March 2006 - Wine-Devel - list.winehq.org

Errors in Win API Stats

by Peter Åstrand

I've noticied that http://www.winehq.org/site/winapi_stats is not quite correct. For example, it claims that "amstream" is 100% implemented, which is not correct: There are 21 stubs or something like that. How is this page generated? Regards, -- Peter Åstrand ThinLinc Chief Developer Cendio http://www.cendio.se Teknikringen 3 583 30 Linköping Phone: +46-13-21 46 00

March 6, 2006

Re: shdocvw: Added get_LocationURL implementation.

by Jacek Caban

Robert Shearman wrote: > > BSTR and LPWSTR are not interchangable. A lot of the time you can get > away with it, but if anything tries to access the length that prefixes > a BSTR on the LPWSTR then bad things will happen. > Right, thanks. I don't know, why I have changed it. I think I passed BSTR in one of modifications in my tree. I'll send a fixed version. Thanks, Jacek

March 6, 2006

Re: shdocvw: Added get_LocationURL implementation.

by Robert Shearman

Jacek Caban wrote: >c24b7d0d0c498ec98785b93eba54b41d53ab2c9b >diff --git a/dlls/shdocvw/navigate.c b/dlls/shdocvw/navigate.c >index 9c4893c..40d4a06 100644 >--- a/dlls/shdocvw/navigate.c >+++ b/dlls/shdocvw/navigate.c >@@ -316,7 +316,7 @@ static IBindStatusCallback *create_callb > return BINDSC(ret); > } > >-static void on_before_navigate2(WebBrowser *This, PBYTE post_data, ULONG post_data_len, >+static void on_before_navigate2(WebBrowser *This, LPWSTR url, PBYTE post_data, ULONG post_data_len, > LPWSTR headers, VARIANT_BOOL *cancel) > { > VARIANT var_url, var_flags, var_frame_name, var_post_data, var_post_data2, var_headers; >@@ -367,7 +367,7 @@ static void on_before_navigate2(WebBrows > V_VT(params+5) = (VT_BYREF|VT_VARIANT); > V_VARIANTREF(params+5) = &var_url; > V_VT(&var_url) = VT_BSTR; >- V_BSTR(&var_url) = SysAllocString(This->url); >+ V_BSTR(&var_url) = url; > > V_VT(params+6) = (VT_DISPATCH); > V_DISPATCH(params+6) = (IDispatch*)WEBBROWSER2(This); > > BSTR and LPWSTR are not interchangable. A lot of the time you can get away with it, but if anything tries to access the length that prefixes a BSTR on the LPWSTR then bad things will happen. -- Rob Shearman

March 6, 2006

Re: FONTS: fix Courier font

by Dmitry Timoshkov

"Vitaly Lipatov" <lav(a)etersoft.ru> wrote: > > What are you trying to fix? > There is screenshot from notepad with Courier font selected in attachment. Do you see the same problem with Windows font? -- Dmitry.

March 6, 2006

Re: configure: add --with-fontfir (try3) with patch!

by Alexandre Julliard

Vitaly Lipatov <lav(a)etersoft.ru> writes: > В сообщении от 6 марта 2006 14:18 Alexandre Julliard написал(a): >> Vitaly Lipatov <lav(a)etersoft.ru> writes: >> > I don't like WINE fonts in %_datadir/fonts/wine, I it crossed >> > with ALT Linux font policy and so on. >> >> Where do you install them then? > %_datadir/wine/fonts > %_datadir/fonts is default fontconfig's path for font search. Well yes, that was the idea, that they should be found by fontconfig. Now of course if we make the code look in that path explicitly then I agree we might as well put them in datadir/wine. -- Alexandre Julliard julliard(a)winehq.org

March 6, 2006

Re: FONTS: fix Courier font

by Vitaly Lipatov

В сообщении от 6 марта 2006 15:07 Вы написали: > "Vitaly Lipatov" <lav(a)etersoft.ru> wrote: > > > Last time I asked Huw said that this is a bug in fontforge, > > > the font files have correct values. > > > > If the bug in fontforge will exists for ages... > > Well, I will use the patch as fast hack for courier font correcting. > > What are you trying to fix? There is screenshot from notepad with Courier font selected in attachment. -- Lav Виталий Липатов Россия, Санкт-Петербург. www.etersoft.ru GNU! ALT Linux Team! WINE! WIKI! LaTeX! LyX!

March 6, 2006

How to link and run applications using WinLib

by Arun Kumar Khandelwal

Hi All, I am able to compile and link my test program using winemaker. It generates a test.exe.so and not test.exe. I do not know how to run this *.so? What should I do to link and run program using winlib? Thanks, Arun

March 6, 2006

Re: Feedback request on wininet patch

by Robert Shearman

Victor Pelt wrote: >@@ -2447,7 +2485,18 @@ BOOL WINAPI InternetSetOptionW(HINTERNET > TRACE("Option INTERNET_OPTION_DISABLE_PASSPORT_AUTH: harmless stub, since not enabled\n"); > break; > case INTERNET_OPTION_RECEIVE_TIMEOUT: >- FIXME("Option INTERNET_OPTION_RECEIVE_TIMEOUT: STUB\n"); >+ { >+ unsigned long conns=*(unsigned long*)lpBuffer; >+ /* have HINTERNET hInternet , handle where to set the options */ >+ int result = setsockopt(INET_getSocketHelper(lpwhh), INET_getProtocolHelper(lpwhh), SO_RCVTIMEO, lpBuffer, (int)sizeof(lpBuffer)); >+ >+ if (result == 0) { >+ ret = TRUE; >+ } else { >+ /* Translate the result in the right error to return */ >+ ret = FALSE; >+ } >+ } > > It would be much better to implement this in netconnection.c. Perhaps calling it something like NETCONNECTION_set_timeout and allow both the send and receive timeouts to be set with it. -- Rob Shearman

March 6, 2006

Re: Error while compiling code using wineg++ / winegcc on solaris

by Dimi Paun

On Mon, 2006-03-06 at 14:47 +0530, Vikas Gera wrote: > testuser(a)testbox:~/test >wineg++ test.cpp > winegcc: -fshort-wchar failed. What version of gcc are you using? -- Dimi Paun <dimi(a)lattica.com> Lattica, Inc.

March 6, 2006

Re: Feedback request on wininet patch

by Vijay Kiran Kamuju

Hi Victor, If you are interested in implementing more features. You can have a look at my wininet pacthes, almost all of them have been rejected. I hope you might improve those or get new ideas from those. As i am going to pick some other well documented portion in windows api. Leaving the wininet for the time being, i might come back to it sometime lateer (in distant future). bye, Vijay On 3/6/06, Mike McCormack <mike(a)codeweavers.com> wrote: > > Victor Pelt wrote: > > > - FIXME("Option INTERNET_OPTION_RECEIVE_TIMEOUT: STUB\n"); > > + { > > It would be nice if you could use indentation consistent with the rest > of the file, and avoid tabs. > > > + unsigned long conns=*(unsigned long*)lpBuffer; > > What's 'conns' for? > > > + /* have HINTERNET hInternet , handle where to set the options */ > > + int result = setsockopt(INET_getSocketHelper(lpwhh), INET_getProtocolHelper(lpwhh), SO_RCVTIMEO, lpBuffer, (int)sizeof(lpBuffer)); > > + > > + if (result == 0) { > > + ret = TRUE; > > + } else { > > + /* Translate the result in the right error to return */ > > + ret = FALSE; > > + } > > + } > > How about "ret = (result == 0);"? > > > @@ -2455,6 +2504,9 @@ BOOL WINAPI InternetSetOptionW(HINTERNET > > case INTERNET_OPTION_CONNECT_RETRIES: > > FIXME("Option INTERNET_OPTION_CONNECT_RETRIES: STUB\n"); > > break; > > + case INTERNET_OPTION_CONTEXT_VALUE: > > + FIXME("Option INTERNET_OPTION_CONTEXT_VALUE; STUB\n"); > > + break; > > default: > > FIXME("Option %ld STUB\n",dwOption); > > INTERNET_SetLastError(ERROR_INVALID_PARAMETER); > > This chunk should be in a separate patch, since it's unrelated. > > > +/* Helper functions to get unix backend stuff based in windows handles (LPWININETHANDLEHEADER s) */ > > +int INET_getSocketHelper(LPWININETHANDLEHEADER); > > +int INET_getProtocolHelper(LPWININETHANDLEHEADER); > > Why not just make these static, and get rid of these declarations? > > Mike > > >

March 6, 2006

Re: Wine compilation error (flex related)

by Christian Lachner

ah, found something: http://sourceforge.net/tracker/index.php?func=detail&aid=1439346&group_id=9… greets On 3/6/06, Christian Lachner <gladiac.lists(a)gmail.com> wrote: > I have the same problem with a gcc-4.0 and flex 2.5.33. 2.5.4a works > ok ;). I don't think it's a wine-problem but the flex-releases are as > rare as a solar ecplipse so the fix must be in wine (maybe a > workaround or reformatting of the code since it's a syntaxing > problem). But i could also be wrong. > > On 2/27/06, Elvis Pranskevichus <elprans(a)gmail.com> wrote: > > Hello! > > > > Both emerge =wine-0.9.8 and the latest git workdir fail to compile on current > > Gentoo with: > > > > gcc -c -I. -I. -I../../include -I../../include > > -DINCLUDEDIR="\"/usr/local/include/wine\"" -Wall -pipe -fno-strict-aliasing > > -gstabs+ -Wdeclaration-after-statement -Wpointer-arith -g -O2 -o lex.yy.o > > lex.yy.c > > lex.yy.c:2610: error: parse error before numeric constant > > lex.yy.c: In function `yy_scan_string': > > lex.yy.c:2611: error: number of arguments doesn't match prototype > > lex.yy.c:305: error: prototype declaration > > lex.yy.c:2613: warning: passing arg 1 of `strlen' makes pointer from integer > > without a cast > > lex.yy.c:2613: warning: passing arg 1 of `yy_scan_bytes' makes pointer from > > integer without a cast > > make[2]: *** [lex.yy.o] Error 1 > > make[2]: Leaving directory `/home/el/projects/wine/wine-git/tools/wrc' > > make[1]: *** [wrc] Error 2 > > make[1]: Leaving directory `/home/el/projects/wine/wine-git/tools' > > make: *** [tools] Error 2 > > > > The flex --version is flex 2.5.33 (the latest one, released on February 21, > > 2006). > > gcc --version is gcc (GCC) 3.4.4 (Gentoo 3.4.4-r1, ssp-3.4.4-1.0, pie-8.7.8) > > > > widl and wmc lexers compiled fine. > > > > Downgrading to flex 2.5.4a makes everything compile again. > > Is this a flex issue or is a Wine problem? > > > > Thanks, > > -- > > Elvis Pranskevichus - elprans(a)gmail.com > > > > > > >

March 6, 2006

Re: Wine compilation error (flex related)

by Christian Lachner

I have the same problem with a gcc-4.0 and flex 2.5.33. 2.5.4a works ok ;). I don't think it's a wine-problem but the flex-releases are as rare as a solar ecplipse so the fix must be in wine (maybe a workaround or reformatting of the code since it's a syntaxing problem). But i could also be wrong. On 2/27/06, Elvis Pranskevichus <elprans(a)gmail.com> wrote: > Hello! > > Both emerge =wine-0.9.8 and the latest git workdir fail to compile on current > Gentoo with: > > gcc -c -I. -I. -I../../include -I../../include > -DINCLUDEDIR="\"/usr/local/include/wine\"" -Wall -pipe -fno-strict-aliasing > -gstabs+ -Wdeclaration-after-statement -Wpointer-arith -g -O2 -o lex.yy.o > lex.yy.c > lex.yy.c:2610: error: parse error before numeric constant > lex.yy.c: In function `yy_scan_string': > lex.yy.c:2611: error: number of arguments doesn't match prototype > lex.yy.c:305: error: prototype declaration > lex.yy.c:2613: warning: passing arg 1 of `strlen' makes pointer from integer > without a cast > lex.yy.c:2613: warning: passing arg 1 of `yy_scan_bytes' makes pointer from > integer without a cast > make[2]: *** [lex.yy.o] Error 1 > make[2]: Leaving directory `/home/el/projects/wine/wine-git/tools/wrc' > make[1]: *** [wrc] Error 2 > make[1]: Leaving directory `/home/el/projects/wine/wine-git/tools' > make: *** [tools] Error 2 > > The flex --version is flex 2.5.33 (the latest one, released on February 21, > 2006). > gcc --version is gcc (GCC) 3.4.4 (Gentoo 3.4.4-r1, ssp-3.4.4-1.0, pie-8.7.8) > > widl and wmc lexers compiled fine. > > Downgrading to flex 2.5.4a makes everything compile again. > Is this a flex issue or is a Wine problem? > > Thanks, > -- > Elvis Pranskevichus - elprans(a)gmail.com > > >

March 6, 2006

Re: FONTS: fix Courier font

by Dmitry Timoshkov

"Vitaly Lipatov" <lav(a)etersoft.ru> wrote: > > Last time I asked Huw said that this is a bug in fontforge, > > the font files have correct values. > If the bug in fontforge will exists for ages... > Well, I will use the patch as fast hack for courier font correcting. What are you trying to fix? -- Dmitry.

March 6, 2006

Re: configure: add --with-fontfir (try3) with patch!

by Vitaly Lipatov

В сообщении от 6 марта 2006 14:18 Alexandre Julliard написал(a): > Vitaly Lipatov <lav(a)etersoft.ru> writes: > > I don't like WINE fonts in %_datadir/fonts/wine, I it crossed > > with ALT Linux font policy and so on. > > Where do you install them then? %_datadir/wine/fonts %_datadir/fonts is default fontconfig's path for font search. -- Lav Виталий Липатов Россия, Санкт-Петербург. www.etersoft.ru GNU! ALT Linux Team! WINE! WIKI! LaTeX! LyX!

March 6, 2006

Floating point problem in WineEngGetGlyphOutline

by Jeff L

I have started to get a floating point problem "First chance exception: invalid float operation in 32-bit code (0x00e6b6f9).". I don't know a lot about floating point but from what I have read, the FLDCW most likely is unmasking an interrupt when it reloads the control word. Winedbg does not seem to have support for displaying floating point registers and control word etc. Can someone point me in the direction of getting to the bottom os this? Jeff Latimer Wine-dbg> 0x00e6b6d1 WineEngGetGlyphOutline+0x1d9 [/winf/dlls/gdi/freetype.c:2818] in gdi32: movl 0xffffff00(%ebp),%eax 2818 left = (INT)(ft_face->glyph->metrics.horiBearingX * widthRatio) & -64; Wine-dbg> 0x00e6b6d7 WineEngGetGlyphOutline+0x1df [/winf/dlls/gdi/freetype.c:2818] in gdi32: movl 0x54(%eax),%ecx 2818 left = (INT)(ft_face->glyph->metrics.horiBearingX * widthRatio) & -64; Wine-dbg> 0x00e6b6da WineEngGetGlyphOutline+0x1e2 [/winf/dlls/gdi/freetype.c:2818] in gdi32: movl 0x20(%ecx),%edx 2818 left = (INT)(ft_face->glyph->metrics.horiBearingX * widthRatio) & -64; Wine-dbg> 0x00e6b6dd WineEngGetGlyphOutline+0x1e5 [/winf/dlls/gdi/freetype.c:2818] in gdi32: fld %st(0) 2818 left = (INT)(ft_face->glyph->metrics.horiBearingX * widthRatio) & -64; Wine-dbg> 0x00e6b6df WineEngGetGlyphOutline+0x1e7 [/winf/dlls/gdi/freetype.c:2818] in gdi32: pushl %edx 2818 left = (INT)(ft_face->glyph->metrics.horiBearingX * widthRatio) & -64; Wine-dbg> 0x00e6b6e0 WineEngGetGlyphOutline+0x1e8 [/winf/dlls/gdi/freetype.c:2818] in gdi32: fimuls 0x0(%esp) 2818 left = (INT)(ft_face->glyph->metrics.horiBearingX * widthRatio) & -64; Wine-dbg> 0x00e6b6e3 WineEngGetGlyphOutline+0x1eb [/winf/dlls/gdi/freetype.c:2818] in gdi32: fnstcw 0xffffff92(%ebp) 2818 left = (INT)(ft_face->glyph->metrics.horiBearingX * widthRatio) & -64; Wine-dbg> 0x00e6b6e6 WineEngGetGlyphOutline+0x1ee [/winf/dlls/gdi/freetype.c:2818] in gdi32: movw 0xffffff92(%ebp),%ax 2818 left = (INT)(ft_face->glyph->metrics.horiBearingX * widthRatio) & -64; Wine-dbg> 0x00e6b6ea WineEngGetGlyphOutline+0x1f2 [/winf/dlls/gdi/freetype.c:2818] in gdi32: movb $0xc,%ah 2818 left = (INT)(ft_face->glyph->metrics.horiBearingX * widthRatio) & -64; Wine-dbg> 0x00e6b6ec WineEngGetGlyphOutline+0x1f4 [/winf/dlls/gdi/freetype.c:2818] in gdi32: movw %ax,0xffffff90(%ebp) 2818 left = (INT)(ft_face->glyph->metrics.horiBearingX * widthRatio) & -64; Wine-dbg> 0x00e6b6f0 WineEngGetGlyphOutline+0x1f8 [/winf/dlls/gdi/freetype.c:2818] in gdi32: fldcw 0xffffff90(%ebp) 2818 left = (INT)(ft_face->glyph->metrics.horiBearingX * widthRatio) & -64; Wine-dbg> 0x00e6b6f3 WineEngGetGlyphOutline+0x1fb [/winf/dlls/gdi/freetype.c:2818] in gdi32: fistps 0xffffff10(%ebp) 2818 left = (INT)(ft_face->glyph->metrics.horiBearingX * widthRatio) & -64; Wine-dbg> 0x00e6b6f9 WineEngGetGlyphOutline+0x201 [/winf/dlls/gdi/freetype.c:2818] in gdi32: fldcw 0xffffff92(%ebp) 2818 left = (INT)(ft_face->glyph->metrics.horiBearingX * widthRatio) & -64; Wine-dbg> Register dump: CS:0073 SS:007b DS:007b ES:007b FS:1007 GS:0033 EIP:00e6b6f9 ESP:7fc5cce0 EBP:7fc5ce2c EFLAGS:00200206( - 00 - -IP1) EAX:7d820c72 EBX:00e94bc8 ECX:7d82ed70 EDX:00030840 ESI:7fedcb40 EDI:0000001a Wine-dbg>stepi First chance exception: invalid float operation in 32-bit code (0x00e6b6f9). Register dump: CS:0073 SS:007b DS:007b ES:007b FS:1007 GS:0033 EIP:00e6b6f9 ESP:7fc5cce0 EBP:7fc5ce2c EFLAGS:00210306( - 00 -RITP1) EAX:7d820c72 EBX:00e94bc8 ECX:7d82ed70 EDX:00030840 ESI:7fedcb40 EDI:0000001a Stack dump: 0x7fc5cce0: 00030840 7fe90020 7fe90000 7fc5cd0c 0x7fc5ccf0: 0026d2aa 00000000 00000000 00000002 0x7fc5cd00: 00e94bc8 00000122 7fc5ce10 7fc5cd6c 0x7fc5cd10: 00e71fee 7fe90000 00000000 7fedc218 0x7fc5cd20: 7feddac2 001716ae 00e8edda 7d828d90 0x7fc5cd30: 0000000e 00000010 7d828f00 7d828e68 0200: sel=1007 base=7befc000 limit=00001fff 32-bit rw- Backtrace: =>1 0x00e6b6f9 WineEngGetGlyphOutline+0x201 [/winf/dlls/gdi/freetype.c:2818] in gdi32 (0x00e6b6f9) 2 0x00e72810 WineEngGetTextExtentPoint+0xd0(font=0x7fedc6b0, wstr=0x7fc5e3bc, count=0xa, size=0x7fc5cf54) [/winf/dlls/gdi/freetype.c:3685] in gdi32 (0x00e72810) 3 0x00e64347 GetTextExtentPoint32W+0x47(hdc=0x1c0, str=0x7fc5e3bc, count=0xa, size=0x7fc5cf54) [/winf/dlls/gdi/font.c:1104] in gdi32 (0x00e64347) 4 0x00e67010 GetCharacterPlacementW+0x134(hdc=0x1c0, lpString=0x7fc5e3bc, uCount=0xa, nMaxExtent=0x0, lpResults=0x7fc5e398, dwFlags=0x8103b) [/winf/dlls/gdi/font.c:2780] in gdi32 (0x00e67010) 5 0x004022b5 in script_test (+0x22b5) (0x004022b5) 6 0x00401141 in script_test (+0x1141) (0x00401141) 7 0x3266f6e6 in cc3260 (+0x6f6e6) (0x3266f6e6) 8 0x00000000 (0x00000000) 9 0x00793ccb wine_switch_to_stack+0x17 in libwine.so.1 (0x00793ccb) 0x00e6b6f9 WineEngGetGlyphOutline+0x201 [/winf/dlls/gdi/freetype.c:2818] in gdi32: fldcw 0xffffff92(%ebp) 2818 left = (INT)(ft_face->glyph->metrics.horiBearingX * widthRatio) & -64;

March 6, 2006

Re: FONTS: fix Courier font

by Vitaly Lipatov

В сообщении от 6 марта 2006 14:38 Dmitry Timoshkov написал(a): > "Vitaly Lipatov" <lav(a)etersoft.ru> wrote: > > The main change is > > -Ascent: 11 > > -Descent: 2 > > +Ascent: 10 > > +Descent: 3 > > Last time I asked Huw said that this is a bug in fontforge, > the font files have correct values. If the bug in fontforge will exists for ages... Well, I will use the patch as fast hack for courier font correcting. I tested with fontforge-20060209. -- Vitaly Lipatov, ALT Linux Team Russia, Saint-Petersburg, www.etersoft.ru

March 6, 2006

Re: FONTS: fix Courier font

by Dmitry Timoshkov

"Vitaly Lipatov" <lav(a)etersoft.ru> wrote: > The main change is > -Ascent: 11 > -Descent: 2 > +Ascent: 10 > +Descent: 3 Last time I asked Huw said that this is a bug in fontforge, the font files have correct values. -- Dmitry.

March 6, 2006

Re: configure: add --with-fontfir (try3) with patch!

by Alexandre Julliard

Vitaly Lipatov <lav(a)etersoft.ru> writes: > I don't like WINE fonts in %_datadir/fonts/wine, I it crossed > with ALT Linux font policy and so on. Where do you install them then? -- Alexandre Julliard julliard(a)winehq.org

March 6, 2006

Re: WINE & prelink

by Mike McCormack

Vitaly Lipatov wrote: > Can anyone explain here or at http://wiki.winehq.com/ about prelink usage in > WINE. Is it necessary for build process? > I see it is recommended at http://wiki.winehq.com/Recommended_Packages It's used in an attempt to fix the load address of a few builtin dlls (ntdll, kernel32, user) so that Windows programs that rely on the address being the same across different programs work correctly. Mike

March 6, 2006

WINE & prelink

by Vitaly Lipatov

Can anyone explain here or at http://wiki.winehq.com/ about prelink usage in WINE. Is it necessary for build process? I see it is recommended at http://wiki.winehq.com/Recommended_Packages -- Vitaly Lipatov, ALT Linux Team Russia, Saint-Petersburg, www.etersoft.ru

March 6, 2006

Error while compiling code using wineg++ / winegcc on solaris

by Vikas Gera

Hi!! I am not able to compile a test program on solaris using winegcc tool. Can anybody help? Here's the program source code: // file test.cpp #include <stdio.h> #include <windows.h> int main () { SYSTEMTIME lpSystemTime; GetSystemTime(&lpSystemTime); printf("Today is: %d/%d/%d\n", lpSystemTime.wYear, lpSystemTime.wMonth, lpSystemTime.wDay); return 0; } When I try to compile it using winegcc, it gives following error message: testuser(a)testbox:~/test >wineg++ test.cpp winegcc: -fshort-wchar failed. On adding a flag '-v' with wineg++, a detailed error message is displayed: testuser(a)testbox:~/test >wineg++ -v test.cpp -fshort-wchar -DWINE_UNICODE_NATIVE -D_REENTRANT -fPIC -DWIN32 -D_WIN32 -D__WIN32 -D__WIN32__ -D__WINNT -D__WINNT__ -D__stdcall=__attribute__((__stdcall__)) -D__cdecl=__attribute__((__cdecl__)) -D__fastcall=__attribute__((__fastcall__)) -D_stdcall=__attribute__((__stdcall__)) -D_cdecl=__attribute__((__cdecl__)) -D_fastcall=__attribute__((__fastcall__)) -D__declspec(x)=__declspec_##x -D__declspec_align(x)=__attribute__((aligned(x))) -D__declspec_allocate(x)=__attribute__((section(x))) -D__declspec_deprecated=__attribute__((deprecated)) -D__declspec_dllimport=__attribute__((dllimport)) -D__declspec_dllexport=__attribute__((dllexport)) -D__declspec_naked=__attribute__((naked)) -D__declspec_noinline=__attribute__((noinline)) -D__declspec_noreturn=__attribute__((noreturn)) -D__declspec_nothrow=__attribute__((nothrow)) -D__declspec_novtable=__attribute__(()) -D__declspec_selectany=__attribute__((weak)) -D__declspec_thread=__thread -D__WINE__ -D__int8=char -D__int16=short -D__int32=int -D__int64=long long -c -o test-9pH6Aj.o -v test.cpp -isystem/opt/cfw/wine/include/wine/windows winegcc: -fshort-wchar failed. I can see a lot of compile flags/options. I want to be able to enable/disable these options. Or if there is some other way to be able to compile the code using wingcc, pleae tell me. The code works well on windows. I even managed to compile it using gcc, but it resulted in coredump. If somebody can guide me how to compile an application using winelib, which can run on solaris. I am getting linker/loader errors. thanks!! Vikas Gera

March 6, 2006

Re: configure: add --with-fontfir (try3) with patch!

by Alexandre Julliard

"StartCom Ltd." <eddy_nigg(a)startcom.org> writes: > I would like to through in my support for Vitaly on this issue (and > maybe similar ones) which happen to be the source of some discussions > lately (rpath was one, I think). > > More configuration options makes wine to be more easy configured, build > and adjusted, making it more scalable....Distribution based patches > should be the last resort - not the first. If a configure option, such > as in this case can be so easily implemented, it should supported...As > long as it hard coded during make, than there should be a configuration > option for this (font path) but also others.... The problem with adding an option for every single thing that someone might want to change is that you end up with a million of them, burying the few important options in a mass of useless details. So for things that are only used to build some special package, it's perfectly reasonable IMO to ask the packager to pass options to make instead of configure, since it's all done from a script anyway. -- Alexandre Julliard julliard(a)winehq.org

March 6, 2006

FW: Re: Wine testing and Guild Wars: Factions Beta weekend March 24th

by Jeremiah Powell

Please see below. >From: Alexandre Julliard <julliard(a)winehq.org> >To: "Jeremiah Powell" <waveclaw(a)hotmail.com> >Subject: Re: Wine testing and Guild Wars: Factions Beta weekend March 24th > >"Jeremiah Powell" <waveclaw(a)hotmail.com> writes: > > > I would like to extend this opportunity to someone in the wine > > development community. Who would like a chance to collect data on the > > behavior of running Guild Wars under wine? > > > > How about for free? > >That sounds good, you should ask on wine-devel, I'm sure people would >be interested. > >-- >Alexandre Julliard >julliard(a)winehq.org I have 3 guest keys for the Guildwars: Factions BWE starting March 24th. Everyone I know already plays Guildwars on Windows. The Wine D3D support has gotten to the point of running Guildwars in a playable state[1]. I'd like offer these keys to the wine developers for debug purposes. In particular, I'd like committers on the D3D subsystem and the sound systems to enjoy a little free game time while collecting some valuable data[2]. There are only 3 keys. These are free TRIAL accounts, good for only that weekend. If you buy the game, you can open the accounts back up at a later time. The game client generates the account-contact emails from personal email addresses I give it, but they are in no other way assocaited with my account(s). Please reply to me directly if you want the keys, but keep the spam to a minimum. A list of people getting the keys can be posted if the list wants. JD 'too much hotmail spam anyway' Powell --- 1. While not stable, normal play is possible. I use SuSE OSS 10 with ATI's fglx drivers, and Cedega chokes on for this 'supported' game during screen changes, which the Wine D3D code handles gracefully. 2. From the appdb: Guildwars is a pure DirectX game that uses no opengl, no SDL, etc. Guildwars was functional under wine until the September 2005 Sorrow's Furnace update. Now game client has two modes in which it can be run. Normally, guildwars tried to use DirectX's shading language. This causes it to fail under Wine. With the '-noshaders' option, the game falls back on only the older DirectX calls. Other interesting features include toggling sound support with '-nosound' and output of triangles, fps, networking information with 'perfs' and starting in a window with '-windowed.' A 3rd party sound expansion is available that integrates with the DRM system in Media Player 9, but can be patched out by moving the DLL to another location.

March 6, 2006

Re: wine 0.9.9 vs. Ned For Speed V -- already better

by Tom Wickline

On 3/5/06, Stefan Dösinger <stefandoesinger(a)gmx.at> wrote: > Am Sonntag, 5. März 2006 20:36 schrieb Samium Gromoff: > > manages to allow NFS to initialize the video mode, apparently. > You can try my directdraw patches: http://stud4.tuwien.ac.at/~e0526822/ > But wait before downloading, I have some updates I'll upload within the next > hours. > > Is a demo of that game available? Need for Speed V is: Porsche Unleashed The Demo: nfspudemo.zip : http://ftp.gameaholic.com/pub/demos/?N=D http://miningco.gameaholic.com/idgames.d/gamedemos/?offset=14 http://www.sports-gaming.com/downloads/demos/nfspudemo.shtml http://www.gamepro.com/computer/pc/1713/info.shtml http://www.fileshack.com/file.x?fid=701 http://www.gamershell.com/download_4366.shtml http://files.gameaholic.com/idgames.d/gamedemos/ http://www.ausgamers.com/files/demos.php http://www.3dgamers.com/ Tom > > Stefan > > > > > > >

March 5, 2006

Re: Feedback request on wininet patch

by Mike McCormack

Victor Pelt wrote: > - FIXME("Option INTERNET_OPTION_RECEIVE_TIMEOUT: STUB\n"); > + { It would be nice if you could use indentation consistent with the rest of the file, and avoid tabs. > + unsigned long conns=*(unsigned long*)lpBuffer; What's 'conns' for? > + /* have HINTERNET hInternet , handle where to set the options */ > + int result = setsockopt(INET_getSocketHelper(lpwhh), INET_getProtocolHelper(lpwhh), SO_RCVTIMEO, lpBuffer, (int)sizeof(lpBuffer)); > + > + if (result == 0) { > + ret = TRUE; > + } else { > + /* Translate the result in the right error to return */ > + ret = FALSE; > + } > + } How about "ret = (result == 0);"? > @@ -2455,6 +2504,9 @@ BOOL WINAPI InternetSetOptionW(HINTERNET > case INTERNET_OPTION_CONNECT_RETRIES: > FIXME("Option INTERNET_OPTION_CONNECT_RETRIES: STUB\n"); > break; > + case INTERNET_OPTION_CONTEXT_VALUE: > + FIXME("Option INTERNET_OPTION_CONTEXT_VALUE; STUB\n"); > + break; > default: > FIXME("Option %ld STUB\n",dwOption); > INTERNET_SetLastError(ERROR_INVALID_PARAMETER); This chunk should be in a separate patch, since it's unrelated. > +/* Helper functions to get unix backend stuff based in windows handles (LPWININETHANDLEHEADER s) */ > +int INET_getSocketHelper(LPWININETHANDLEHEADER); > +int INET_getProtocolHelper(LPWININETHANDLEHEADER); Why not just make these static, and get rid of these declarations? Mike

March 5, 2006