December 2002 - Wine-Devel - list.winehq.org

Re: Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library

by Dimitrie O. Paun

On December 20, 2002 02:07 am, Francois Gouget wrote: > Does this work? It seems to work very well! It can go to Alexandre as far as I'm concerned. Thanks. -- Dimi.

Dec. 20, 2002

Re: rpcdce.h fixes

by Dimitrie O. Paun

On December 20, 2002 04:55 am, Ove Kaaven wrote: > It does, but it was more difficult to get things (rpcrt4 in particular) to > compile without warnings (or huge piles of typecasts) this way, so I > decided to just use LPSTR/LPWSTR instead in the rpc headers. Didn't think > it would cause problems. It does, in C++ because of stronger typing than in C. We can do a conditional thing #ifdef __WINE__ ... Alexandre? -- Dimi.

Dec. 20, 2002

Re: ThunkTheTemplateHandle and Kazaa

by Ender

> It's works for me on Wine-cvs-20021219. Kazaa was broken for a time, > but has now work for at least 2-weeks. I don't think the problem was in > Wine, the solution was to load "winedefault.reg". Yes I've got all that. Kazaa 2.0.2 (or is it 2.2... the new one anyway) calls ThunkTheTemplateHandle when running in Win98 mode. This function needs an implementation to work - a lot of people have similar problems if you read the compatability db and forums... I suspect their problem is also this function. So getting a working implementation (and as I said, from my uneducated debugging it seems to be nothing much more than a wrapper function) into WINE would be very nice :) Cheers, Ender

Dec. 20, 2002

Conformance tests failing

by Paul Millar

Hi everyone, The extra support added by Chris in: http://cvs.winehq.com/patch.py?id=6743 doesn't seem be enough for wine to pass all of his new conformance tests added in: http://cvs.winehq.com/patch.py?id=6744 A few of them failed, caused WRT reported: locale.c:164: Test failed 1 of 1 time (100%): [test instance 1] locale.c:164: Test failed: GetTimeFormat: got 6 instead of 8 locale.c:146: Test failed 1 of 1 time (100%): [test instance 1] locale.c:146: Test failed: GetTimeFormat got '08' instead of '8 AM' locale.c:147: Test failed 1 of 1 time (100%): [test instance 1] locale.c:147: Test failed: GetTimeFormat: got 3 instead of 5 locale.c:163: Test failed 1 of 1 time (100%): [test instance 1] locale.c:163: Test failed: GetTimeFormat got '08:56' instead of '8:56 AM' for 95, 98, me, NT, Win2k, ... (the database is still updating at the moment) Details of the build are available at: http://www.astro.gla.ac.uk/users/paulm/WRT/buildinfo.php?id=271 ---- Paul Millar

Dec. 20, 2002

Re: I need help debugging COM

by Ove Kaaven

On Fri, 20 Dec 2002, Mehmet YASAR wrote: > Hi, > > I'm working on enhancing Wine's Ddraw surface code. I have not enough > experience of COM (and very small knowledge asm) to understand why I > have a crash on the following line in my test app test.exe : > > lpov->Lock(NULL, &SurfaceDesc, DDLOCK_WAIT, NULL); > > Here is more info on the crash : > > Wine-dbg>c > First chance exception: page fault on read access to 0x00000064 in > 32-bit code (0x00401a7b). I'd say this probably points to that the vtbl of the interface pointed to by lpov has not been initialized (the vtbl seems to be a null pointer). Perhaps you should look into the surface constructor code.

Dec. 20, 2002

I need help debugging COM

by Mehmet YASAR

Hi, I'm working on enhancing Wine's Ddraw surface code. I have not enough experience of COM (and very small knowledge asm) to understand why I have a crash on the following line in my test app test.exe : lpov->Lock(NULL, &SurfaceDesc, DDLOCK_WAIT, NULL); Here is more info on the crash : Wine-dbg>c First chance exception: page fault on read access to 0x00000064 in 32-bit code (0x00401a7b). Register dump: CS:0023 SS:002b DS:002b ES:002b FS:008f GS:0007 EIP:00401a7b ESP:406d28e8 EBP:406d2c68 EFLAGS:00010206( R- 00 I - -P1 ) EAX:403c670c EBX:4086b8bc ECX:403c670c EDX:00000000 ESI:406d2bec EDI:406d2c64 Stack dump: 0x406d28e8 (NTDLL.DLL.memcpy+0x4287b8): 403c670c 00000000 406d2bfc 00000001 0x406d28f8 (NTDLL.DLL.memcpy+0x4287c8): 00000000 00020022 00000000 40098c19 0x406d2908 (NTDLL.DLL.memcpy+0x4287d8): 4086b478 400fa444 406d2934 400be095 0x406d2918 (NTDLL.DLL.memcpy+0x4287e8): 4086b478 00000002 400fa25f 40110360 0x406d2928 (NTDLL.DLL.memcpy+0x4287f8): 4086b8bc 00000000 406d29f0 406d2944 0x406d2938 (NTDLL.DLL.memcpy+0x428808): 408116fc 4086b478 4086b8bc 406d29d8 0x406d2948 (NTDLL.DLL.memcpy+0x428818): 0011: sel=008f base=40110360 limit=00000fff 32-bit rw- Backtrace: =>0 0x00401a7b (test.exe..text+0xa7b in Y:\devel\test.exe) (ebp=406d2c68) 1 0x004028ee (test.exe..text+0x18ee in Y:\devel\test.exe) (ebp=406d2d0c) 2 0x4081a3c3 (WINPROC_wrapper+0x17 in user32.dll.so) (ebp=406d2d30) 3 0x4081a45d (WINPROC_CallWndProc+0x8d(proc=0x402860, hwnd=0x20022, msg=0x708, wParam=0x0, lParam=0x0) [winproc.c:183] in user32.dll.so) (ebp=406d2d60) ll.so) (ebp=406d2d60) 4 0x40820e52 (CallWindowProcA+0xb2(func=0x40e80458, hwnd=0x20022, msg=0x708, wParam=0x0, lParam=0x0) [winproc.c:2789] in user32.dll.so) (ebp=406d2d94) l.so) (ebp=406d2d94) 5 0x40802114 (DispatchMessageA+0x134(msg=0x406d2de8) [message.c:1094] in user32.dll.so) (ebp=406d2dd8) 6 0x00402747 (test.exe..text+0x1747 in Y:\devel\test.exe) (ebp=406d2e04) 7 0x00402fe5 (test.exe.EntryPoint+0xce in Y:\devel\test.exe) (ebp=406d2e90) 8 0x400b92bf (start_process+0x24f [process.c:564] in libntdll.dll.so) (ebp=406d2f38) 9 0x400bda27 (call_on_thread_stack+0x27(func=0x400b9070) [sysdeps.c:112] in libntdll.dll.so) (ebp=406d2ff4) 10 0x400bdbe0 (SYSDEPS_CallOnStack+0x14 in libntdll.dll.so) (ebp=00000000) 0x00401a7b (test.exe..text+0xa7b in Y:\devel\test.exe): call *0x64(%edx) Wine-dbg>disas 0x00401a7b (test.exe..text+0xa7b in Y:\devel\test.exe): call *0x64(%edx) 0x00401a7e (test.exe..text+0xa7e in Y:\devel\test.exe): movl %eax,0xffffff5c(%ebp) 0x00401a84 (test.exe..text+0xa84 in Y:\devel\test.exe): movl 0xffffff5c(%ebp),%ecx 0x00401a8a (test.exe..text+0xa8a in Y:\devel\test.exe): pushl %ecx 0x00401a8b (test.exe..text+0xa8b in Y:\devel\test.exe): pushl $0x40a164 0x00401a90 (test.exe..text+0xa90 in Y:\devel\test.exe): pushl $0x1 0x00401a92 (test.exe..text+0xa92 in Y:\devel\test.exe): call 0x00402510 (test.exe..text+0x1510 in Y:\devel\test.exe) 0x00401a97 (test.exe..text+0xa97 in Y:\devel\test.exe): addl $12,%esp 0x00401a9a (test.exe..text+0xa9a in Y:\devel\test.exe): cmpl $0x88760104,0xffffff5c(%ebp) 0x00401aa4 (test.exe..text+0xaa4 in Y:\devel\test.exe): jz 0x00401ab2 (test.exe..text+0xab2 in Y:\devel\test.exe) I have certainly introduced bugs in ddraw code, but where should I look ? Mehmet

Dec. 20, 2002

Re: rpcdce.h fixes

by Ove Kaaven

On Fri, 20 Dec 2002, Dimitrie O. Paun wrote: > MinGW has them defined this way, and wxWindows likes them like this. > Can someone please check that this matches the Windows headers? It does, but it was more difficult to get things (rpcrt4 in particular) to compile without warnings (or huge piles of typecasts) this way, so I decided to just use LPSTR/LPWSTR instead in the rpc headers. Didn't think it would cause problems. > RPCRTAPI RPC_STATUS RPC_ENTRY > - RpcStringFreeA(LPSTR* String); > + RpcStringFreeA(unsigned char** String); > RPCRTAPI RPC_STATUS RPC_ENTRY > - RpcStringFreeW(LPWSTR* String); > + RpcStringFreeW(unsigned short** String); > #define RpcStringFree WINELIB_NAME_AW(RpcStringFree)

Dec. 20, 2002

Re: ThunkTheTemplateHandle and Kazaa

by Acke Carlsson

On Friday 20 December 2002 09:34 am, Ender wrote: > Hi all. > > Trying to run any recent version of Kazaa (Lite) under WINE fails for me, > as it trys to call an unimplemented krnl386.exe stub... ordinal 538 > (ThunkTheTemplateHandle). > > After doing a little minimal debugging, all I could see that ordinal doing > was messing with some segment pointers and calling ordinal 630 (C16ThkSL). > > I really have no idea what any of this stuff does (*cough*), although > copying the line for ordinal 630 into ordinal 538 makes Kazaa work for me > > :) > > I don't suppose anybody with a little more knowledge of this bit of > Windows knows what actually goes on here? It'd be nice to see this fixed > (properly :) in CVS, as Kazaa is a fairly popular application. > > Cheers, > Ender It's works for me on Wine-cvs-20021219. Kazaa was broken for a time, but has now work for at least 2-weeks. I don't think the problem was in Wine, the solution was to load "winedefault.reg". ${HOME}/.wine/config ---------------SNIP--------------- [AppDefaults\\kazaa.exe\\DllOverrides] "commctrl" = "native" "comctl32" = "native" "shdoclc" = "native" "shdocvw" = "native" "shlwapi" = "native" "ole32" = "native" "oleaut32" = "native" "rpcrt4" = "native" ---------------SNIP---------------

Dec. 20, 2002

ThunkTheTemplateHandle and Kazaa

by Ender

Hi all. Trying to run any recent version of Kazaa (Lite) under WINE fails for me, as it trys to call an unimplemented krnl386.exe stub... ordinal 538 (ThunkTheTemplateHandle). After doing a little minimal debugging, all I could see that ordinal doing was messing with some segment pointers and calling ordinal 630 (C16ThkSL). I really have no idea what any of this stuff does (*cough*), although copying the line for ordinal 630 into ordinal 538 makes Kazaa work for me :) I don't suppose anybody with a little more knowledge of this bit of Windows knows what actually goes on here? It'd be nice to see this fixed (properly :) in CVS, as Kazaa is a fairly popular application. Cheers, Ender

Dec. 20, 2002

Re: Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library

by Dimitrie O. Paun

On December 20, 2002 02:07 am, Francois Gouget wrote: > Does this work? It might, but if I try to compile with the glibc headers, I get other problems not in that _stricmp, and _strdup are not declared. Which is odd, as I seem to have compiled with glibc headers a few files before... It's way past my bed time, I'll into it tomorrow. -- Dimi.

Dec. 20, 2002

Re: HELP: DEVMODE def

by Dimitrie O. Paun

On December 19, 2002 07:48 pm, Dan Kegel wrote: > gcc3.2 has them. I just verified that a fragment like that compiles fine > with g++3.2. Thank you. ChangeLog Enable nameless structures in C++ for gcc >= 3.x. (Discovery and testing done by Dan Kegel). Index: include/winnt.h =================================================================== RCS file: /var/cvs/wine/include/winnt.h,v retrieving revision 1.147 diff -u -r1.147 winnt.h --- include/winnt.h 19 Dec 2002 21:12:35 -0000 1.147 +++ include/winnt.h 20 Dec 2002 06:38:42 -0000 @@ -140,8 +140,8 @@ # define NONAMELESSSTRUCT # define NONAMELESSUNION #else -/* Anonymous struct support starts with gcc 2.96 */ -# if !defined(NONAMELESSSTRUCT) && (defined(__GNUC__) && (defined(__cplusplus) || ((__GNUC__ < 2) || ((__GNUC__ == 2) && (__GNUC_MINOR__ < 96))))) || defined(__SUNPRO_C) || defined(__SUNPRO_CC) +/* Anonymous struct support starts with gcc 2.96 or gcc/g++ 3.x */ +# if !defined(NONAMELESSSTRUCT) && (defined(__GNUC__) && ((__GNUC__ < 2) || ((__GNUC__ == 2) && (defined(__cplusplus) || (__GNUC_MINOR__ < 96))))) || defined(__SUNPRO_C) || defined(__SUNPRO_CC) # define NONAMELESSSTRUCT # endif /* Anonymous unions support starts with gcc 2.96/g++ 2.95 */ -- Dimi.

Dec. 20, 2002

Re: Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library

by Francois Gouget

On Wed, 18 Dec 2002, Dimitrie O. Paun wrote: > Folks, > > This does not work (<= is include): > winsock.h > <= sys/types.h > <= sys/select.h > > BANG! FD_CLR gets defined before we have a chance to manipulate it. > > Any ideas on how to fix this? Does this work? There might still be problems with FreeBSD but I did not have time to check today... Index: include/winsock.h =================================================================== RCS file: /home/wine/wine/include/winsock.h,v retrieving revision 1.49 diff -u -r1.49 winsock.h --- include/winsock.h 21 Nov 2002 23:44:19 -0000 1.49 +++ include/winsock.h 20 Dec 2002 01:14:39 -0000 @@ -56,29 +56,83 @@ /* * This section defines the items that conflict with the Unix headers. */ +#if !defined(USE_WS_PREFIX) && !defined(__WINE_USE_MSVCRT) +/* We are not using the WS_ prefix and not using the MSVCRT either so we + * risk getting conflicts for everything related to select. + */ +# ifdef FD_CLR +/* Too late, the Unix version of stdlib.h was included before winsock.h. + * This means select and all the related stuff is already defined and we + * cannot override types and function prototypes. + * All we can do is disable all these symbols so that they are not used + * inadvertantly. + */ +# undef FD_SETSIZE +# undef FD_CLR +# undef FD_SET +# undef FD_ZERO +# undef FD_ISSET -#ifndef __WINE_USE_MSVCRT -/* Get the u_xxx types from the Unix headers. They will do and doing it - * this way will avoid redefinitions. But on FreeBSD we may get macros - * and prototypes for htonl & co. This means the functions will not be - * called because of the macros. So this should not harm us too much unless - * we try to define our own prototypes (different calling convention). +# define FD_SETSIZE Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library +# define FD_CLR Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library +# define FD_SET Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library +# define FD_ZERO Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library +# define FD_ISSET Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library +# define fd_set Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library +# define select Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library +# include <sys/types.h> +# ifdef __MINGW_H + /* MinGW doesn't define the u_xxx types */ +# define WS_DEFINE_U_TYPES +# endif +# else +/* stdlib.h has not been included yet so it's not too late. Include it now + * making sure that none of the select symbols is affected. Then we can + * define them with our own values. */ +# define fd_set unix_fd_set +# define timeval unix_timeval +# define select unix_select +# include <sys/types.h> +# ifndef FD_CLR + /* On Windows-style platforms, sys/types.h does not define FD_CLR + * and does not define the u_xxx types + */ +# define WS_DEFINE_U_TYPES +# endif +# undef fd_set +# undef timeval +# undef select +# undef FD_SETSIZE +# undef FD_CLR +# undef FD_SET +# undef FD_ZERO +# undef FD_ISSET + +# define WS_DEFINE_SELECT +# endif /* FD_CLR */ + +#elif !defined(__WINE_USE_MSVCRT) +# define WS_DEFINE_SELECT # include <sys/types.h> -# if defined(USE_WS_PREFIX) || !defined(htonl) -# define WS_DEFINE_HTONL -# endif /* htonl */ +# ifndef FD_CLR + /* On Windows-style platforms, sys/types.h does not define FD_CLR + * and does not define the u_xxx types + */ +# define WS_DEFINE_U_TYPES +# endif #else -/* Since we are using the MSVCRT headers, we must define the u_xxx - * types ourselves. - */ +# define WS_DEFINE_SELECT + /* Our MSVCRT headers don't define the u_xxx types */ +# define WS_DEFINE_U_TYPES +#endif /* !USE_WS_PREFIX && !__WINE_USE_MSVCRT */ + +#ifdef WS_DEFINE_U_TYPES typedef unsigned char u_char; typedef unsigned short u_short; typedef unsigned int u_int; typedef unsigned long u_long; -# define WS_DEFINE_HTONL -#endif /* __WINE_USE_MSVCRT */ - +#endif /* @@ -349,55 +403,6 @@ * Select */ -#if !defined(USE_WS_PREFIX) && !defined(__WINE_USE_MSVCRT) -/* We are not using the WS_ prefix and not using the MSVCRT either so we - * risk getting conflicts for everything related to select. - */ -# ifdef FD_CLR -/* Too late, the Unix version of stdlib.h was included before winsock.h. - * This means select and all the related stuff is already defined and we - * cannot override types and function prototypes. - * All we can do is disable all these symbols so that they are not used - * inadvertantly. - */ -# undef FD_SETSIZE -# undef FD_CLR -# undef FD_SET -# undef FD_ZERO -# undef FD_ISSET - -# define FD_SETSIZE Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library -# define FD_CLR Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library -# define FD_SET Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library -# define FD_ZERO Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library -# define FD_ISSET Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library -# define fd_set Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library -# define select Include_winsock_h_before_stdlib_h_or_use_the_MSVCRT_library -# else -/* stdlib.h has not been included yet so it's not too late. Include it now - * making sure that none of the select symbols is affected. Then we can - * define them with our own values. - */ -# define fd_set unix_fd_set -# define timeval unix_timeval -# define select unix_select -# include <stdlib.h> -# undef fd_set -# undef timeval -# undef select -# undef FD_SETSIZE -# undef FD_CLR -# undef FD_SET -# undef FD_ZERO -# undef FD_ISSET - -# define WS_DEFINE_SELECT -# endif /* FD_CLR */ - -#else -# define WS_DEFINE_SELECT -#endif /* !USE_WS_PREFIX && !__WINE_USE_MSVCRT */ - #ifdef WS_DEFINE_SELECT /* Define our own version of select and the associated types and macros */ @@ -929,12 +934,16 @@ int WINAPI WS(shutdown)(SOCKET,int); SOCKET WINAPI WS(socket)(int,int,int); -#ifdef WS_DEFINE_HTONL +#if defined(htonl) && !defined(USE_WS_PREFIX) +# undef htonl +# undef htons +# undef ntohl +# undef ntohs +#endif u_long WINAPI WS(htonl)(u_long); u_short WINAPI WS(htons)(u_short); u_long WINAPI WS(ntohl)(u_long); u_short WINAPI WS(ntohs)(u_short); -#endif #if defined(__WINE__) || !defined(__WINE_WINSOCK2__) /* Stuff specific to winsock.h */ -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ It really galls me that most of the computer power in the world is wasted on screen savers. Chris Caldwell from the GIMPS project http://www.mersenne.org/prime.htm

Dec. 20, 2002

Re: Listview report mode tweaks

by Dimitrie O. Paun

On December 19, 2002 07:48 pm, Duane Clark wrote: > I also altered LISTVIEW_GetSubItemRect so that it reports correct > values, as compared against WinNT on ControlSpy. Problem is that LISTVIEW_GetSubItemRect is documented to return same thing for LVIR_LABEL and LVIR_BOUNDS. Of course, incorrect documentation is known to exist, so as long as you checked it... However, when we implement things different from the documented behavior, a clear comment explaining why we deviate is appropriate. Otherwise, someone might get rid of it in the future. :( -- Dimi.

Dec. 20, 2002

Re: strcasecmp

by David Fraser

Dimitrie O. Paun wrote: >On December 19, 2002 09:26 am, Waldek Hebisch wrote: > > >>If we are going to seriously support mingw then we need "mingw headers". >>I am not sure if true mingw headres are different enuugh from normal >>wine and msvcrt headers to warrant extra subdirectory, but logically >>they form a separate set. I suggest to put conditionls (if needed) in >>msvcrt headers corresponding to mingw headers, and make a separate >>subdirectory for mingw headers which emulate Unix headers. >> >> > >Right. I haven't looked at the mingw headers, but I assume they are >rather similar to the ones in glibc. If that's the case, we can >reuse those (if you have gcc which you have to anyway if you want to >use the mingwrap stuff): > >[dimi(a)dimi wine.src]$ sed -n 290,296p /usr/lib/gcc-lib/i386-redhat-linux/3.2/include/stddef.h >#ifndef __WCHAR_TYPE__ >#define __WCHAR_TYPE__ int >#endif >#ifndef __cplusplus >typedef __WCHAR_TYPE__ wchar_t; >#endif >#endif > >Which means we can simply -D__WCHAR_TYPE__=unsigned\ short > >But we still can not link. For that, we need to reimplement all wchar_t >using functions, and link those in before the glibc ones. Linking tricks >are not my strenght, anyone know how we can do that? Does it work to >just place a lib with those symbols first in the -l list? > > > Not sure exactly what you're looking for here - but I think you're wanting to know about the way it resolves symbols. Basically symbol lookups are only done in libraries that appear later on the command line than the one needing the symbol. So if libb.a depends on liba.a, then gcc bladibla -lb -la will work while gcc bladibla -la -lb will fail. Hope that's relevant David

Dec. 20, 2002

IE5.5 and 6.0sp1 error [Unimplimented]

by Boris

I just downloaded the latest cvs on a system without windows files. I tried to install IE5.5 and IE 6.0sp1 as win2k or winxp and as it tries to install, the following error occurs Unimplimented Kernel32.dll.VerSetConditionMask in 32bit mode. Anyone looking to write an update to kernel32.dll to add the function?

Dec. 19, 2002

Re: RECT def

by Dmitry Timoshkov

"Dimitrie O. Paun" <dpaun(a)rogers.com> wrote: > > ChangeLog > > MS uses LONG for the members of RECT. > > Of course, this causes a sh*tload of warnings. This may be a > good time to convert most of these to the debugrect function > that I used in listview.c. Sounds like a plan? Personally I like this idea. -- Dmitry.

Dec. 19, 2002

Re: Wine 0.9 TODO v0.7

by Dmitry Timoshkov

"Steven Edwards" <steven_ed4153(a)yahoo.com> wrote: > We still need to double check the latest import lib and winebuild changes on Mingw32. > I am going to be swamped for the next week or two. Alexandre told me that it works for him with binutils 2.13.90. I have tested it under Cygwin and can confirm that it works for me too after latest winebuild import fixes from Alexandre. Btw, you can suggest ReactOS people to look how handling of import libs via .def files done in Wine and get rid of that ugly .def/.edf stuff. -- Dmitry.

Dec. 19, 2002

About WineLib : there is a lack of API in wine

by Kevin

hi ::) I have an appliction of Windows. In the appliction i call createRemoteThread 、SetDCPenColor and setDCBrushcolor，but wine don't support those API functions ,if there has the related function in wine?? Thanks kevin wine-devel(a)winehq.com ----- Original Message ----- From: <wine-devel-request(a)winehq.com> To: <wine-devel(a)winehq.com> Sent: Friday, December 20, 2002 2:00 AM Subject: wine-devel digest, Vol 1 #1519 - 5 msgs > Send wine-devel mailing list submissions to > wine-devel(a)winehq.com > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.winehq.com/mailman/listinfo/wine-devel > or, via email, send a message with subject or body 'help' to > wine-devel-request(a)winehq.com > > You can reach the person managing the list at > wine-devel-admin(a)winehq.com > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of wine-devel digest..." > > > Today's Topics: > > 1. Re: [D3D8 PATCH] d3d8 shaders update - VS2 (David Laight) > 2. Re: [D3D8 PATCH] d3d8 shaders update - VS2 (Zsolt Rizsanyi) > 3. RE: Win2000 Conformance Test Results... (Kye Lewis) > 4. Re: strcasecmp (Dimitrie O. Paun) > 5. Re: Update to patches sgml (Tony Lambregts) > > --__--__-- > > Message: 1 > Date: Thu, 19 Dec 2002 14:54:21 +0000 > From: David Laight <david(a)l8s.co.uk> > To: "Dimitrie O. Paun" <dpaun(a)rogers.com> > Cc: Fabian Cenedese <Cenedese(a)indel.ch>, wine-devel(a)winehq.com > Subject: Re: [D3D8 PATCH] d3d8 shaders update - VS2 > > On Thu, Dec 19, 2002 at 07:08:40AM -0500, Dimitrie O. Paun wrote: > > On December 19, 2002 05:29 am, David Laight wrote: > > > No, and I would use it if I did :-) > ^^^ > not > > Finger trouble... > > > http://www.vim.org > > Probably... > > > David > > -- > David Laight: david(a)l8s.co.uk > > --__--__-- > > Message: 2 > From: Zsolt Rizsanyi <rizsanyi(a)myrealbox.com> > To: Fabian Cenedese <Cenedese(a)indel.ch>, wine-devel(a)winehq.com > Subject: Re: [D3D8 PATCH] d3d8 shaders update - VS2 > Date: Thu, 19 Dec 2002 16:06:51 +0100 > > On Thursday 19 December 2002 14:27, Fabian Cenedese wrote: > > Considering the age of these two it's maybe understandable why they > > don't have this feature. Tell me an editor in Windows and I may switch. > > There is windows version of Vim. Try that :) > > > --__--__-- > > Message: 3 > From: "Kye Lewis" <kyethespy(a)softhome.net> > To: "'Francois Gouget'" <fgouget(a)free.fr> > Cc: wine-devel(a)winehq.org > Subject: RE: Win2000 Conformance Test Results... > Date: Thu, 19 Dec 2002 16:30:59 +1100 > > XP Results Round 2: > > ** If a test is not listed here, then it has passed ** > > advapi32_test registry: No Change > > Note: Test now always has 3 failures, > No cases of more failures are apparent. > > kernel32_test process: No Change > > netapi32_test access: No Change > > shell32_test shfileop: No Change > > user32_test sysparams: No Change > > Thanks, > Kye Lewis > LinksOS Project Administrator > LinksOS Web Team Leader > kyethespy(a)softhome.net > (New email address! kyethespy(a)liquid2k.com will still work, but please > send all emails now to new email) > > "640k ought to be enough for anyone!" > - Bill Gates > > ================================= > LinksOS: Linking you to the world > http://linksos.cjb.net > http://linksos.sourceforge.net/ > http://linksos.oxygen-x.net/ > Now with Bugzilla and phpBB installed :) > ================================= > > > -----Original Message----- > From: wine-devel-admin(a)winehq.com [mailto:wine-devel-admin(a)winehq.com] > On Behalf Of Francois Gouget > Sent: Friday, December 13, 2002 8:58 AM > To: David Fraser > Cc: Kye Lewis; wine-devel(a)winehq.org > Subject: Re: Win2000 Conformance Test Results... > > > Hmmm, I meant to send this two days ago... > > > On Tue, 10 Dec 2002, David Fraser wrote: > [...] > > for kernel32_test path, I got this additional error (before all the > others) > > I have made tons of fixes to the kernel/path test and it should now work > much better. I just uploaded the new version of the tests compiled from > my sources. One can download them from the usual URL: > > http://fgouget.free.fr/wine/winetests.zip > > > [...] > > >process: 116 tests executed, 0 marked as todo, 4 failures. > > > > > Exactly the same except I have 10 messages interspersed with these, > saying: > > tests/process.c: 1 tests executed, 0 marked as todo, 0 failures. > > Why these extra messages? > > Because this executable invokes itself to test CreateProcess. And each > time it exits it prints statistics about how many tests passed and > failed, although the child processes are not actually performing any > test. So these messages can be ignored. > > > > >h:\wine\wine\dlls\shlwapi\tests\shreg.c:218: Test failed: (6) > > >h:\wine\wine\dlls\shlwapi\tests\shreg.c:219: Test failed: (6,43) > > >h:\wine\wine\dlls\shlwapi\tests\shreg.c:220: Test failed: > (3435973836) > [...] > > Same but I get 0 for all your 3435973836 which is 0xcccccccc. Must be > a > > debug/release build thing - variable not being initialized. > > Should not be the case. The code does: > dwType = -1; > dwRet = SHQueryValueExA( hKey, "Test3", NULL, &dwType, NULL, > &dwSize); > ok( dwType == REG_SZ, "(%lu)" , dwType); > > So we should either get -1 or something that makes sense, not > 0xcccccccc. This needs to be investigated more. > > > > Should we print out some of these results in hex as well to make life > > easier? > > We should print them in just one format, decimal or hex, whichever makes > more sense. > > > > >I will now be concentrating on working out what is wrong with these > > >tests :) > > Yep, I think this is priority number one now. Running the tests on more > platforms would only return pretty much the same list of bugs again and > again. > > > -- > Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ > Any sufficiently advanced bug is indistinguishable from a feature. > -- from some indian guy > > > > > > > --__--__-- > > Message: 4 > From: "Dimitrie O. Paun" <dpaun(a)rogers.com> > Reply-To: dpaun(a)rogers.com > Organization: DSSD Software Inc. > To: Waldek Hebisch <hebisch(a)math.uni.wroc.pl>, wine-devel(a)winehq.com > Subject: Re: strcasecmp > Date: Thu, 19 Dec 2002 11:48:23 -0500 > > On December 19, 2002 09:26 am, Waldek Hebisch wrote: > > If we are going to seriously support mingw then we need "mingw headers". > > I am not sure if true mingw headres are different enuugh from normal > > wine and msvcrt headers to warrant extra subdirectory, but logically > > they form a separate set. I suggest to put conditionls (if needed) in > > msvcrt headers corresponding to mingw headers, and make a separate > > subdirectory for mingw headers which emulate Unix headers. > > Right. I haven't looked at the mingw headers, but I assume they are > rather similar to the ones in glibc. If that's the case, we can > reuse those (if you have gcc which you have to anyway if you want to > use the mingwrap stuff): > > [dimi(a)dimi wine.src]$ sed -n 290,296p /usr/lib/gcc-lib/i386-redhat-linux/3.2/include/stddef.h > #ifndef __WCHAR_TYPE__ > #define __WCHAR_TYPE__ int > #endif > #ifndef __cplusplus > typedef __WCHAR_TYPE__ wchar_t; > #endif > #endif > > Which means we can simply -D__WCHAR_TYPE__=unsigned\ short > > But we still can not link. For that, we need to reimplement all wchar_t > using functions, and link those in before the glibc ones. Linking tricks > are not my strenght, anyone know how we can do that? Does it work to > just place a lib with those symbols first in the -l list? > > -- > Dimi. > > > --__--__-- > > Message: 5 > Date: Thu, 19 Dec 2002 10:36:14 -0700 > From: Tony Lambregts <tony_lambregts(a)telusplanet.net> > To: Joerg Mayer <jmayer(a)loplof.de> > CC: "Dimitrie O. Paun" <dpaun(a)rogers.com>, wine-devel(a)winehq.com > Subject: Re: Update to patches sgml > > Joerg Mayer wrote: > > >On Thu, Dec 19, 2002 at 01:10:52AM -0500, Dimitrie O. Paun wrote: > > > > > >>This should read "Patches should be inlined (if you can configure your email > >>client to not wrap lines), or attached as text/plain attachements...." > >> > >> > > > >Maybe you should include a step by step instruction for outlook users > >how to do that. It seems that outlook is the leading misattachment generator, > >so instead of just telling people what NOT to do, maybe you can tell > >people how to do it right. > > > > > > > Well I suppose I could if I had those instructions. However it seems out > of scope for a style section. AFAIK just giving the patch a .txt > extention and attaching will solve the problem for most mailers > including outlook. > > -- > > Tony Lambregts > > > > > > --__--__-- > > _______________________________________________ > wine-devel mailing list - wine-devel(a)winehq.com > http://www.winehq.com/mailman/listinfo/wine-devel > > > End of wine-devel Digest > >

Dec. 19, 2002

Re: HELP: DEVMODE def

by Dan Kegel

> On December 19, 2002 03:20 pm, Ove Kaaven wrote: >> union { >> struct { >> short dmOrientation; >> short dmPaperSize; >> short dmPaperLength; >> short dmPaperWidth; >> }; >> POINTL dmPosition; >> }; > > This is a problem, because it seems we don't have nameless structs > in g++, and I have no idea how to solve this. gcc3.2 has them. I just verified that a fragment like that compiles fine with g++3.2. Red Hat 8.0, Suse 8.1, and Mandrake 9.0 have already standardized on gcc3.2. Debian supports gcc3.2 only in their 'testing' distribution, but historically, a lot of people switch to 'testing' when they need new stuff. I have a feeling enough distributions have updated to gcc3.2 that it might be ok to start depending on it. - Dan -- Dan Kegel Linux User #78045 http://www.kegel.com

Dec. 19, 2002

Re: RECT def

by Dimitrie O. Paun

On December 19, 2002 02:35 pm, Dimitrie O. Paun wrote: > ChangeLog > MS uses LONG for the members of RECT. Of course, this causes a sh*tload of warnings. This may be a good time to convert most of these to the debugrect function that I used in listview.c. Sounds like a plan? -- Dimi.

Dec. 19, 2002

Re: HELP: DEVMODE def

by Ove Kaaven

On Thu, 19 Dec 2002, Dimitrie O. Paun wrote: > Thanks for the help guys, > > Here is another one: the DEVMODE structure. The full MSVC++ definition: typedef struct _devicemodeA { BYTE dmDeviceName[CCHDEVICENAME]; WORD dmSpecVersion; WORD dmDriverVersion; WORD dmSize; WORD dmDriverExtra; DWORD dmFields; union { struct { short dmOrientation; short dmPaperSize; short dmPaperLength; short dmPaperWidth; }; POINTL dmPosition; }; short dmScale; short dmCopies; short dmDefaultSource; short dmPrintQuality; short dmColor; short dmDuplex; short dmYResolution; short dmTTOption; short dmCollate; BYTE dmFormName[CCHFORMNAME]; WORD dmLogPixels; DWORD dmBitsPerPel; DWORD dmPelsWidth; DWORD dmPelsHeight; DWORD dmDisplayFlags; DWORD dmDisplayFrequency; #if(WINVER >= 0x0400) DWORD dmICMMethod; DWORD dmICMIntent; DWORD dmMediaType; DWORD dmDitherType; DWORD dmReserved1; DWORD dmReserved2; #if (WINVER >= 0x0500) || (_WIN32_WINNT >= 0x0400) DWORD dmPanningWidth; DWORD dmPanningHeight; #endif #endif /* WINVER >= 0x0400 */ } DEVMODEA, *PDEVMODEA, *NPDEVMODEA, *LPDEVMODEA;

Dec. 19, 2002

Re: Import or ntdll

by Alexandre Julliard

"Jeff Smith" <whydoubt(a)hotmail.com> writes: > Realized that ntdll was not needed in the IMPORTS list. I have tested > this in Linux and have not encountered any problems. It is needed. You don't see a problem on Linux because winebuild adds that import automatically in most cases, but it will break Reactos etc. -- Alexandre Julliard julliard(a)winehq.com

Dec. 19, 2002

Re: HELP: DEVMODE def

by Jeff Smith

At a glance, for wine's to conform with VC6's: 1.) "SHORT" -> "short" 2.) dmICMMethod and below only if WINVER >= 0x0400 3.) dmPanningWidth and below only if WINVER >= 0x0500 or _WIN32_WINN >= 0x0400 4.) Windows has no DUMMY*NAME*'s -- Jeff S >From: "Dimitrie O. Paun" <dpaun(a)rogers.com> > >Thanks for the help guys, > >Here is another one: the DEVMODE structure. _________________________________________________________________ Add photos to your messages with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail

Dec. 19, 2002

Re: HELP: DEVMODE def

by Vincent Béron

Dimitrie O. Paun a écrit: > Thanks for the help guys, > > Here is another one: the DEVMODE structure. > short rather than SHORT. The union is present. I have dmReserved1 and 2 in mine, along with the last 2 fields. (SDK downloaded last week). Vincent

Dec. 19, 2002

Re: another wtypes addition

by Ove Kaaven

On Thu, 19 Dec 2002, Dimitrie O. Paun wrote: > On December 19, 2002 02:32 pm, Ove Kaaven wrote: > > This is the last of the definition shuffling I had to do before my > > (rewind) build tree compiled with a replaced objidl.h. > > So we should be able to get rid of the include/wine/obj_*.h soon? Eventually. My objidl.h replacement only lets you get rid of the files currently referenced by it (obj_misc.h, obj_channel.h, obj_clientserver.h, obj_storage.h, obj_marshal.h, obj_moniker.h, obj_propertystorage.h, obj_dataobject.h), and indirectly obj_base.h, but many such files still remain and several more IDL files need to be written to replace those, for example oaidl.idl must be written before we can remove obj_oleaut.h, oleidl.idl is needed to get rid of yet a few more, and so on. I've already written an oaidl.idl, but not oleidl.idl (or any other of the remaining 53 IDL files in my copy of MSVC++).

Dec. 19, 2002

Re: HELP: RECT definition

by Vincent Béron

Dimitrie O. Paun a écrit: > Can someone please check the MS headers for the RECT definition? > > In our definition, we use INT, while the MinGW people use LONG. > Which one is correct? > LONG Vincent

Dec. 19, 2002

HELP: DEVMODE def

by Dimitrie O. Paun

Thanks for the help guys, Here is another one: the DEVMODE structure. Ours: typedef struct { BYTE dmDeviceName[CCHDEVICENAME]; WORD dmSpecVersion; WORD dmDriverVersion; WORD dmSize; WORD dmDriverExtra; DWORD dmFields; union { struct { SHORT dmOrientation; SHORT dmPaperSize; SHORT dmPaperLength; SHORT dmPaperWidth; } DUMMYSTRUCTNAME1; POINTL dmPosition; } DUMMYUNIONNAME1; SHORT dmScale; SHORT dmCopies; SHORT dmDefaultSource; SHORT dmPrintQuality; SHORT dmColor; SHORT dmDuplex; SHORT dmYResolution; SHORT dmTTOption; SHORT dmCollate; BYTE dmFormName[CCHFORMNAME]; WORD dmLogPixels; DWORD dmBitsPerPel; DWORD dmPelsWidth; DWORD dmPelsHeight; DWORD dmDisplayFlags; DWORD dmDisplayFrequency; DWORD dmICMMethod; DWORD dmICMIntent; DWORD dmMediaType; DWORD dmDitherType; DWORD dmReserved1; DWORD dmReserved2; DWORD dmPanningWidth; DWORD dmPanningHeight; } DEVMODEA, *PDEVMODEA, *LPDEVMODEA; MinGW: typedef struct _devicemodeA { BYTE dmDeviceName[CCHDEVICENAME]; WORD dmSpecVersion; WORD dmDriverVersion; WORD dmSize; WORD dmDriverExtra; DWORD dmFields; short dmOrientation; short dmPaperSize; short dmPaperLength; short dmPaperWidth; short dmScale; short dmCopies; short dmDefaultSource; short dmPrintQuality; short dmColor; short dmDuplex; short dmYResolution; short dmTTOption; short dmCollate; BYTE dmFormName[CCHFORMNAME]; WORD dmLogPixels; DWORD dmBitsPerPel; DWORD dmPelsWidth; DWORD dmPelsHeight; DWORD dmDisplayFlags; DWORD dmDisplayFrequency; DWORD dmICMMethod; DWORD dmICMIntent; DWORD dmMediaType; DWORD dmDitherType; DWORD dmICCManufacturer; DWORD dmICCModel; } DEVMODEA,*LPDEVMODEA,*PDEVMODEA; -- Dimi.

Dec. 19, 2002

Re: Update to patches sgml

by Joerg Mayer

On Thu, Dec 19, 2002 at 12:47:51PM -0600, Jeff Smith wrote: > Leading spaces are getting dropped when I inline and even when I attach > as a TXT file. The only way it will let all my patches through > unharmed is by leaving the extension as DIFF. I have several patches > in my queue, but have been reluctant to send them out until I get this > resolved. I don't think that resolving the problems should be a requirement to actually sned patches. Send them out now and just write a short note about currently not being able to "properly" attach them. The attachment requirement is intended to increase the speed of development, not slow it down. ciao Jörg -- Joerg Mayer <jmayer(a)loplof.de> I found out that "pro" means "instead of" (as in proconsul). Now I know what proactive means.

Dec. 19, 2002

Re: another wtypes addition

by Dimitrie O. Paun

On December 19, 2002 02:32 pm, Ove Kaaven wrote: > This is the last of the definition shuffling I had to do before my > (rewind) build tree compiled with a replaced objidl.h. So we should be able to get rid of the include/wine/obj_*.h soon? -- Dimi.

Dec. 19, 2002

HELP: RECT definition

by Dimitrie O. Paun

Can someone please check the MS headers for the RECT definition? In our definition, we use INT, while the MinGW people use LONG. Which one is correct? Ours: typedef struct tagRECT { INT left; INT top; INT right; INT bottom; } RECT, *PRECT, *LPRECT; MinGW: typedef struct tagRECT { LONG left; LONG top; LONG right; LONG bottom; } RECT,*PRECT,*LPRECT; -- Dimi.

Dec. 19, 2002

Re: Update to patches sgml

by Joerg Mayer

On Thu, Dec 19, 2002 at 10:36:14AM -0700, Tony Lambregts wrote: > Joerg Mayer wrote: > >Maybe you should include a step by step instruction for outlook users > >how to do that. It seems that outlook is the leading misattachment generator, > >so instead of just telling people what NOT to do, maybe you can tell > >people how to do it right. > > Well I suppose I could if I had those instructions. I jsut dug out the following mail: ---------------- Message-ID: <01f801c27fc4$3060c100$c4823bd5(a)dmitry> From: "Dmitry Timoshkov" <dmitry(a)baikal.ru> To: "Patrik Stridvall" <ps(a)leissner.se> Cc: <wine-devel(a)winehq.com> References: <9B72374533931944B8501EFE41A391E5A50CE2(a)knasen.leissner.se> Subject: Re: winapi update X-Spam-Level: "Patrik Stridvall" <ps(a)leissner.se> wrote: > > On October 29, 2002 11:27 am, Patrik Stridvall wrote: > > > Anyway, I can't inline my patches the because Microsoft > > Outlook wraps > > > the lines and I sure a lot of other people that use attachments have > > > similar problems. > > > > And can't be turned off?!? Hmmm, anyway, just attach them as > > text/plain. > > I have no idea how or if it is possible to do either. This is a part of my mail sent to Guy in order to help him to send patches using Outlook Express. I hope that it will be helpful for Outlook users too. <QUOTE> Since I was using OE to send patches, I have found a way to send diffs as attachments and without \n vs. \r\n problem. You need following two things to make it work. 1. Make sure that .diff files have \r\n line ends, because if OE detects that there is no \r\n line endings it switches to quoted-printable format attachments. 2. Using regedit add key "Content Type" with value "text/plain" to the .diff extension under HKEY_CLASSES_ROOT (same as for .txt extension). This tells OE to use Content-Type: text/plain instead of application/octet-stream. </QUOTE> Item #1 is important. After you hit "Send" button, go to "Outbox" and using "Properties" verify the message source to make sure that the mail has correct format. You might want to send several test e-mails to yourself too. -- Dmitry. ---------------- > However it seems out of scope for a style section. You are right, but only in theory. The reality is, that people easily do the "wrong" thing while trying to do the "right" thing. And that's not because they didsomething stupid, but because but because M$ products tend to know better what the users needs to do thatn the user himself :-/ > AFAIK just giving the patch a .txt > extention and attaching will solve the problem for most mailers > including outlook. Then this should be tested by people who have access to O and OE and added afterwards. Ciao Jörg -- Joerg Mayer <jmayer(a)loplof.de> I found out that "pro" means "instead of" (as in proconsul). Now I know what proactive means.

Dec. 19, 2002

Re: Wine 0.9 TODO v0.7

by Steven Edwards

We still need to double check the latest import lib and winebuild changes on Mingw32. I am going to be swamped for the next week or two. --- "Dimitrie O. Paun" <dpaun(a)rogers.com> wrote: > Hi folks, > > The next version of the 0.9 TODO is way past due, so I've > decided to release it today: > http://www.dssd.ca/wine/Wine-0.9-TODO-0.7.html > > ChangeLog > * Bill finished moving CVS docu from User Guide to Devel Guide (B3) > * Alexandre implemented import libraries (D3), finished winebuild (D4) > * I fixed header location (mostly done by Alexandre :)) (E9) > * Status update from Jaco (C2), and Ender (E7) > * Brand new Introduction > * New tasks added (D5, D6, D7) > > Check it out, it's fun! :) > > -- > Dimi. > > __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com

Dec. 19, 2002

Re: Update to patches sgml (one more time)

by Tony Lambregts

Dimitrie O. Paun wrote: >On December 19, 2002 01:56 am, Tony Lambregts wrote: > > >>Not a bad idea. how's this then. >> >> > >Better. Did you look at this: > http://www.winehq.com/hypermail/wine-devel/2002/09/0127.html > >The rules detailed in there are still valid: > -- unified diff only (required) > -- have a decent subject (recommended) > -- a long description (optional, if the change warrants it) > -- a meaningful ChangeLog entry (required) > -- new files, if any, included in patch, diffed against /dev/null (required) > -- patch inlined at the end of the message (recommended) > -- one changeset per message > Most of this is covered in the Patch format section and are in point form so I added most of what was missing there. > >What about including them in point-form for clarity, with some rationale >(as you did) for things like patch inlining, etc. Idea being that one >shouldn't have read the entire story if he/she only wants to see the rules. > >The story is good for first timers that want to understand why we have these >rules/recommendations, so we can easily point them to them instead of having >to answer it in long email every time... :) > > > The whole thing isn't much more that a couple hundred words and in point form. I think most developrers are capable of getting the idea without having to go back to it each time they send a patch <g> Change Log: update patches.sgml to include more information on style and format of patches Files: documentation/patches.sgml -- Tony Lambregts Index: patches.sgml =================================================================== RCS file: /home/wine/wine/documentation/patches.sgml,v retrieving revision 1.4 diff -u -r1.4 patches.sgml --- patches.sgml 17 Sep 2002 00:07:03 -0000 1.4 +++ patches.sgml 19 Dec 2002 18:36:29 -0000 @@ -14,18 +14,28 @@ <itemizedlist> <listitem> <para> - a description of what was wrong and what is now better - (and now broken :). + A meaningfull subject (very short description of patch) + </para> + </listitem> + <listitem> + <para> + A long (paragraph) description of what was wrong and what is now + better (and now broken :). (recomended) + </para> + </listitem> + <listitem> + <para> + Change Log: A short description of what was changed. </para> </listitem> <listitem> <para> - your contact information ( Name/Handle and e-mail ) + Your contact information ( Name/Handle and e-mail ) </para> </listitem> <listitem> <para> - the patch in <command>diff -u</command> format (it happens...) + The patch in <command>diff -u</command> format (it happens...) </para> </listitem> </itemizedlist> @@ -47,6 +57,82 @@ <para> For removals, list the files. </para> + <para> + Since wine is constantly changing due to development it is strongly + recomended that you use cvs for patches, if you cannot use cvs for + some reason, you can submit patches against the latest tarball. + To do this make a copy of the files that you will be modifying and + <command>diff -u</command> against the old file. I.E. + </para> + <screen> +diff -u file.old file.c > file.txt + </screen> + </sect1> + + <sect1 id="Style-notes"> + <title>Some notes about style</title> + + <para> + There are a few conventions that about coding style that have been + adopted over the years of development. The rational for these + <quote>rules</quote> is explained for each one. + </para> + <itemizedlist> + <listitem> + <para> + Only one change set per patch. Patches should address only one + bug/problem at a time. If a lot of changes need to be made then it + is perfered to break it into a series of patches. This makes it + easier to find regressions. + </para> + </listitem> + <listitem> + <para> + Tabs are not forbidden but are defined as 8 charaters and the usual + amount of indentation is 4 characters. + </para> + </listitem> + <listitem> + <para> + C++ style comments are discouraged since some compilers choke on + them. + </para> + </listitem> + <listitem> + <para> + Commenting out a block of code is usually done by enclosing it in + <command>#if 0 ... #endif</command> Statements. For example. + </para> + <screen> +/* note about reason for commenting block */ +#if 0 +code +code /* comments */ +code +#endif + </screen> + <para> + The reason for using this method is that it does not require that + you edit comments that may be inside the block of code. + </para> + </listitem> + <listitem> + <para> + Patches should be inlined (if you can configure your email client to + not wrap lines), or attached as plain text attachements so they can + be read inline. This may mean some more work for you. However it + allows others to review your patch easily and decreases the chances + of it being overlooked or forgotten. + </para> + </listitem> + <listitem> + <para> + Code is usually limited to 80 columns. This helps prevent mailers + mangling patches by line wrap. Also it generally makes code easier + to read. + </para> + </listitem> + </itemizedlist> </sect1> <sect1 id="patch-quality">

Dec. 19, 2002

Re: Update to patches sgml

by Jeff Smith

Hotmail better get their act together or I will have to go elsewhere. Leading spaces are getting dropped when I inline and even when I attach as a TXT file. The only way it will let all my patches through unharmed is by leaving the extension as DIFF. I have several patches in my queue, but have been reluctant to send them out until I get this resolved. -- Jeff S >From: Tony Lambregts <tony_lambregts(a)telusplanet.net> > >Joerg Mayer wrote: > >>On Thu, Dec 19, 2002 at 01:10:52AM -0500, Dimitrie O. Paun wrote: >> >> >>>This should read "Patches should be inlined (if you can configure your >>>email >>>client to not wrap lines), or attached as text/plain attachements...." >>> >>> >> >>Maybe you should include a step by step instruction for outlook users >>how to do that. It seems that outlook is the leading misattachment >>generator, >>so instead of just telling people what NOT to do, maybe you can tell >>people how to do it right. >> >> >> >Well I suppose I could if I had those instructions. However it seems out of >scope for a style section. AFAIK just giving the patch a .txt extention and >attaching will solve the problem for most mailers including outlook. > >-- > >Tony Lambregts _________________________________________________________________ MSN 8 with e-mail virus protection service: 2 months FREE* http://join.msn.com/?page=features/virus

Dec. 19, 2002

Re: Wineconsole 'regression'

by Eric Pouech

> I looked at the wineconsole code and it does not seem that there is a > configure option or any other mean to have wineconsole act as before (ie to > use the slow but working USER driver). that's a feature (see previous discussion on wineconsole implementation) we could add an option for that, but it's better to fix actually the curses implementation *first* does this attached patch solves some issues ? (it won't solve all of the issues for ida) A+ -- Eric Pouech Index: curses.c =================================================================== RCS file: /home/cvs/cvsroot/wine/wine/programs/wineconsole/curses.c,v retrieving revision 1.1 diff -u -r1.1 curses.c --- curses.c 13 Dec 2002 23:37:06 -0000 1.1 +++ curses.c 19 Dec 2002 18:29:25 -0000 @@ -136,22 +136,28 @@ { int x, y; CHAR_INFO* cell; - WORD color; + DWORD attr; for (y = tp; y <= bm; y++) { cell = &data->cells[y * data->curcfg.sb_width]; for (x = 0; x < data->curcfg.sb_width; x++) { - color = 0; - if (cell[x].Attributes & FOREGROUND_RED) color |= COLOR_RED; - if (cell[x].Attributes & FOREGROUND_BLUE) color |= COLOR_BLUE; - if (cell[x].Attributes & FOREGROUND_GREEN) color |= COLOR_GREEN; - if (cell[x].Attributes & BACKGROUND_RED) color |= COLOR_RED << 3; - if (cell[x].Attributes & BACKGROUND_BLUE) color |= COLOR_BLUE << 3; - if (cell[x].Attributes & BACKGROUND_GREEN) color |= COLOR_GREEN << 3; + if (cell[x].Char.UnicodeChar >= 0x20 && cell[x].Char.UnicodeChar <= 0xFF) + attr = LOBYTE(cell[x].Char.UnicodeChar); + else + attr = ' '; + + if (cell[x].Attributes & FOREGROUND_RED) attr |= COLOR_PAIR(COLOR_RED); + if (cell[x].Attributes & FOREGROUND_BLUE) attr |= COLOR_PAIR(COLOR_BLUE); + if (cell[x].Attributes & FOREGROUND_GREEN) attr |= COLOR_PAIR(COLOR_GREEN); + if (cell[x].Attributes & BACKGROUND_RED) attr |= COLOR_PAIR(COLOR_RED << 3); + if (cell[x].Attributes & BACKGROUND_BLUE) attr |= COLOR_PAIR(COLOR_BLUE << 3); + if (cell[x].Attributes & BACKGROUND_GREEN) attr |= COLOR_PAIR(COLOR_GREEN << 3); - PRIVATE(data)->line[x] = (unsigned char)cell[x].Char.UnicodeChar | COLOR_PAIR(color); + if (cell[x].Attributes & FOREGROUND_INTENSITY) attr |= A_BOLD; + + PRIVATE(data)->line[x] = attr; } mvwaddchnstr(PRIVATE(data)->pad, y, 0, PRIVATE(data)->line, data->curcfg.sb_width); } @@ -282,10 +288,9 @@ static COORD pos /* = {0, 0} */; MEVENT mevt; - BOOL ret = 0; if (getmouse(&mevt) == ERR) - return FALSE; + return FALSE; WINE_TRACE("[%u]: (%d, %d) %08lx\n", mevt.id, mevt.x, mevt.y, (unsigned long)mevt.bstate); @@ -296,7 +301,6 @@ #define BTN3_BIT FROM_LEFT_2ND_BUTTON_PRESSED #define BTN4_BIT 0 /* not done yet */ - /* FIXME: to be checked */ if (mevt.bstate & BUTTON1_PRESSED) bstate |= BTN1_BIT; if (mevt.bstate & BUTTON1_RELEASED) bstate &= ~BTN1_BIT; if (mevt.bstate & BUTTON2_PRESSED) bstate |= BTN2_BIT; @@ -304,28 +308,12 @@ if (mevt.bstate & BUTTON3_PRESSED) bstate |= BTN3_BIT; if (mevt.bstate & BUTTON3_RELEASED) bstate &= ~BTN3_BIT; - /* for the clicked & double click events, since we'll generate automatically - * the release event, we don't have to store the state - */ - if ((mevt.bstate & (BUTTON1_CLICKED|BUTTON1_DOUBLE_CLICKED)) && !(bstate & BTN1_BIT)) - { - ret = BTN1_BIT; - } - if ((mevt.bstate & (BUTTON2_CLICKED|BUTTON2_DOUBLE_CLICKED)) && !(bstate & BTN2_BIT)) - { - ret = BTN2_BIT; - } - if ((mevt.bstate & (BUTTON3_CLICKED|BUTTON3_DOUBLE_CLICKED)) && !(bstate & BTN3_BIT)) - { - ret = BTN3_BIT; - } - ir->EventType = MOUSE_EVENT; ir->Event.MouseEvent.dwMousePosition.X = mevt.x; ir->Event.MouseEvent.dwMousePosition.Y = mevt.y; - ir->Event.MouseEvent.dwButtonState = (bstate | ret); - + ir->Event.MouseEvent.dwButtonState = bstate; + /* partial conversion */ ir->Event.MouseEvent.dwControlKeyState = 0; if (mevt.bstate & BUTTON_SHIFT) ir->Event.MouseEvent.dwControlKeyState |= SHIFT_PRESSED; @@ -338,19 +326,23 @@ */ ir->Event.MouseEvent.dwEventFlags = 0; +#if 0 + /* won't work, we need to still generate the first click event */ if ((mevt.bstate & BUTTON1_DOUBLE_CLICKED) && ((bstate|ret) & BTN1_BIT)) ir->Event.MouseEvent.dwEventFlags |= DOUBLE_CLICK; if ((mevt.bstate & BUTTON2_DOUBLE_CLICKED) && ((bstate|ret) & BTN2_BIT)) ir->Event.MouseEvent.dwEventFlags |= DOUBLE_CLICK; if ((mevt.bstate & BUTTON3_DOUBLE_CLICKED) && ((bstate|ret) & BTN3_BIT)) ir->Event.MouseEvent.dwEventFlags |= DOUBLE_CLICK; - if (mevt.x != pos.X || mevt.y != pos.Y) +#endif + if (!(mevt.bstate & (BUTTON1_PRESSED|BUTTON1_RELEASED|BUTTON2_PRESSED|BUTTON2_RELEASED|BUTTON3_PRESSED|BUTTON3_RELEASED)) && + (mevt.x != pos.X || mevt.y != pos.Y)) { ir->Event.MouseEvent.dwEventFlags |= MOUSE_MOVED; } pos.X = mevt.x; pos.Y = mevt.y; - - return ret; + + return FALSE; } /****************************************************************** @@ -659,7 +651,8 @@ nodelay(stdscr, TRUE); keypad(stdscr, TRUE); mousemask(0xffffffff, &PRIVATE(data)->initial_mouse_mask); - + /* no click generation... (--hmmm man page says -1 to disable) */ + mouseinterval(0); return TRUE; }

Dec. 19, 2002

Wine 0.9 TODO v0.7

by Dimitrie O. Paun

Hi folks, The next version of the 0.9 TODO is way past due, so I've decided to release it today: http://www.dssd.ca/wine/Wine-0.9-TODO-0.7.html ChangeLog * Bill finished moving CVS docu from User Guide to Devel Guide (B3) * Alexandre implemented import libraries (D3), finished winebuild (D4) * I fixed header location (mostly done by Alexandre :)) (E9) * Status update from Jaco (C2), and Ender (E7) * Brand new Introduction * New tasks added (D5, D6, D7) Check it out, it's fun! :) -- Dimi.

Dec. 19, 2002

Re: Update to patches sgml

by Tony Lambregts

Joerg Mayer wrote: >On Thu, Dec 19, 2002 at 01:10:52AM -0500, Dimitrie O. Paun wrote: > > >>This should read "Patches should be inlined (if you can configure your email >>client to not wrap lines), or attached as text/plain attachements...." >> >> > >Maybe you should include a step by step instruction for outlook users >how to do that. It seems that outlook is the leading misattachment generator, >so instead of just telling people what NOT to do, maybe you can tell >people how to do it right. > > > Well I suppose I could if I had those instructions. However it seems out of scope for a style section. AFAIK just giving the patch a .txt extention and attaching will solve the problem for most mailers including outlook. -- Tony Lambregts

Dec. 19, 2002

Re: strcasecmp

by Dimitrie O. Paun

On December 19, 2002 09:26 am, Waldek Hebisch wrote: > If we are going to seriously support mingw then we need "mingw headers". > I am not sure if true mingw headres are different enuugh from normal > wine and msvcrt headers to warrant extra subdirectory, but logically > they form a separate set. I suggest to put conditionls (if needed) in > msvcrt headers corresponding to mingw headers, and make a separate > subdirectory for mingw headers which emulate Unix headers. Right. I haven't looked at the mingw headers, but I assume they are rather similar to the ones in glibc. If that's the case, we can reuse those (if you have gcc which you have to anyway if you want to use the mingwrap stuff): [dimi(a)dimi wine.src]$ sed -n 290,296p /usr/lib/gcc-lib/i386-redhat-linux/3.2/include/stddef.h #ifndef __WCHAR_TYPE__ #define __WCHAR_TYPE__ int #endif #ifndef __cplusplus typedef __WCHAR_TYPE__ wchar_t; #endif #endif Which means we can simply -D__WCHAR_TYPE__=unsigned\ short But we still can not link. For that, we need to reimplement all wchar_t using functions, and link those in before the glibc ones. Linking tricks are not my strenght, anyone know how we can do that? Does it work to just place a lib with those symbols first in the -l list? -- Dimi.

Dec. 19, 2002

Re: [D3D8 PATCH] d3d8 shaders update - VS2

by Zsolt Rizsanyi

On Thursday 19 December 2002 14:27, Fabian Cenedese wrote: > Considering the age of these two it's maybe understandable why they > don't have this feature. Tell me an editor in Windows and I may switch. There is windows version of Vim. Try that :)

Dec. 19, 2002

Re: [D3D8 PATCH] d3d8 shaders update - VS2

by David Laight

On Thu, Dec 19, 2002 at 07:08:40AM -0500, Dimitrie O. Paun wrote: > On December 19, 2002 05:29 am, David Laight wrote: > > No, and I would use it if I did :-) ^^^ not Finger trouble... > http://www.vim.org Probably... David -- David Laight: david(a)l8s.co.uk

Dec. 19, 2002

Re: [D3D8 PATCH] d3d8 shaders update - VS2

by Zsolt Rizsanyi

On Thursday 19 December 2002 10:14, Fabian Cenedese wrote: > > > >/* note about reason for commenting block */ > > > >#if 0 > > > >code > > > >code /* comments */ > > > >code > > > >#endif > > > > > > Whereas this is the solution with the least work it's also hard to > > > > spot. If > > > > > using > > > a text editor with syntax coloring it all looks the same. But if a code > > > > isn't > > > > > effective I'd like to see that. > > > >Actually it could be considered as an editor fault if it changes color for > >commented out code but doesn't recognize the #if 0/#endif construct. > > I wouldn't say so. A comment is easily recognized by the chars /* */ or //. > But to get a #if 0 or #if _ANYTHING you really need a language dependant > preprocessor which goes beyond a normal text editor, even one with > syntax coloring. > Do you have such a text editor that finds #if's? Yes :) Vim parses #if 0. So #if 0 ... #endif is colored like a comment. Of course that does not mean that all #if -s will be parsed, but IMO only #if 0 should be colored like a comment since other #if -s are not comments. Eg. if you have something like #if defined(__linux__) ... #elif defined(__bsd__) ... #endif You dont want to comment any out any part of the code since all code will be compiled on some OS. Regards Zsolt

Dec. 19, 2002

Re: strcasecmp

by Waldek Hebisch

> On December 18, 2002 05:06 pm, Francois Gouget wrote: > > There is the same issue gethostname(). Visual C++ does not define it in > > the msvcrt headers so how does one compile wxWindows with visual C++? > > Because these are apps that compile under MinGW using their headers for > libc, not the crtdll/msvcrt ones. > > In theory, we should compile such apps using the native libc headers, > but that does not work. And it does not work because of -fwchar-short, > which is a must... > > So we are in a bit of a bind: we _have_ to use msvcrt, but the app > expects functions that are present only in the Unix/MinGW headers. > What do we do? > > -- > Dimi. If we are going to seriously support mingw then we need "mingw headers". I am not sure if true mingw headres are different enuugh from normal wine and msvcrt headers to warrant extra subdirectory, but logically they form a separate set. I suggest to put conditionls (if needed) in msvcrt headers corresponding to mingw headers, and make a separate subdirectory for mingw headers which emulate Unix headers. -- Waldek Hebisch hebisch(a)math.uni.wroc.pl or hebisch(a)hera.math.uni.wroc.pl

Dec. 19, 2002

Re: How long does it take *you* to build wine?

by P. Christeas

Build (after a cvs update, make depend) on an Athlon XP 1800 w 256 Mb DDR: 1038.43user 72.66system 19:41.06elapsed 94%CPU That is 18.5 min. (the process was 'niced', so elapsed is more..)

Dec. 19, 2002

Re: [D3D8 PATCH] d3d8 shaders update - VS2

by Fabian Cenedese

>Yes, the version of vim that came with (I believe) Red Hat 6.0 >has the definitions in its syntax file for proper "#if 0" >highlighting. As common a practice as this has become, any >syntax-highlighting editor with some degree of maturity that >did *not* have this in its default rulesets for C/C++ would >surprise me. > >What language-independent, syntax-highlighting text editor are >you using? > >-- Jeff S >> >>I wouldn't say so. A comment is easily recognized by the chars /* */ or //. >>But to get a #if 0 or #if _ANYTHING you really need a language dependant >>preprocessor which goes beyond a normal text editor, even one with >>syntax coloring. >>Do you have such a text editor that finds #if's? There goes the editor war again :) As Windows developer I'm mostly using MSVC, still version 6, About says it's from 98. I don't know about .net studio if this supports #if 0. For our own product we use an open-source editor for Windows called CodeMax. An expanded version is here http://www.ticz.com/homes/users/nlewis/index.html?target=download Considering the age of these two it's maybe understandable why they don't have this feature. Tell me an editor in Windows and I may switch. As said before they don't color all files the same, they get the language from the file extension and I guess vim does the same. Recognizing the #if 0 is the only difference (well, at least concerning coloring :) On Linux I used Kate so far. Hmm, Kate also supports #if 0, never noticed that... bye Fabi A: No, see http://www.netmeister.org/news/learn2quote.html Q: Should I include quotations after my reply ?

Dec. 19, 2002

Re: [D3D8 PATCH] d3d8 shaders update - VS2

by Jeff Smith

Yes, the version of vim that came with (I believe) Red Hat 6.0 has the definitions in its syntax file for proper "#if 0" highlighting. As common a practice as this has become, any syntax-highlighting editor with some degree of maturity that did *not* have this in its default rulesets for C/C++ would surprise me. What language-independent, syntax-highlighting text editor are you using? -- Jeff S >From: Fabian Cenedese <Cenedese(a)indel.ch> > >> > >/* note about reason for commenting block */ >> > >#if 0 >> > >code >> > >code /* comments */ >> > >code >> > >#endif >> > >> > Whereas this is the solution with the least work it's also hard to >>spot. If >> > using >> > a text editor with syntax coloring it all looks the same. But if a code >>isn't >> > effective I'd like to see that. >> >>Actually it could be considered as an editor fault if it changes color for >>commented out code but doesn't recognize the #if 0/#endif construct. > >I wouldn't say so. A comment is easily recognized by the chars /* */ or //. >But to get a #if 0 or #if _ANYTHING you really need a language dependant >preprocessor which goes beyond a normal text editor, even one with >syntax coloring. >Do you have such a text editor that finds #if's? > >bye Fabi _________________________________________________________________ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail

Dec. 19, 2002

Re: critical section timeout for the dsound_test

by Sylvain Petreolle

I posted a thread on nvidia forum. --- Enrico Horn <EnricoHorn(a)t-online.de> a écrit : > Hi, > On Wednesday 18 December 2002 12:23, Sylvain Petreolle wrote: > > I have same issue, but the debugger manages to start and "ends" in > the > > X video binary. > > > no debugger for me, just the message. > > > Could be a stupid question, but what is your VIDEO driver ? > > Mine is nVidia binary, and I have got some problems with it. > > I have reproductible crashes with it (like this test) for a long > time. > > Don't worry, I know this is a dsound test... :) > > > yes i use the nvidia binary driver too, mine is version 1.0-3123 > Which one do you use? > Did you try to use the included nv driver? > I get some crashes in libGL esp. when closing an app. > I have always ignored those since they seemed nvidia's fault to me. > > Ciao > Farmboy > farmboy1(a)subdimension.com > > > > ===== > > Sylvain Petreolle > > spetreolle(a)users.sourceforge.net > > Fight against Spam ! http://www.euro.cauce.org/en/index.html > > ICQ #170597259 > > > > "Don't think you are. Know you are." Morpheus in Matrix, chapter > 15. > > > > ___________________________________________________________ > > Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! > > Yahoo! Mail : http://fr.mail.yahoo.com > ===== Sylvain Petreolle spetreolle(a)users.sourceforge.net Fight against Spam ! http://www.euro.cauce.org/en/index.html ICQ #170597259 "Don't think you are. Know you are." Morpheus in Matrix, chapter 15. ___________________________________________________________ Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! Yahoo! Mail : http://fr.mail.yahoo.com

Dec. 19, 2002

Re: Update to patches sgml

by Dimitrie O. Paun

On December 19, 2002 01:56 am, Tony Lambregts wrote: > Not a bad idea. how's this then. Better. Did you look at this: http://www.winehq.com/hypermail/wine-devel/2002/09/0127.html The rules detailed in there are still valid: -- unified diff only (required) -- have a decent subject (recommended) -- a long description (optional, if the change warrants it) -- a meaningful ChangeLog entry (required) -- new files, if any, included in patch, diffed against /dev/null (required) -- patch inlined at the end of the message (recommended) -- one changeset per message What about including them in point-form for clarity, with some rationale (as you did) for things like patch inlining, etc. Idea being that one shouldn't have read the entire story if he/she only wants to see the rules. The story is good for first timers that want to understand why we have these rules/recommendations, so we can easily point them to them instead of having to answer it in long email every time... :) -- Dimi.

Dec. 19, 2002

Re: [D3D8 PATCH] d3d8 shaders update - VS2

by Dimitrie O. Paun

On December 19, 2002 05:29 am, David Laight wrote: > No, and I would use it if I did :-) http://www.vim.org -- Dimi.

Dec. 19, 2002

Re: Update to patches sgml

by Joerg Mayer

On Thu, Dec 19, 2002 at 01:10:52AM -0500, Dimitrie O. Paun wrote: > This should read "Patches should be inlined (if you can configure your email > client to not wrap lines), or attached as text/plain attachements...." Maybe you should include a step by step instruction for outlook users how to do that. It seems that outlook is the leading misattachment generator, so instead of just telling people what NOT to do, maybe you can tell people how to do it right. Ciao Jörg -- Joerg Mayer <jmayer(a)loplof.de> I found out that "pro" means "instead of" (as in proconsul). Now I know what proactive means.

Dec. 19, 2002

Re: [D3D8 PATCH] d3d8 shaders update - VS2

by David Laight

> >Actually it could be considered as an editor fault if it changes color for > >commented out code but doesn't recognize the #if 0/#endif construct. I'd tend to agree... OTOH I don't like colour :-) > I wouldn't say so. A comment is easily recognized by the chars /* */ or //. No it isn't! You still have to know the language in order to get quoting right. Try editing an m4 source file! > But to get a #if 0 or #if _ANYTHING you really need a language dependant > preprocessor which goes beyond a normal text editor, even one with > syntax coloring. The only difficulty is knowing which half of the #if ... #else ... #endif to parse. Parsing both is wrong and will lead to miscoloration. > Do you have such a text editor that finds #if's? No, and I would use it if I did :-) David -- David Laight: david(a)l8s.co.uk

Dec. 19, 2002