Re: dead stores in netapi32
On Wednesday 19 November 2008 00:10:52 Juan Lang wrote: Does that netbios code actually work? IIRC you need to bind to port 137 to do netbios requests. That's not strictly true, you can be a client without binding to port 137. You just can't advertise any services. I haven't tried it myself in a long, long time, though I did hear from a couple people that were using it successfully a few years ago. Oh, ok. I've only ever seen 137 in traces, but I guess that makes sense on Windows, where browsing is a system service anyway. Thanks for pointing that out, Juan. Ricardo, I'm currently looking at the netapi32 files you pointed out. Thanks for the catch. Cheers, Kai -- Kai Blin WorldForge developer http://www.worldforge.org/ Wine developerhttp://wiki.winehq.org/KaiBlin Samba team member http://www.samba.org/samba/team/ -- Will code for cotton. signature.asc Description: This is a digitally signed message part.
Re: Fedora 9 and MingW
On Tuesday 18 November 2008 22:48:11 Paul Vriens wrote: https://fcp.surfsite.org/modules/newbb/viewtopic.php?topic_id=61801forum=11 I've installed these rpm's but the executables use a different naming convention: i686-pc-mingw32* instead of what we check for: i686-mingw32*. Should we add this to our configure? Looks promising! If Fedora accepts it under that name it's certainly worthwhile to add it to configure. -Hans
RE: netapi32: Actually return a couple of return values.
it seems you two fixed the same, juan sent a patch in a previous wine-devel email. you might want to check that because you two have different solutions for the first problem, line 415. From: [EMAIL PROTECTED] Date: Wed, 19 Nov 2008 12:41:50 +0100 Subject: netapi32: Actually return a couple of return values. To: [EMAIL PROTECTED] ricardo filipe [EMAIL PROTECTED] found a couple of dead return value assignments in nbt.c and netbios.c Fix up these error paths. --- dlls/netapi32/nbt.c | 12 +++- dlls/netapi32/netbios.c |2 +- 2 files changed, 12 insertions(+), 2 deletions(-) _ Receba GRÁTIS as mensagens do Messenger no seu celular quando você estiver offline. Conheça o MSN Mobile! http://mobile.live.com/signup/signup2.aspx?lc=pt-br
Regarding ./configure file
Sir/Madam, I need to add few functions related to libwnck. It requires some header files to be included. I included them in wine source code. But when compiled it says header files not found. I haven't updated configure file. How can i update it? please tell me what r the changes that i need to made? Regards, Sainath.
Re: dlls/oleaut32: add tests for OleLoadPictureFile(Ex)
On Tue, 18 Nov 2008, Michael Karcher wrote: Looks OK to me, but what happened to your finding about INET_E_RESOURCE_NOT_FOUND? Was it a red herring? I saw that on XP, but I couldn't reproduce it on Windows 2000. I need to investigate this further. Is there a mechansim to do windows-version-specific tests? I ended up having to add OleLoadPicutreFile and OleLoadPictureFileEx to the .spec and .c files as actual functions (for now just returning E_NOTIMPL) in order to get the test to even compile. Might be that they don't appear in the ELF export table otherwise, and as we don't crossbuild the tests, the imports are done by the ELF linker/loader. BTW: In general, if you add stubs like this, please add a FIXME call showing the parameters and saying stub. How do I show the VARIANT in the FIXME call? This fragment repeats over and over, sometimes with PICTYPE_BITMAP instead. Might be worth refactoring that into a function. todo_wine works for function boundaries. If you refactor, you should pass that function __LINE__ information and print it in your OK calls to know which of the tests is failing if something fails. You do a lot of tests like this. Setting disp to (void*)0xdeadbeef instead before allows you to differentiate between disp unmodified and disp explicitly set to zero. Good ideas. I'll do that.
Re: dlls/msxml3: check some dead store results (llvm/clang)
ricardo filipe wrote: only return value checking in this patch. _ Confira vídeos com notícias do NY Times, gols direto do Lance, videocassetadas e muito mais no MSN Video! http://video.msn.com/?mkt=pt-br You are mixing tabs and spaces. -- Cheers, Paul.
RE: Regarding ./configure file
Don't modify ./configure directly, edit configure.ac instead. The autoconf documentation describes how this file works. configure is then created from configure.ac by running autoconf From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of shanmukha sainath addepalli Sent: Tuesday, November 18, 2008 10:27 AM To: wine-devel@winehq.org Subject: Regarding ./configure file Sir/Madam, I need to add few functions related to libwnck. It requires some header files to be included. I included them in wine source code. But when compiled it says header files not found. I haven't updated configure file. How can i update it? please tell me what r the changes that i need to made? Regards, Sainath.
spec file syntax for VARIANT parameter
In order to implement OleLoadPictureFile and OleLoadPictureFileEx, it will be necessary to change their entries in oleaut32.spec from stub to stdcall. The stdcall syntax wants a list of parameter types, which puts these functions in an unususal situation: they take a VARIANT (which is a struct, in case anyone is not familiar) by value as their first parameter. There does not seem to be any type defined in the spec file that fits this situation. I have been using ptr in my last couple of patches which add stub implementations for these functions, but when I tried to build this on cygwin I got errors like: undefined reference to [EMAIL PROTECTED]' undefined reference to [EMAIL PROTECTED]' Looking at the functions that are present in the .o file, I see they are really [EMAIL PROTECTED] and [EMAIL PROTECTED] Changing the spec file to list each VARIANT as 4 longs fixes these errors. My question is, should my next version of the patch list the VARIANT as 4 longs, or is there some better way to specify this that I'm not aware of? If not, is there a comment syntax for the spec file where I can explain why there are 4 long params where there should be one VARIANT? Thanks Jeremy
Re: spec file syntax for VARIANT parameter
On Wed, Nov 19, 2008 at 01:11:05PM -0800, Jeremy Drake wrote: In order to implement OleLoadPictureFile and OleLoadPictureFileEx, it will be necessary to change their entries in oleaut32.spec from stub to stdcall. The stdcall syntax wants a list of parameter types, which puts these functions in an unususal situation: they take a VARIANT (which is a struct, in case anyone is not familiar) by value as their first parameter. There does not seem to be any type defined in the spec file that fits this situation. I have been using ptr in my last couple of patches which add stub implementations for these functions, but when I tried to build this on cygwin I got errors like: undefined reference to [EMAIL PROTECTED]' undefined reference to [EMAIL PROTECTED]' Hmm, use as much long specifiers to fit the VARIANT struct (number of bytes/4). I think VARIANT has 16 byte, so use long long long long. My question is, should my next version of the patch list the VARIANT as 4 longs, or is there some better way to specify this that I'm not aware of? If not, is there a comment syntax for the spec file where I can explain why there are 4 long params where there should be one VARIANT? Currently no other way. Ciao, Marcus
Re: Regarding ./configure file
I've looked into configure and configure.ac and I've seen that configure is quite more recent than configure.ac and it seems to have been changed manually. Am I wrong ? Max Stefan Dösinger ha scritto: Don't modify ./configure directly, edit configure.ac instead. The autoconf documentation describes how this file works. configure is then created from configure.ac by running autoconf *From:* [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] *On Behalf Of *shanmukha sainath addepalli *Sent:* Tuesday, November 18, 2008 10:27 AM *To:* wine-devel@winehq.org *Subject:* Regarding ./configure file Sir/Madam, I need to add few functions related to libwnck. It requires some header files to be included. I included them in wine source code. But when compiled it says header files not found. I haven't updated configure file. How can i update it? please tell me what r the changes that i need to made? Regards, Sainath.
Re: dlls/ole32: fixed dead assignments (llvm/clang)
2008/11/18 ricardo filipe [EMAIL PROTECTED]:
@@ -1951,6 +1951,12 @@ HRESULT WINAPI CoRegisterClassObject(
newClass-pMarshaledData,
flags
(REGCLS_MULTIPLEUSE|REGCLS_MULTI_SEPARATE),
newClass-RpcRegistration);
+ MSHLFLAGS_TABLESTRONG);
This won't compile.
+ if (hr) {
+ FIXME(RPC_StartLocalServer failed, %x!\n,hr);
You should change this to:
ERR(RPC_StartLocalServer failed with error 0x%x\n,hr);
+ return hr;
+ }
+
}
return S_OK;
}
diff --git a/dlls/ole32/ole16.c b/dlls/ole32/ole16.c
index a8fa59b..21fb1fb 100644
--- a/dlls/ole32/ole16.c
+++ b/dlls/ole32/ole16.c
@@ -598,13 +598,13 @@ HRESULT WINAPI CLSIDFromProgID16(LPCOLESTR16 progid,
LPCLSID riid)
buf = HeapAlloc(GetProcessHeap(),0,strlen(progid)+8);
sprintf(buf,%s\\CLSID,progid);
- if ((err=RegOpenKeyA(HKEY_CLASSES_ROOT,buf,xhkey))) {
+ if (RegOpenKeyA(HKEY_CLASSES_ROOT,buf,xhkey)) {
Since you are changing this anyway, you should explicitly compare the
result against ERROR_SUCCESS to improve readability of the code here.
HeapFree(GetProcessHeap(),0,buf);
return CO_E_CLASSSTRING;
}
HeapFree(GetProcessHeap(),0,buf);
buf2len = sizeof(buf2);
- if ((err=RegQueryValueA(xhkey,NULL,buf2,buf2len))) {
+ if (RegQueryValueA(xhkey,NULL,buf2,buf2len)) {
Ditto.
RegCloseKey(xhkey);
return CO_E_CLASSSTRING;
}
--
Rob Shearman
Re: dlls/rpcrt4: fixed dead increments (llvm/clang)
2008/11/18 ricardo filipe [EMAIL PROTECTED]: diff --git a/dlls/rpcrt4/rpc_transport.c b/dlls/rpcrt4/rpc_transport.c index 3d66740..815ed64 100644 --- a/dlls/rpcrt4/rpc_transport.c +++ b/dlls/rpcrt4/rpc_transport.c @@ -479,7 +479,6 @@ static size_t rpcrt4_ncacn_np_get_top_of_tower(unsigned char *tower_data, memcpy(tower_data, networkaddr, networkaddr_size); else tower_data[0] = 0; -tower_data += networkaddr_size; return size; } @@ -692,7 +691,6 @@ static size_t rpcrt4_ncalrpc_get_top_of_tower(unsigned char *tower_data, pipe_floor-count_rhs = endpoint_size; memcpy(tower_data, endpoint, endpoint_size); -tower_data += endpoint_size; return size; } These increments are intentional in case the code is copy and pasted and some more data is added to the end, so please don't remove them. -- Rob Shearman
Re: dlls/rpcrt4: fixed dead increments (llvm/clang)
Rob Shearman wrote: 2008/11/18 ricardo filipe [EMAIL PROTECTED]: diff --git a/dlls/rpcrt4/rpc_transport.c b/dlls/rpcrt4/rpc_transport.c index 3d66740..815ed64 100644 --- a/dlls/rpcrt4/rpc_transport.c +++ b/dlls/rpcrt4/rpc_transport.c @@ -479,7 +479,6 @@ static size_t rpcrt4_ncacn_np_get_top_of_tower(unsigned char *tower_data, memcpy(tower_data, networkaddr, networkaddr_size); else tower_data[0] = 0; -tower_data += networkaddr_size; return size; } @@ -692,7 +691,6 @@ static size_t rpcrt4_ncalrpc_get_top_of_tower(unsigned char *tower_data, pipe_floor-count_rhs = endpoint_size; memcpy(tower_data, endpoint, endpoint_size); -tower_data += endpoint_size; return size; } These increments are intentional in case the code is copy and pasted and some more data is added to the end, so please don't remove them. But those are dead code in the actual context and thus distract the reader. Shouldn't those be made comments instead? bye michael
