATH characters. To extend this limit to 32,767 wide characters, call the Unicode version of the function and prepend "\\?\" to the path. For more information, see Naming a File. Windows 95/98/Me: This string must not exceed MAX_PATH characters.

3 2

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

wine-devel