- Wine-Devel - list.winehq.org

Re: PATCH: server/named_pipe.c (RESENT)
by Alexandre Julliard Oct. 28, 2003

Oct. 28, 2003

Gerald Pfeifer <gerald(a)pfeifer.com> writes: > was there any problem with the following patch, or did it just fall > through the cracks? ;-) > > Gerald > > ChangeLog: > Remove unused variable pipe_client_fd_ops. Well, either the client side should really use pipe_client_fd_ops, or pipe_server_fd_ops should be renamed to show that it is used by both client and server. I'm not sure which one was intended, could the named pipe experts please check that? -- Alexandre Julliard julliard(a)winehq.com

1 0

Support for TLS in CrossOver
by Mike Hearn Oct. 28, 2003

Oct. 28, 2003

Hi Alexandre, I noticed the ChangeLog for the new CXOffice release (congrats on that btw!) said it now works with all glibc 2.3 variants. Does that mean the TLS related named pipe breakage we saw earlier is now fixed? If so, do you know when the fix will be in WineHQ CVS? thanks -mike

3 3

WineHQ: PATCH to fix IPersistFile::Load() and associated functions
by Subhobroto Sinha Oct. 28, 2003

Oct. 28, 2003

Hello all I finally produced the "official patch" after verifying that everything is OK. This patch has been inlined as well as attached as a ZIP file, so suit yourself. All reports regaring this patch is to be sent to <subhobrotosinha at yahoo.com> with a proper subject prefixed by "IPersistFile patch" in the subject header. Just don't simply hit the 'Reply' button ! This patch fixes: (1) Stream_LoadString: Conflict of unicode strings being taken as LPSTR and AGAIN being converted to Unicode due to wrong value of SCF_UNICODE .(I have really NOT changed the original code, but redirected this call to a fixed function which I named wineStream_LoadString. This should remain so until SCF_UNICODE value is corrected) (2) Stream_LoadLocation: Previously the GetPath() failed. It works great now. For the uninitiated, here's what has to be done: Please make sure that you have the 20031016 source (latest tarball) (1)Copy this .diff to the root folder of the Wine source (2)Run 'patch' on this diff file (3)It should patch "dlls/shell32/shelllink.c". Change to that directory (4)Type 'make' to rebuild 'shell32.dll.so' (5)'su' out and overwrite/symlink the current 'shell32.dll.so' over the installed one Now IPersistFile::Load() should work as intended. Try out McCormack's 'winemenubuilder' or my WineLib 'linkresolve' (NOT my C++ version) If the diff is not proper, etc, please alert me at the mentioned address PS: As I am not a member of 'wine-patches',can the moderator be kind enough... :=) I really want these functions to get fixed. That's all. Regards Subhobroto Sinha __________________________________ Do you Yahoo!? Exclusive Video Premiere - Britney Spears http://launch.yahoo.com/promos/britneyspears/ --- dlls/shell32/shelllink.c.original 2003-10-22 17:56:31.000000000 +0530 +++ dlls/shell32/shelllink.c 2003-10-26 16:30:57.000000000 +0530 @@ -401,9 +401,49 @@ return S_OK; } +static HRESULT wineStream_LoadString( IStream* stm, LPWSTR *pstr ) +{ + DWORD count=0; + USHORT len; + LPWSTR str=NULL; + HRESULT r; + /* + From Win98 upwards unicode==TRUE.(Atleast for shortcuts..) + However, the value of SCF_UNICODE does NOT seem to be 0x1000. + Until then, we assume it true. + Anybody differing mail to <subhobrotosinha at yahoo.com>. + */ + TRACE("%p\n", stm); + + r = IStream_Read(stm, &len, sizeof(len), &count); + if(FAILED(r)||(count != sizeof(len))) return E_FAIL; + + len *= sizeof(WCHAR); + + TRACE("reading %d\n", len); + str = HeapAlloc(GetProcessHeap(), 0, len+sizeof(WCHAR)); + if(!str) return E_OUTOFMEMORY; + + count = 0; + r = IStream_Read(stm, str, len, &count); + if( FAILED(r)||( count!=len) ) + { + HeapFree( GetProcessHeap(), 0, str ); + return E_FAIL; + } + + str[count/2]=0; + *pstr = str; + TRACE("read %s\n", debugstr_w(str)); + return S_OK; +} static HRESULT Stream_LoadString( IStream* stm, BOOL unicode, LPWSTR *pstr ) { +/*This is the original function short circuited to the above till we find out correct 'SCF_UNICODE'. +Now it's assumed that the data's unicode. (This assumption failes iff the OS was <=Win95*/ +return (wineStream_LoadString(stm,pstr)); +#if 0 DWORD count; USHORT len; LPVOID temp; @@ -453,6 +493,7 @@ *pstr = str; return S_OK; +#endif } static HRESULT Stream_LoadLocation( IStream* stm ) @@ -533,6 +574,14 @@ if( FAILED( r ) ) return r; } + ////////////////////////////////////////////////////////////////////////////////////// + char szTemp[MAX_PATH]={0}; + SHGetPathFromIDListA(This->pPidl,szTemp); + This->sPath=HeapAlloc( GetProcessHeap(), 0,(strlen(szTemp)+1)*sizeof(WCHAR)); + MultiByteToWideChar(CP_ACP,0,szTemp,(strlen(szTemp)+1),This->sPath,(strlen(szTemp)+1)); + //TRACE("%s\n",szTemp); + ////////////////////////////////////////////////////////////////////////////////////// + This->wHotKey = hdr.wHotKey; This->iIcoNdx = hdr.nIcon; FileTimeToSystemTime (&hdr.Time1, &This->time1);

3 3

Winesetuptk 0.7 released
by Ivan Leo Murray-Smith Oct. 28, 2003

Oct. 28, 2003

Hi lists, Winesetuptk 0.7 is at the wine sourceforge page http://sourceforge.net/project/showfiles.php?group_id=6241 This new version contains an updated winedefault.reg, you don't need to run regedit winedefault.reg any more, and it's updated to configurate all the latest and greatest features of wine. This release was possible thanks to the help and contributions of Vincent Béron and Alex Pasadyn. Without them, winesetuptk 0.7 wouldn't exist. Download and enjoy!

1 0

Status of HeapAlloc Cleanup in shell32
by Steven Edwards Oct. 28, 2003

Oct. 28, 2003

Is the HeapAlloc/ReAlloc stuff being tracked anywhere? I think it is the cause of a lot of problems I have with shell32 under Windows and ReactOS. Applications linked to wine shell32.dll under Windows2000 run fine but when you exit they get illigal op's where the memory could not be "read" kind of like the bug I had in uninstaller.exe that started the HeapAlloc cleanup. Thanks Steven __________________________________ Do you Yahoo!? Exclusive Video Premiere - Britney Spears http://launch.yahoo.com/promos/britneyspears/

2 1

Re: wine/ tools/winebuild/relay.c dlls/ntdll/signa ...
by Robert Shearman Oct. 28, 2003

Oct. 28, 2003

This has fixed the problem slightly (see message titled "Bug in cxx_frame_handler") - the program now displays: err:seh:setup_exception stack overflow 592 bytes in thread 000e eip 401c425a esp 40640db0 stack 0x40540000-0x40750000 And then exits. I still believe there is a problem in dlls/msvcrt/cppexcept.c with the following lines (359-364): if (!exc_type) /* nested exception, fetch info from original exception */ { rec = msvcrt_get_thread_data()->exc_record; exc_type = (cxx_exception_type *)rec->ExceptionInformation[2]; } The problem is that msvcrt_get_thead_data()->exc_record returns NULL (probably because there is no nested exception?). Should there just be a check there for a NULL pointer or is it part of a bigger problem? Rob On Tuesday 28 October 2003 12:30 am, you wrote: > ChangeSet ID: 9891 > CVSROOT: /opt/cvs-commit > Module name: wine > Changes by: julliard(a)wine.codeweavers.com 2003/10/27 18:30:55 > > Modified files: > tools/winebuild: relay.c > dlls/ntdll : signal_i386.c > > Log message: > Switch back to the main process stack before calling exception > handlers. > > Patch: http://cvs.winehq.com/patch.py?id=9891 > > Old revision New revision Changes Path > 1.31 1.32 +27 -28 wine/tools/winebuild/relay.c > 1.73 1.74 +297 -165 wine/dlls/ntdll/signal_i386.c

1 0

Improper Focus handling
by Cogman Oct. 28, 2003

Oct. 28, 2003

I have noticed this bug in wine for some time, (winex has corrected it btw). It seems that new windows, child windows, and message boxes are incorrectly handled in wine, when created they are often put at the bottem of the stack of wine windows and not given focus. This is bad expecialy in Install shield programs that bring the messages and windows up one after the other, but you cant see the others or move them because the first is covering them. Also Keyboard focus does not seem to be working properly all the time (sometimes a window gets keyboardfocus and sometimes it doesnt) One thing I would like to point out is the fact that if you click on a window in windows it gets both keyboard and mouse focus. A few examples of where the focus issue becomes a problem are in Cstike and Pharaoh. In Cstrike (or any HL game for that matter) if you loose focus on the main window (alt-tab or something else happens) you loose keyboardfocus, but not always mouse focus, So you endup stuck in a game with the keyboard focused on some other window (usually on a konsole so killall is the only way to get out other then changing virtual windows) In pharaoh the problem is when a new window pops up. The keyboard does not send messages to it, and you can click on it and everything but not be able to type anything (save game window). Now if you run games in desktop mode it will sometimes work. but I dont think that you should have to run games in desktop just to get proper focus handling. I really would love to see this bug smushed :)

1 0

wine msvcrt on Library initialization
by Boaz Harrosh Oct. 28, 2003

Oct. 28, 2003

I have only managed to compile and link MFC with the msvcrt.dll the std way would not work. Now the MFC has lots of: <somefoo.cpp> void CleanFoo() { } int init = atexit( CleanFoo ) ; </somefoo.cpp> The call to atexit will in-turn call _lock if we look at _lock.c : <lock.c> ... /* If the lock doesn't exist yet, create it */ if( lock_table[ locknum ].bInit == FALSE ) { /* Lock while we're changing the lock table */ _lock( _LOCKTAB_LOCK ); .... </lock.c> Since we are calling on library initialization, (global variable) it might happen that the call is made before the initialization of the _LOCKTAB_LOCK (before a call to lock.c::msvcrt_init_mt_locks) and therefor I get an endless recursion. Now this is why we have the -wrap option with the delayed load of winelib apps. [Q}Should I submit a patch to protect that endless recursion in msvcrt. or should it be noted that all DLL s used by the winelib-application should be also linked to the wrapper (to have a chance to initialize) and submit a patch to winemaker? P.S Some similar problem occurs with native msvcrt.dll. That one is hard to pinpoint exactly. In any way making the wrapper link to msvcrt will solve the problem in both cases. Free Life Boaz

1 0

Proposal to resolve the issue of the broken wine-users gateway
by Jeremy White Oct. 28, 2003

Oct. 28, 2003

Hi Folks, Say, Just chatting on IRC today, and ShadowHawk pointed out that a host of other sites (Franks corner, etc) have sprung up almost entirely because there are no user based forums on winehq.org. This is, in part, due to the fact that the gateway between the newsgroup and wine-users is busted. And it is that way, because the gateway we were using is a dead project, and Jer hasn't been able to wrestle it back into working (and doesn't really want to). Further, I think it may be time to recognize that cool looking forums are what users expect; that mailing lists and newsgroups are really becoming the 'old' way of doing things. Further, there is a PHP based forum system that Jer would be happy to put up on winehq.org. This would neatly side step the whole problem and perhaps provide a nice upgrade for winehq.org. What do others think? Cheers, Jeremy

5 6

Re: DMUSIC: fix in dmloader's behavior
by Jerry Jenkins Oct. 28, 2003

Oct. 28, 2003

Rok Mandeljc wrote: > + /* OK, MSDN says that search order is the following: > + - current directory (DONE) > + - windows search path (FIXME: how do I get that?) Try this : DWORD GetEnvironmentVariable( LPCTSTR lpName, // environment variable name LPTSTR lpBuffer, // buffer for variable value DWORD nSize // size of buffer ); or NTSTATUS WINAPI RtlQueryEnvironmentVariable_U( PWSTR env, PUNICODE_STRING name, PUNICODE_STRING value ); > + - loader's search path (DONE) > + */

2 1