September 2003 - Wine-Devel

Re: Latest CVS doesnt compile
by Rolf Kalbermatter Sept. 7, 2003

Sept. 7, 2003

Jon Brandenburg <jbrandenburg(a)yahoo.com> wrote: >On Sat, 2003-09-06 at 16:36, Ian Goldby wrote: >> >> and occurs 2-3 hours into the build. Thanks for any further suggestions. > >Speaking of compile times.. what are the average/min/max times for >compiling the complete system? I was surprise when it took my system 12 >minutes to do it do a full cvs 'make depend && make'. Now after seeing >this message I have much to be grateful for. On my Athlon 2000+ with standard Suse 8.2 AMD Athlon kernel 2.4.19, I get about 15 minutes or so for everything from ./configure until make, after a make clean. As to the original problem. I had some to. Turned out to be missing include statements in not yet committed source files after Fridays header include cleanup. Rolf Kalbermatter

1 0

Papers on legal aspects of reverse engineering
by Dan Kegel Sept. 7, 2003

Sept. 7, 2003

Somebody was asking about what is allowed with respect to using Microsoft SDKs when writing Wine code. I have no answer for his question, but did spend a few minutes digging up interesting bookmarks on reverse engineering in general. Maybe they'll be of some interest. http://reengineer.org/ http://www.chillingeffects.org/reverse/faq.cgi http://www.digital-law-online.com/lpdi1.0/treatise25.html http://www.digital-law-online.com/lpdi1.0/treatise48.html http://www.sims.berkeley.edu/~pam/papers/l&e%20reveng5.pdf http://www.law.cornell.edu/commentary/intelpro/chn95int.htm http://www.lawplusplus.com/column10.htm http://www.lgu.com/publications/softcopy/12.shtml http://research.sun.com/people/cristina/legal-computing-publications.html - Dan -- Dan Kegel http://www.kegel.com http://counter.li.org/cgi-bin/runscript/display-person.cgi?user=78045

1 0

Porting radio driver from MS to Linux
by Arjen Verweij Sept. 7, 2003

Sept. 7, 2003

Hi, I am looking to port a MSVC++ driver to Linux. The recommended approach, to convert the Makefile and compile it didn't work for me, so I am trying to compile seperate cpp.files by implementing the winapi functions I need in a seperate linuxapi.cpp There are about 30 functions I have to port at the moment, and some of them are easier than others. Basically they are part of the DLL (so), thread/event/mutex, File*Find apis, with some miscellaneous ones. I think I could do myself a favor by asking if the MSVC++ code could use libpthread instead of the MS threads, so I'm not too worried about those, and simulating DLL with the dl* api seems easy enough as well. Still, there are problems with Find*File, FindClose, GetModuleFileName and so on. I'm not too familiar with winapi myself, because I learn C/C++ in a college environment, where most programming assignments are geared towards Linux. Can anyone comment sensibly on this? If you are of the opinion that I should just rewrite the app from scratch, I agree in part, but the assignment is to leave as much of the original code as it is, apart from isolated cases where g++ can't quite follow the MSVC++ compiler. At the end of the email is a list of prototypes included, that I ripped from winbase.h Looking forward to read your thoughts on this matter, Regards, Arjen Verweij Here is the list: DWORD GetModuleFileName(HMODULE, LPTSTR, DWORD); HANDLE FindFirstFile(LPCSTR lpFileName, LPWIN32_FIND_DATA lpFindFileData); BOOL FindNextFile(HANDLE hFindFile, LPWIN32_FIND_DATA lpFindFileData); BOOL FindClose(HANDLE hFindFile); BOOL SetEvent(HANDLE hEvent); FARPROC GetProcAddress(HMODULE hModule, LPCSTR lpProcName); HMODULE LoadLibrary(LPCSTR lpLibFileName); BOOL FreeLibrary(HMODULE hLibModule); HANDLE CreateMutex(LPSECURITY_ATTRIBUTES lpMutexAttributes, BOOL bInitialOwner, LPCSTR lpName); DWORD WaitForMultipleObjects(DWORD nCount, CONST HANDLE *lpHandles, BOOL bWaitAll, DWORD dwMilliseconds); BOOL CloseHandle(HANDLE hObject); BOOL ReleaseMutex(HANDLE hMutex); DWORD GetCurrentProcessId(VOID); HANDLE CreateEvent(LPSECURITY_ATTRIBUTES lpEventAttributes, BOOL bManualReset, BOOL bInitialState, LPCSTR lpName); VOID Sleep(DWORD dwMilliseconds); BOOL ResetEvent(HANDLE hEvent); DWORD _beginthread(void(__cdecl *start_address) (void*), unsigned, void*); VOID OutputDebugString(LPCSTR lpOutputString); DWORD WaitForSingleObject(HANDLE hHandle, DWORD dwMilliseconds); void _endthread(void); BOOL ReleaseSemaphore(HANDLE hSemaphore, LONG lReleaseCount, LPLONG lpPreviousCount); HANDLE CreateSemaphore(LPSECURITY_ATTRIBUTES lpSemaphoreAttributes, LONG lInitialCount, LONG lMaximumCount, LPCSTR lpName); BOOL UnmapViewOfFile(LPCVOID lpBaseAddress); HANDLE CreateFileMapping(HANDLE hFile, LPSECURITY_ATTRIBUTES lpFileMappingAttributes, DWORD flProtect, DWORD dwMaximumSizeHigh, DWORD dwMaximumSizeLow, LPCSTR lpName); DWORD GetLastError(VOID); LPVOID MapViewOfFile(HANDLE hFileMappingObject, DWORD dwDesiredAccess, DWORD dwFileOffsetHigh, DWORD dwFileOffsetLow, DWORD dwNumberOfBytesToMap);

4 3

Dynamically altering drive mappings at runtime
by Mike Hearn Sept. 7, 2003

Sept. 7, 2003

.... and on a slightly related note, I am wondering how hard it would be to allow wine to alter drive mappings at runtime. The reason I ask is that it'd be nice for the "add drive" dialog in winecfg to use the shell directory chooser that shows all directories on the system rather than forcing the user to type in a path manually, which is error prone. To do that, of course we need a few things: * the ability to map / if it's not already mapped (or maybe we should just force / to always be mapped to drive Z in the wine code) * the ability to switch off drive letters showing through in the directory chooser dialog, so it's less confusing to see this magic letter appear. Thoughts on how hard this is? thanks -mike

3 2

Using fontconfig
by Mike Hearn Sept. 7, 2003

Sept. 7, 2003

Hi to anybody who knows about fonts, It'd be nice to use fontconfig in future to locate font installation paths. There is a simple API: FcStrList FcConfigGetFontDirs (FcConfig *config); which should let us use a small part of it, even if we don't use all of it. It means we don't need to make the user manually configure font paths. Does anybody know how hard this would be? thanks -mike

2 1

Problems with libartsc.so
by jeff_latimer Sept. 7, 2003

Sept. 7, 2003

I have just upgraded to RedHat 9 and when I make wine it can't link in libartsc.so . Looking in /usr/lib there is a libartsc.so.0 not libartsc.so . Is this normal? Jeff ---------------- Powered by telstra.com

2 1

Re: Display detection bug
by Ivan Leo Murray-Smith Sept. 6, 2003

Sept. 6, 2003

> The 'Linux community' of the start where all the people gave something back > (useful and advanced bug reports, fixes, help, ...) is being slowly killed > by people who choose Linux just because it's free (as in beer) and who do > not give back anything. Thats because most people don't want to give anything to computers at all, most people just use them as tools. And the best way to avoid hundreds of silly questions is to make software more user friendly. So maybe wine mime type for gnome and KDE would be nice in 1.0, and again that's the more users, better software thing.

1 0

Re: Display detection bug
by Ivan Leo Murray-Smith Sept. 6, 2003

Sept. 6, 2003

>Well, it's more a 'good old days' feelings than a real theory. Ie the >biggest problem I see with 'Linux on everybody's Desktop' is that it's >damaging for the community (due to the fact that the developper / user ratio >is getting more and more skewed to the user side whereas 'in the golden >days' anyone using Linux was a developper). More users means more feedback, better software, more users, more developers, more software, more feedback...

3 2

Re: [winecfg 5] Implement transactions support,remove some crack x11drv prefs,complete desktop sizing code
by Ivan Leo Murray-Smith Sept. 6, 2003

Sept. 6, 2003

.The UI > standard on free software desktops for instant apply dialogs is to have > just a Close dialog, with a Revert button in cases where the user might > not be able to undo the changes themselves. For the main window, just a > Close button is enough. That would be inconsistent with KDE,that uses apply and OK buttons, and cancel to not apply and changes and exit. The Mandrake Control Center also works with OK buttons.

2 1

Wine fun projects - explorer clone update
by Steven Edwards Sept. 6, 2003

Sept. 6, 2003

Hello All, I doubt this will ever make it in to WINE as Martin has been using C++ to implement our explorer clone but you might want to have a look at it if you want a explorer clone for WINE. http://www.sky.franken.de/explorer/index.html You will want to get the current source from ReactOS CVS. Martin has made a lot of changes to support both WINE and ReactOS and been working on improving WINEs shell32. Thanks Steven __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com

1 0

DGA input events dropped
by Thomas J. Moore Sept. 6, 2003

Sept. 6, 2003

For a very long time now, enabling DGA causes mouse/keyboard events to be ignored. There have been a few threads on this issue in the past, and it seems nobody is willing to actually fix this problem (even though some have submitted patches to fix it, nothing has ever made it into CVS). Why is this? Overall, the problem seems to be related to the use of separate per-thread X display handles, along with a global (separate) "GDI" display. At first I tried just making the first thread's display the "global" display, but since that didn't work, and I don't have the patience to figure out why this dichotomy exists, here's a simple patch which at least works for a few games I tried (but still has mouse problems with a few other games I've tried - there was a mouse fix on an earlier thread(1) on this subject, but it doesn't work -- perhaps it has something to do with the fact that DGA only returns relative mouse movement, and non-DGA returns absolute movement). On the other hand, you could just at least make your default behavior and sample config not enable DGA. Note that this patch is against wine-20030813, but I did check CVS first and it still applies. Also, I wasn't sure about the wine-patches message policy, so I'm just sending this to wine-devel. (1) http://www.winehq.com/hypermail/wine-devel/2002/06/0427.html ---------------------------------------------------------------------------- Thomas J. Moore, Hacker/SysAdmin | TV & video games have destroyed my mind inet: dark(a)mama.indstate.edu | unix ada asm c/c++ mcu ee amiga tired ----------------------------------------------------------------------------

2 1

Re: LPT0 support with DOSFS_OpenDevice
by Mike McCormack Sept. 6, 2003

Sept. 6, 2003

Hi Ed, If the code is essentially the same anyway, it might be a good idea just to add a switch in server/serial.c and to the create_serial request to tell the server whether the device is a serial port or a parallel port, then change serial/parallel behaviour depending on that switch. You can also use serial_get_info to return a flag saying that you're dealing with a parallel port if necessary. It's probably better to avoid duplication of the code unless necessary. Mike Ed Lea wrote: > This patch provides initial support for accessing LPT0 via /dev/lp0. It > provides this support in very much the same way serial port support is > provided - most of the functions are subtly altered duplicates of that > code. Currently only syncronous accesses are supported, although the > infra-structure for asyncronous methods is in place it has not been > tested. It would be a good starting point for anyone to get this working > at a later date though. > > The patch enables programs that communicate with parallel port devices > directly through LPT0 to work. > > Ed Lea <ed(a)orangebits.cx>

1 0

RE: TreeView & WM_NOTIFY
by Igor Grahek Sept. 5, 2003

Sept. 5, 2003

It looks like comctl32.dll is seting up some hooks to forward messages to requested window because even if you use --dll comctl32=n the owner is set up to "GetAncestor( cs->hwndParent, GA_ROOT )" Same thing happends under Windows so it seems that everything is OK with windows/win.c I tried to figure out what is going on with hooks but I don't have much experience with windows hooks so I will appreciate any help. Thanks, Igor > -----Original Message----- > From: Igor Grahek [mailto:igorg(a)cadlink.com] > Sent: September 5, 2003 11:29 AM > To: 'wine-devel(a)winehq.org' > Subject: RE: TreeView & WM_NOTIFY > > > Here is an example in attachment (Visual C++ project) . > If you click on "tst" button it opens TreeView control with > WS_POPUP and Cview as parent (to process WM_NOTIFY messages) > With current code in windows/win.c owner is set to > "GetAncestor( cs->hwndParent, GA_ROOT )" and WM_NOTIFY goes > to wrong window. If I change code so that owner is > "WIN_GetFullHandle(cs->hwndParent)" then there is problem > with Z ordering (keeping popup window on top) but WM_NOTIFY > goes to Cview as it should. > > However if I use --dll comctl32=n and owner is set to > "GetAncestor( cs->hwndParent, GA_ROOT )" everything works OK > - popup is on top and CView receives WM_NOTIFY from it. > > I know I'm repeating myself but I'm not sure where I should > look for the problem: is it in dll/comctl32/treeview.c (as > --dll comctl32=n suggesting) or is it in windows/win.c and > dll/x11drv/* ? > > Igor > > > -----Original Message----- > > From: Igor Grahek [mailto:igorg(a)cadlink.com] > > Sent: September 4, 2003 1:50 PM > > To: 'Dimitrie O. Paun' > > Cc: 'wine-devel(a)winehq.org' > > Subject: RE: TreeView & WM_NOTIFY > > > > > > I think I found the source of my problem: > > ------------- > > Wine/windows/win.c ::CreateWindowEx > > > > if ((cs->style & (WS_CHILD|WS_POPUP)) == WS_CHILD) > > parent = WIN_GetFullHandle(cs->hwndParent); > > else > > owner = GetAncestor( cs->hwndParent, GA_ROOT ); > > -------------------------- > > Why owner is not set to "WIN_GetFullHandle(cs->hwndParent)" ? > > > > If I change this everything works OK except there is a > > problem with Z ordering (it is hiding popup window behind its > > owner when Managed=N) Is the source of problem somewhere in > > wine/dll/x11drv/winpos.c (I'm currently examining > > SWP_DoOwnedPopups function) and why it is working with --dll > > comctl32=n - no problems with Z order there?! > > > > BTW I'm working with 20030618 build. > > > > Igor > > > > > -----Original Message----- > > > From: Dimitrie O. Paun [mailto:dimi(a)intelliware.ca] > > > Sent: September 2, 2003 6:42 PM > > > To: Igor Grahek > > > Cc: wine-devel(a)winehq.org > > > Subject: Re: TreeView & WM_NOTIFY > > > > > > > > > On Tue, 2 Sep 2003, Igor Grahek wrote: > > > > > > > My problem is very similar to bug 1227 > > > > (http://bugs.winehq.org/show_bug.cgi?id=1227) > > > > Treeview in application is sending WM_NOTIFY on double > > > click (NM_DBLCLK) > > > > to the wrong > > > > Window handle. GetParent() function returns handle to the > > > root window > > > > and root window is not forwarding/broadcasting that message > > > and message > > > > is lost. Everything works fine under Windows > > > > or with --dll comctl32=n > > > > > > I'm afraid you'll need to debug this and post a lot more > > detail, maybe > > > a code fragment showing the undesirable behaviour. > > > > > > -- > > > Dimi. > > > > > > > > >

1 0

Help required regarding compilation of windows project in Linux via WineLib
by Anjan Sarkar Sept. 5, 2003

Sept. 5, 2003

Hello, I am using wine for my personal interest and need your help regarding the following problem I have faced. I have following configuration, a) Red Had Linux 9.0 b) Wine (wine-20030813.tar.gz) is extracted in /ww/wine-20030813 c) It creates a directory /c for windows and /root/.wine d) All the bin files are at /usr/local/bin Wine runs fine in my machine. Now I try to complie windows project in Linux via WineLib. I have made the following things, copy all the .c, .h, .rc files from programs/winemine dir to programs/winemine2 then from winemine2 run the following commands, $ winemaker --lower-uppercase . $ ./configure --with-wine=/usr/local/wine It gives me the error, "checking for windef.h configure:error couldnot find Wine headers(windef.h) I have also copy the file from /usr/local/include/wine/windows to current directory. Still the problem exists. Kindly suggest me what I have to do? Again please excuse me for this silly question. Thanking you, Anjan Sarkar Need a new email address that people can remember Check out the new EudoraMail at http://www.eudoramail.com

4 3

Display detection bug
by hatky Sept. 5, 2003

Sept. 5, 2003

Some games are failing at detecting a 3D card or sometimes the screen dept, have a look at bug : http://bugs.winehq.org/show_bug.cgi?id=1696 getting into another bugzila/testing prenzy, hatky. __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com

5 6

Re: Get rid of W->A calls
by Rolf Kalbermatter Sept. 5, 2003

Sept. 5, 2003

hatky(a)users.sourceforge.net wrote: > Ok, one last line: > > strcpy(((char*)lpwh) + sizeof(WINHELP), lpHelpFile); I'm also not sure about the context here and it seems a little strange but what you would want to do here is probably: strcpyW((LPWSTR)((char*)lpwh + sizeof(WINHELP)), lpHelpFile); Rolf Kalbermatter

1 0

Re: Get rid of W->A calls
by hatky Sept. 5, 2003

Sept. 5, 2003

> Let us know if you need any help. Just to be sure I got it right, I moved the DirectDrawEnumerateExA stuff into DirectDrawEnumerateExW since there is nothing ascii or unicode specific in it I think it's ok on it's own now in DirectDrawEnumerateExA I have these variables: LPDDENUMCALLBACKEXA lpCallback pointer to a callback function, the DirectDrawEnumerateExW get a LPDDENUMCALLBACKEXW (not match a suprise I guess) so do I need to convert it? just cast it as (DirectDrawEnumerateExW)? something else? LPVOID lpContext Address of an application-defined context to be passed to the enumeration callback function each time that it is called when they called from W -> A they did: DirectDrawEnumerateProcData epd; epd.lpCallback = (LPVOID) lpCallback; epd.lpContext = lpContext; and sent in as: (LPVOID) &epd Should it just be sent as is? DWORD dwFlags I assume it should be passed as is, for some reason the opposite call sent 0 Hatky. __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com

3 11

Re: [winecfg 5] Implement transactions support, remove some crack x11drv prefs, complete desktop sizing code
by Dimitrie O. Paun Sept. 5, 2003

Sept. 5, 2003

On 5 Sep 2003, Mike Hearn wrote: > This patch will add some warnings, and yeah, the commenting out of lots > of code is ugly, but don't worry, it's all temporary until I've finished > bringing the code across from the old global config struct. At that > point it'll all be deleted in favour of the transaction system. Cool. I should put you up as a worker on this task... > The transactions system is very simple, at the moment the default is > instant apply, despite the presence of OK and Cancel buttons. That's a > bit misleading, but the PropertySheet() API doesn't seem to have any way > to express the need for "Close" and maybe "Revert" buttons. The UI > standard on free software desktops for instant apply dialogs is to have > just a Close dialog, with a Revert button in cases where the user might > not be able to undo the changes themselves. For the main window, just a > Close button is enough. Would it be OK to implement a Wine extension > here Alexandre? Maybe we should just commit on OK, but that will be inconsistent with Gnome. I guess that at some point we need some extensions, one for the edit control to work with selections would be nice :) > ChangeLog: > - Remove "PerfectGraphics", "Use XSHM extension", and "Use a private > colour map", people who know what these actually are can probably use > regedit (we'll need default reg entries, don't let me forget!) Yeah, we should create a table with all the options, and keep track of wich ones were review, default values, static/dynamic behaviour, etc. I said I'd do it, but I didn't get around to it yet... -- Dimi.

1 0

WIDL problems
by Rolf Kalbermatter Sept. 5, 2003

Sept. 5, 2003

I'm trying to get a new shell object into Wine and seeing that most of the object definition stuff has recently been moved to IDL files I wondered if I could define the object in shobjidl.idl. However I seem to get some problems compiling shobjidl.idl with with widl. I tried ./tools/widl/widl -I ./include ./include/shobjidl.idl and get an error. To be sure it wasn't caused by my editing I made sure the entire include dir is an unmodified version of the CVS tree and tried it there and I still get the error ./include/shobjidl.idl:426: Error: method index mismatch in write_proxy What is the problem here? Should I just sent in the patch to shobjidl.idl and the header file gets generated automatically with some more capable tool? By the way I noticed also that IPersistFolder3 got probably overlooked when moving the definitions to idl files. Rolf Kalbermatter

2 1

RE: Get rid of W->A calls
by Robert Shearman Sept. 5, 2003

Sept. 5, 2003

> -----Original Message----- > From: wine-devel-admin(a)winehq.org [mailto:wine-devel-admin(a)winehq.org]On > Behalf Of Dimitrie O. Paun > Sent: 05 September 2003 17:40 > To: hatky(a)users.sourceforge.net > Cc: wine-devel(a)winehq.org > Subject: Re: Get rid of W->A calls > > > why did they call them wcommand in WinHelpA and > > command in WinHelpW ?? (msdn calls them uCommand but > > the code uses wcommand...) > > I don't see why, it should be named the same in > both versions if it's encoding independent. I think the confusing thing is the 'w' prefix. In this case 'w' stands for WORD, i.e. a two-byte integer. Also, MSDN isn't clear as to whether it is encoding independent but I would assume so. > > -hDest = FindWindowA("MS_WINHELP", NULL); > > +hDest = FindWindowW("MS_WINHELP", NULL); > > You can't to that directly, as we don't support > Unicode literals. You need to do that by hand: > > WCHAR clsName[] = { 'M', 'S', '_', 'W', 'I', 'N', 'H', 'E', > 'L', 'P', 0 }; > hDest = FindWindowW(clsName, NULL); > This would be better: static const WCHAR clsName[]... It allows for better optimization by the compiler (assuming that the compiler can't work out that clsName will not be modified) and shows the purpose better to the next person who looks at it. Rob

1 0

Looking for guidance..
by Jon Brandenburg Sept. 5, 2003

Sept. 5, 2003

Hi, I'm interested in doing some development/help for wine and was wondering in what way I could help out? I have a college level understanding of C/C++/Java, have done some professional programming in (ack, VB), am currently in the process of learning C#, but would really like to do some development in unix/linux and I figure this is a good way to get some experience. I'm willing to do whatever needs to be done, but I do have as I'm sure most of you have, other projects/assignments/obligations which I'm sure the rest of you have also. If you could give me a bit of direction on an area of the project that needs some attention and perhaps a little bit of hand holding to begin with I'd like to help out. Thanks! Jon

2 1

WineHQ
by Ivan Leo Murray-Smith Sept. 5, 2003

Sept. 5, 2003

Hi list, A guy that lives in my town found me via the linux counter, and we would like to translate the wine web site to italian, I was thinking of something like microsoft, with a drop menu of languages, where one can choose the destination language. This guy is also a really good web designer, and I could give him a hand as I'm a native english speaker and I could answer any emails we may get on an italian users mailing list, can we do this? Would a flag be better than a drop menu? any ideas/sugestions welcome.

2 1

Re: Display detection bug
by hatky Sept. 5, 2003

Sept. 5, 2003

> What about rather a coding / debugging frenzy ? I considered them both but they are very hard to do for me for a known reason, Documentation. I read them all + a lot of the mailing lists but still do not know where to start with all of this even in the debugging, getting the trace of a certain component is not that easy because there is no place listing them I need to guess if it's +reg or +registry for example, there are places where it's match harder (ex display stuff).... About coding it's even harder since I do not know a lot about linux and in programming for linux for a first, the list of tasks is updated very slowly and usually not very clear to outsiders, the Janitorial list that is suppose to be the easiest is not very clear about what to do and is full of junk already done, not updated very frequently and also has stuff that the gurus here are working months on them (like dll separation) so how does a Janitorial suppose to do that? Don't get me wrong, I am not trying to say anything bad about you or the project or the rest of the wonderful people working on it, but as I see it the reason others programmers do not join it or at least I don't is becouse you do not make this task easy at all, if anyone would want to turn this around and try to pass some of this knowledge I am willing to learn and document (I am not a writer and my primary language is not English...) just to get the view point of a new programmer (ok I HAVE been watching this project for about 4 years but did not get active as a programmer). Ya, this is not a constant signature, sincerely yours, Hatky. __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com

4 3

Get rid of W->A calls
by hatky Sept. 5, 2003

Sept. 5, 2003

Janitorial reporting ;-) In the Get rid of W->A calls task, Say I try dlls/ddraw/main.c: ddraw: DirectDrawEnumerateExW: illegal call to DirectDrawEnumerateExA The misbheaving function is : HRESULT WINAPI DirectDrawEnumerateExW( LPDDENUMCALLBACKEXW lpCallback, LPVOID lpContext, DWORD dwFlags) { DirectDrawEnumerateProcData epd; epd.lpCallback = (LPVOID) lpCallback; epd.lpContext = lpContext; return DirectDrawEnumerateExA(DirectDrawEnumerateExProcW, (LPVOID) &epd, 0); } What sould I do? 1. copy the entire DirectDrawEnumerateExA into it 2. create a 3rd function DirectDrawEnumerateEx that both of them will call 3. somthing else (you tell me) Hatky. __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com

3 2

Case sensitive in wine
by Fabian Cenedese Sept. 5, 2003

Sept. 5, 2003

Hi I just sent an email about a file that isn't found -> Current Working Dir. I don't know if my next problem is related to it, just mentioning it. I tried another app which needs dlls and again I had the error that the file was not found, this time from wine. But the dll sure was in the fake c:\windows\system32. Just as a wild guess I made a copy of this dll and changed the case to all upper case as was the name wine told me. And miraculously the dll was found. So it seems that even while doing "Windows work" wine sometimes behaves case sensitive. FYI: Linux is Suse 8.0, wine is 20030813 (which makes me wonder as I made clean/distclean the old one, cvs up wine, configure, make depend, make, make install... and I never had a wine-rpm) bye Fabi

2 2