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

Re: [QUARTZ] Add implementation for IFilterMapper interface (RESEND)

by Christian Costa

Tom wrote: > Christian Costa wrote: > >> Hi, >> >> This time wih Rob's suggestions. >> >> Bye. >> >> Changelog : >> Implemented IFilterMapper and IEnumRegFilters interfaces. >> Fixed IFilterMapper2_EnumMatchingFilters. >> >> Christian Costa titan.costa(a)wanadoo.fr >> > > Hi Christian, > > Should I add you as a worker on quartz? > > Tom > > > Hi Tom, Yes, you can. :-) Bye, Christian

Feb. 29, 2004

Re: [QUARTZ] Add implementation for IFilterMapper interface (RESEND)

by Tom

Christian Costa wrote: > Hi, > > This time wih Rob's suggestions. > > Bye. > > Changelog : > Implemented IFilterMapper and IEnumRegFilters interfaces. > Fixed IFilterMapper2_EnumMatchingFilters. > > Christian Costa titan.costa(a)wanadoo.fr > Hi Christian, Should I add you as a worker on quartz? Tom

Feb. 29, 2004

Re: help

by Tom

Matt Jaggard wrote: > > hi, I left a message, a longgg time ago on your website, > I was wondering if you could please delete mjaggard(a)blueyonder.co.uk > <mailto:mjaggard(a)blueyonder.co.uk>. As I do not want spammers picking my > address up. Thanks, > The URL my address is on is called. > http://www.winehq.org/hypermail/wine-users/2000/12/0438.html. if you > could delete references to mjaggard(a)blueyonder.co.uk > <mailto:mjaggard(a)blueyonder.co.uk> or the page that would be brill > > Many Thanks > > M Jaggard This is a joke or what? Now you real addy with out the ( _ ) in it is on wine-devel not only by your post but with my reply as well... Your odds of being spammed have just increased by some 200% ... But don't worry spam is really good for you..... :) My dear sweet Grandma use to keep us nice and fat on it when we were just little tikes.. Tom

Feb. 29, 2004

Borland C++ 5.5 fails on Link, any ideas?

by Michael Pyne

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi all, I've been trying to use the free Borland C++ 5.5 command-line tools to compile some Windows applications for my school assignments. To be specific, I am in a graphics class where every turnin has to be Windows. :-( Anyways, Borland C++ runs fine until it tries to link the program in question, at which point ilink32.exe just sits there and eats CPU cycles. I've tried letting it run in case it was simply slower than normal due to the Wine layer, but that doesn't seem to be case (I've let it sit for minutes at a time). This occurs no matter how I run it (with wine or either of wineconsole's two modes). I was wondering if there is a way I can help figure out what the problem is so that we can get this bug fixed. Obligatory specs: OS: Gentoo w/ Linux 2.6.3 (Wine was not compiled using NPTL, but OpenGL support is in) ATI OpenGL drivers XFree86 4.3 GCC 3.3.1 Using fake_windows support Please CC: any replies to me, as I am not subscribed to this list. Also, let me know if there is any other information which would be useful. Regards, - Michael Pyne -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (GNU/Linux) iD8DBQFAQVAQqjQYp5Omm0oRAofOAKDEhWcFavrlg/3VdjMOcx90hF/iiwCgvWUj IS/FYImIKOxThQRj89LvwYc= =+Gxw -----END PGP SIGNATURE-----

Feb. 29, 2004

Re: [QUARTZ] Add implementation for IFilterMapper interface

by Christian Costa

Robert Shearman wrote: >Christian Costa wrote: > > >>Hi, >> >>This patch adds implementation for the IFilterMapper interface. >> >> > >Nice. I have few comments (see below). > > > >>However filters registered with this interface require some work in >>devenum to >>be seen from application. This will be done when time permits. >> >> > >Ok. I had assumed devenum was complete, but obviously not. Does it spit out >any fixme's? > > No, it doesn't. In fact, Devenum enumerates entries in: 1) HROOT\CLSID\{"clsid of categories"]\Instance 2) HCURRENTUSER\Software\Microsoft\ActiveMovie\devenum For Direct Show filters, Devenum retreives filters registered with IFilterMapper and recreates entries in ....\devenum so when enumerating filters we get the IFilterMapper2 filters followed by IFilterMapper ones. I think this is the same for ICM and ACM filters. The current code only do 1 or 2 but not both at the same time. > >... > > > >>@@ -879,7 +893,7 @@ >> if (SUCCEEDED(hrSub)) >> hrSub = >>ICreateDevEnum_CreateClassEnumerator(pCreateDevEnum, &clsidCat, >>&pEnum, 0); >> >>- if (SUCCEEDED(hrSub)) >>+ if (SUCCEEDED(hrSub) && (hrSub == S_OK)) >> >> > >You can remove the SUCCEEDED(hrSub) completely here. > >... > He! Right! :-) > > > >>+ if (SUCCEEDED(hrSub)) >>+ { >>+ len = (strlenW((WCHAR*)&V_UNION(&var, bstrVal))+1) * >>sizeof(WCHAR); >>+ if (!(regfilters[idx].Name = CoTaskMemAlloc(len*2))) >>+ hr = E_OUTOFMEMORY; >>+ } >>+ >>+ if (SUCCEEDED(hrSub)) >>+ { >>+ memcpy(regfilters[idx].Name, &V_UNION(&var, bstrVal), len); >>+ regfilters[idx].Clsid = clsid; >>+ idx++; >>+ } >> >> > >Indentation messed up a bit here and many other places. > > Ok! I will fix that. >... > > >>+ if (cFetched > 0) >>+ { >>+ for(i = 0; i < cFetched; i++) { >>+ /* The string in the REGFILTER structure must be >>allocated in the same block as the REGFILTER structure itself */ >>+ ppRegFilter[i] = >>(REGFILTER*)CoTaskMemAlloc(sizeof(REGFILTER)+(strlenW(This->RegFil >>ters[i].Name)+1)*sizeof(WCHAR)); >>+ if (!ppRegFilter[i]) { >>+ while(i) { >>+ CoTaskMemFree(ppRegFilter[--i]); >>+ ppRegFilter[i] = NULL; >>+ } >>+ return E_OUTOFMEMORY; >>+ } >>+ ppRegFilter[i]->Clsid = This->RegFilters[i].Clsid; >>+ ppRegFilter[i]->Name = >>(WCHAR*)((char*)ppRegFilter[i]+sizeof(REGFILTER)); >>+ CopyMemory(ppRegFilter[i]->Name, >>This->RegFilters[i].Name, >>(strlenW(This->RegFilters[i].Name)+1)*sizeof(WCHAR)); >>+ } >> >> > >Please choose one style of bracket placement and stick to it (this occurs in >a few places too). > I tried to stick to your style. It seems I forgot some brackets. :-) > >Other than these things, it looks fine. > >Rob > > > > Fine. Thanks Rob, Christian

Feb. 28, 2004

Re: GetDefaultPrinter A->W [resubmit]

by Alexandre Julliard

Stefan Leichter <Stefan.Leichter(a)camLine.com> writes: > GetDefaultPrinterW breaks directly when namesize is a NULL pointer and > GetDefaultPrinterA jumps to the end without. Where do is this pointer > accessed when it is not vaild? You are right, I missed the fact that the W function would fail which would cause the A one to bail too. Still it is very confusing, because in the A function you compute the size and allocate a buffer even though it's not going to be used. You should restructure this to make it clear what is happening. -- Alexandre Julliard julliard(a)winehq.org

Feb. 28, 2004

Re: [QUARTZ] Add implementation for IFilterMapper interface

by Robert Shearman

Christian Costa wrote: > > Hi, > > This patch adds implementation for the IFilterMapper interface. Nice. I have few comments (see below). > However filters registered with this interface require some work in > devenum to > be seen from application. This will be done when time permits. Ok. I had assumed devenum was complete, but obviously not. Does it spit out any fixme's? ... > @@ -879,7 +893,7 @@ > if (SUCCEEDED(hrSub)) > hrSub = > ICreateDevEnum_CreateClassEnumerator(pCreateDevEnum, &clsidCat, > &pEnum, 0); > > - if (SUCCEEDED(hrSub)) > + if (SUCCEEDED(hrSub) && (hrSub == S_OK)) You can remove the SUCCEEDED(hrSub) completely here. ... > + if (SUCCEEDED(hrSub)) > + { > + len = (strlenW((WCHAR*)&V_UNION(&var, bstrVal))+1) * > sizeof(WCHAR); > + if (!(regfilters[idx].Name = CoTaskMemAlloc(len*2))) > + hr = E_OUTOFMEMORY; > + } > + > + if (SUCCEEDED(hrSub)) > + { > + memcpy(regfilters[idx].Name, &V_UNION(&var, bstrVal), len); > + regfilters[idx].Clsid = clsid; > + idx++; > + } Indentation messed up a bit here and many other places. ... > + if (cFetched > 0) > + { > + for(i = 0; i < cFetched; i++) { > + /* The string in the REGFILTER structure must be > allocated in the same block as the REGFILTER structure itself */ > + ppRegFilter[i] = > (REGFILTER*)CoTaskMemAlloc(sizeof(REGFILTER)+(strlenW(This->RegFil > ters[i].Name)+1)*sizeof(WCHAR)); > + if (!ppRegFilter[i]) { > + while(i) { > + CoTaskMemFree(ppRegFilter[--i]); > + ppRegFilter[i] = NULL; > + } > + return E_OUTOFMEMORY; > + } > + ppRegFilter[i]->Clsid = This->RegFilters[i].Clsid; > + ppRegFilter[i]->Name = > (WCHAR*)((char*)ppRegFilter[i]+sizeof(REGFILTER)); > + CopyMemory(ppRegFilter[i]->Name, > This->RegFilters[i].Name, > (strlenW(This->RegFilters[i].Name)+1)*sizeof(WCHAR)); > + } Please choose one style of bracket placement and stick to it (this occurs in a few places too). Other than these things, it looks fine. Rob

Feb. 28, 2004

Re: PATCH: Compatibility issue with InternetCrackUrlA

by wire

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; } /***********************************************************************

Feb. 28, 2004

Re: GetDefaultPrinter A->W [resubmit]

by Stefan Leichter

Am Freitag, 27. Februar 2004 01:23 schrieb Alexandre Julliard: > 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. You are right with the last point, but i don't understand what you mean with the first one. GetDefaultPrinterW breaks directly when namesize is a NULL pointer and GetDefaultPrinterA jumps to the end without. Where do is this pointer accessed when it is not vaild? Bye Stefan

Feb. 28, 2004

Re: RPC Update Patch

by Dan Kegel

> 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/

Feb. 28, 2004