February 2004 - Wine-devel - list.winehq.org

Re: GetDefaultPrinter A->W [resubmit]
by Alexandre Julliard 28 Feb '04

28 Feb '04

Stefan Leichter <Stefan.Leichter(a)camLine.com> writes: > Hello, > i sent this patch one week ago, but it does not get commited. Anything wrong > with it? The handling of the buffer sizes looks very suspicious. For instance you check for namesize being NULL, but later on assign something to it anyway. Also allocations have to be in bytes but length arguments (for instance in the GetProfileStringW call) have to be in WCHARs, this isn't done correctly. -- Alexandre Julliard julliard(a)winehq.org

2 2

Re: PATCH: Compatibility issue with InternetCrackUrlA
by wire 28 Feb '04

28 Feb '04

Hi. I have modified my patch about compatibility of InternetCrackUrlA function,to deal with the failed case. If the pointer member of one component is set to point a buffer,but the length member is not big enough,then the component should not be returned, and the length member is set to the value needed. And later i'll post the patch for the conformance test. Bye. ChangeLog: Fill the nPort field with the corresponding scheme; Fix the path and extrainfo components' parse; Deal with the failed case. Index: wine/dlls/wininet/internet.c =================================================================== RCS file: /home/wine/wine/dlls/wininet/internet.c,v retrieving revision 1.73 diff -u -r1.73 internet.c --- wine/dlls/wininet/internet.c 30 Dec 2003 19:16:37 -0000 1.73 +++ wine/dlls/wininet/internet.c 28 Feb 2004 10:22:19 -0000 @@ -845,29 +845,28 @@ DWORD nLength; URL_COMPONENTSW UCW; WCHAR* lpwszUrl; + BOOL result; if(dwUrlLength==0) dwUrlLength=strlen(lpszUrl); lpwszUrl=HeapAlloc(GetProcessHeap(), 0, sizeof(WCHAR)*(dwUrlLength+1)); memset(lpwszUrl,0,sizeof(WCHAR)*(dwUrlLength+1)); nLength=MultiByteToWideChar(CP_ACP,0,lpszUrl,dwUrlLength,lpwszUrl,dwUrlLength+1); - memset(&UCW,0,sizeof(UCW)); - if(lpUrlComponents->dwHostNameLength!=0) - UCW.dwHostNameLength=1; - if(lpUrlComponents->dwUserNameLength!=0) - UCW.dwUserNameLength=1; - if(lpUrlComponents->dwPasswordLength!=0) - UCW.dwPasswordLength=1; - if(lpUrlComponents->dwUrlPathLength!=0) - UCW.dwUrlPathLength=1; - if(lpUrlComponents->dwSchemeLength!=0) - UCW.dwSchemeLength=1; - if(lpUrlComponents->dwExtraInfoLength!=0) - UCW.dwExtraInfoLength=1; - if(!InternetCrackUrlW(lpwszUrl,nLength,dwFlags,&UCW)) - { - HeapFree(GetProcessHeap(), 0, lpwszUrl); - return FALSE; - } + + memcpy(&UCW,lpUrlComponents,sizeof(UCW)); + if (UCW.lpszScheme && UCW.dwSchemeLength) + UCW.lpszScheme = HeapAlloc(GetProcessHeap(), 0, sizeof(WCHAR)*(UCW.dwSchemeLength+1)); + if (UCW.lpszHostName && UCW.dwHostNameLength) + UCW.lpszHostName = HeapAlloc(GetProcessHeap(), 0, sizeof(WCHAR)*(UCW.dwHostNameLength+1)); + if (UCW.lpszUserName && UCW.dwUserNameLength) + UCW.lpszUserName = HeapAlloc(GetProcessHeap(), 0, sizeof(WCHAR)*(UCW.dwUserNameLength+1)); + if (UCW.lpszPassword && UCW.dwPasswordLength) + UCW.lpszPassword = HeapAlloc(GetProcessHeap(), 0, sizeof(WCHAR)*(UCW.dwPasswordLength+1)); + if (UCW.lpszUrlPath && UCW.dwUrlPathLength) + UCW.lpszUrlPath = HeapAlloc(GetProcessHeap(), 0, sizeof(WCHAR)*(UCW.dwUrlPathLength+1)); + if (UCW.lpszExtraInfo && UCW.dwExtraInfoLength) + UCW.lpszExtraInfo = HeapAlloc(GetProcessHeap(), 0, sizeof(WCHAR)*(UCW.dwExtraInfoLength+1)); + result = InternetCrackUrlW(lpwszUrl,nLength,dwFlags,&UCW); + ConvertUrlComponentValue(&lpUrlComponents->lpszHostName, &lpUrlComponents->dwHostNameLength, UCW.lpszHostName, UCW.dwHostNameLength, lpszUrl, lpwszUrl); @@ -888,7 +887,19 @@ lpszUrl, lpwszUrl); lpUrlComponents->nScheme=UCW.nScheme; lpUrlComponents->nPort=UCW.nPort; + lpUrlComponents->dwSchemeLength=UCW.dwSchemeLength; + lpUrlComponents->dwHostNameLength=UCW.dwHostNameLength; + lpUrlComponents->dwUserNameLength=UCW.dwUserNameLength; + lpUrlComponents->dwPasswordLength=UCW.dwPasswordLength; + lpUrlComponents->dwUrlPathLength=UCW.dwUrlPathLength; + lpUrlComponents->dwExtraInfoLength=UCW.dwExtraInfoLength; HeapFree(GetProcessHeap(), 0, lpwszUrl); + HeapFree(GetProcessHeap(), 0, UCW.lpszScheme); + HeapFree(GetProcessHeap(), 0, UCW.lpszHostName); + HeapFree(GetProcessHeap(), 0, UCW.lpszUserName); + HeapFree(GetProcessHeap(), 0, UCW.lpszPassword); + HeapFree(GetProcessHeap(), 0, UCW.lpszUrlPath); + HeapFree(GetProcessHeap(), 0, UCW.lpszExtraInfo); TRACE("%s: scheme(%s) host(%s) path(%s) extra(%s)\n", lpszUrl, debugstr_an(lpUrlComponents->lpszScheme,lpUrlComponents->dwSchemeLength), @@ -896,7 +907,7 @@ debugstr_an(lpUrlComponents->lpszUrlPath,lpUrlComponents->dwUrlPathLength), debugstr_an(lpUrlComponents->lpszExtraInfo,lpUrlComponents->dwExtraInfoLength)); - return TRUE; + return result; } /*********************************************************************** @@ -972,10 +983,14 @@ } else { - INT ncpylen = min((*dwComponentLen)-1, len); - strncpyW(*lppszComponent, lpszStart, ncpylen); - (*lppszComponent)[ncpylen] = '\0'; - *dwComponentLen = ncpylen; + if (*dwComponentLen<len+1) { + **lppszComponent = '\0'; + *dwComponentLen = len+1; + return FALSE; + } + strncpyW(*lppszComponent, lpszStart, len); + (*lppszComponent)[len] = '\0'; + *dwComponentLen = len; } } @@ -999,6 +1014,8 @@ LPWSTR lpszcp = NULL; WCHAR lpszSeparators[3]={';','?',0}; WCHAR lpszSlash[2]={'/',0}; + BOOL result = TRUE; + if(dwUrlLength==0) dwUrlLength=strlenW(lpszUrl); @@ -1027,12 +1044,12 @@ /* Parse <params> */ lpszParam = strpbrkW(lpszap, lpszSeparators); - if (lpszParam != NULL) + if (lpszParam != NULL && (lpUC->dwExtraInfoLength || lpUC->lpszExtraInfo)) { if (!SetUrlComponentValueW(&lpUC->lpszExtraInfo, &lpUC->dwExtraInfoLength, lpszParam, dwUrlLength-(lpszParam-lpszUrl))) { - return FALSE; + result = FALSE; } } @@ -1045,7 +1062,7 @@ lpUC->nScheme = GetInternetSchemeW(lpszUrl, lpszcp - lpszUrl); if (!SetUrlComponentValueW(&lpUC->lpszScheme, &lpUC->dwSchemeLength, lpszUrl, lpszcp - lpszUrl)) - return FALSE; + result = FALSE; /* Eat ':' in protocol. */ lpszcp++; @@ -1112,14 +1129,16 @@ lpszcp++; } - SetUrlComponentValueW(&lpUC->lpszUserName, &lpUC->dwUserNameLength, - lpszUser, lpszPasswd - lpszUser); + if (!SetUrlComponentValueW(&lpUC->lpszUserName, &lpUC->dwUserNameLength, + lpszUser, lpszPasswd - lpszUser)) + result = FALSE; if (lpszPasswd != lpszHost) lpszPasswd++; - SetUrlComponentValueW(&lpUC->lpszPassword, &lpUC->dwPasswordLength, + if (!SetUrlComponentValueW(&lpUC->lpszPassword, &lpUC->dwPasswordLength, lpszPasswd == lpszHost ? NULL : lpszPasswd, - lpszHost - lpszPasswd); + lpszHost - lpszPasswd)) + result = FALSE; lpszcp++; /* Advance to beginning of host */ } @@ -1133,8 +1152,9 @@ entire string up to the first '/' */ if(lpUC->nScheme==INTERNET_SCHEME_RES) { - SetUrlComponentValueW(&lpUC->lpszHostName, &lpUC->dwHostNameLength, - lpszHost, lpszPort - lpszHost); + if (!SetUrlComponentValueW(&lpUC->lpszHostName, &lpUC->dwHostNameLength, + lpszHost, lpszPort - lpszHost)) + result = FALSE; lpUC->nPort = 0; lpszcp=lpszNetLoc; } @@ -1158,12 +1178,29 @@ } else { - SetUrlComponentValueW(&lpUC->lpszHostName, &lpUC->dwHostNameLength, - lpszHost, lpszPort - lpszHost); + if (!SetUrlComponentValueW(&lpUC->lpszHostName, &lpUC->dwHostNameLength, + lpszHost, lpszPort - lpszHost)) + result = FALSE; if (lpszPort != lpszNetLoc) lpUC->nPort = atoiW(++lpszPort); else - lpUC->nPort = 0; + switch(lpUC->nScheme) { + case INTERNET_SCHEME_FTP: + lpUC->nPort = INTERNET_DEFAULT_FTP_PORT; + break; + case INTERNET_SCHEME_HTTPS: + lpUC->nPort = INTERNET_DEFAULT_HTTPS_PORT; + break; + case INTERNET_SCHEME_HTTP: + lpUC->nPort = INTERNET_DEFAULT_HTTP_PORT; + break; + case INTERNET_SCHEME_GOPHER: + lpUC->nPort = INTERNET_DEFAULT_GOPHER_PORT; + break; + default: + lpUC->nPort = 0; + break; + } } } } @@ -1198,7 +1235,7 @@ if (!SetUrlComponentValueW(&lpUC->lpszUrlPath, &lpUC->dwUrlPathLength, lpszcp, len)) - return FALSE; + result = FALSE; } else { @@ -1210,7 +1247,7 @@ debugstr_wn(lpUC->lpszUrlPath,lpUC->dwUrlPathLength), debugstr_wn(lpUC->lpszExtraInfo,lpUC->dwExtraInfoLength)); - return TRUE; + return result; } /***********************************************************************

1 0

Re: RPC Update Patch
by Dan Kegel 27 Feb '04

27 Feb '04

> On Thu, 26 Feb 2004 23:16:23 -0600, Gregory M. Turner wrote: >> They are actually pretty well-documented on MSDN: >> >> http://msdn.microsoft.com/library/default.asp?url=/library/en-us/rpc/rpc/rp… >> >> Looks like a lot of work, Robert; thanks for your efforts. > > Hmm, interesting. I wonder why this doesn't show up when I google for "ndr > format strings". It does if you include the quotes... - Dan -- US citizens: if you're considering voting for Bush, look at these first: http://www.misleader.org/ http://www.cbc.ca/news/background/arar/ http://www.house.gov/reform/min/politicsandscience/

1 0

Re: implement fixme in CreateAcceleratorTable
by Dmitry Timoshkov 27 Feb '04

27 Feb '04

"Pabs" <pabs3(a)bonedaddy.net> wrote: > ChangeLog > > Added some checks to CreateAcceleratorTableA/W (removes fixmes) IMO it's better to use if (!MapVirtualKeyW(lpaccel[i].key, 0)) { WARN("Invalid VK\n"); SetLastError(ERROR_INVALID_PARAMETER); return NULL; } for a check whether VK is a valid one. -- Dmitry.

2 1

catching first chance exception question
by Robert Reif 27 Feb '04

27 Feb '04

I'm looking at how windows deals with invalid parameters in some of the winmm functions and it raises some questions. Sometimes they will check the parameters and return an error code and sometimes they will let a bad parameter cause a first-chance exception and catch it and return an error code. Is it ok to do a simple NULL pointer check and return an error rather than letting it cause an exception and catching it? If not, what's the best way to catch the exception and return the proper error code?

2 1

Re: build: avoid Unix file on MinGW
by Lionel Ulmer 27 Feb '04

27 Feb '04

On Fri, Feb 27, 2004 at 09:19:08AM -0500, Dimitrie O. Paun wrote: > BTW, why do we need the {GLU32,GLUT32,OPENGL}FILES cases? > Should those just be listed under X files? And shouldn't > we just build those, and dload() the Unix libs at run time? We need these special cases as an user can have X installed but not GLU / GLUT or OpenGL. And as to why we cannot dload the Unix lib at run time, it would be a bit like 'why do we not build the X11DRV library all the time and dlsym all the X11 symbols'. Basically, if you do not have OpenGL installed, there is no way you will be able to get Wine's OpenGL library to run. So there is no point to offer some special 'dlopen' magic to be able to load it on these boxes. Lionel -- Lionel Ulmer - http://www.bbrox.org/

1 0

Re: winemaker: use winegcc
by Francois Gouget 27 Feb '04

27 Feb '04

On Fri, 27 Feb 2004, Dimitrie O. Paun wrote: [...] > @@ -801,7 +753,7 @@ > $basename=~ s/\.(dll|exe)$//i; > # This is the default link list of Visual Studio, except odbccp32 > # which we don't have in Wine. > - my @std_imports=qw(advapi32 comdlg32 gdi32 kernel32 odbc32 ole32 oleaut32 shell32 user32 winspool); > + my @std_imports=qw(odbc32 ole32 oleaut32 winspool); > my @std_libraries=qw(uuid); Does winegcc systematically link applications and dlls to advapi32, comdlg32, etc? How does one do if one does not want to link with gdi32 for instance? -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ May your Tongue stick to the Roof of your Mouth with the Force of a Thousand Caramels.

2 1

Re: Help needed to get the application work for native linux editors
by Robert Shearman 27 Feb '04

27 Feb '04

Hi, I can't really give you any advice on using the X API, but I can tell you that the Win32 and X API are completely different. If you want to support native X apps then you will have to either completely rewrite your Windows program in Linux or you will have to compile the Linux editor as a Winelib app and run it under Wine. I suggest you find a good book on the X API and/or contact the appropriate mailing lists to find out which functions you should use. Rob >>> "saravanan" <saravananv(a)erdcitvm.org> 02/26/04 11:17 AM >>> Hai friends, I successfully ran a script manager(keyboard Manager/driver) developed for windows (using VC++) in Linux using wine. Its working in the wordpad application called using wine.I want to make it work in Linux editors. The technical details about the working of the keyboard driver is as follows. 1) The keyboard driver is a keyboard hooker program. The keyboard driver will trap all the keystrokes from the keyboard and convert the Ascii value to the appropriate native language Character code. 2) First we have to get the handle of the window/control where the cursor is currently located, for that the GetFocus API is used. 3) Virtual-key code of the key that generated the keystroke message is converted to appropriate another code for making the keyboard Inscript or phonetic and send using Postmessage API. I have some doubts. Do these APIs written for windows work in linux. In linux , How could i get the handle of the window where the cursor is currently located. What are the changes that i should make in the source code to make it work for linux editors. Looking for your suggessions. Thanks in advance . Saravanan ______________________________________ Scanned and protected by Email scanner

4 3

RPC Update Patch
by Robert Shearman 27 Feb '04

27 Feb '04

Hi, I have uploaded the big stubless RPC patch to a website as it is rather large (~120k). You can get it here: http://www.dcs.warwick.ac.uk/~csvmc/rpcrt4_3.diff As you can probably see, a lot of the stuff in this patch is a bit rough around the edges (lack of copyright notice in some files), but it does seem to work for a lot of cases. Rob P.S. It doesn't work for native OLE32 yet because we need some undocumented interfaces in RPCSS.

3 4

Re: Reset Resolution on Crash (resend)
by Alexandre Julliard 26 Feb '04

26 Feb '04

Robert Shearman <R.J.Shearman(a)warwick.ac.uk> writes: > No comment to this patch last time it was sent. Is it better implemented > some other way? I don't think user32 should be messing with the exception filter. Maybe we could make winedbg do this? -- Alexandre Julliard julliard(a)winehq.org

1 0