May 2010 - Wine-devel - list.winehq.org

Re: d3dx9_36: Implement ID3DXLine_GetDevice with tests
by testbot＠testbot.winehq.org 24 May '10

24 May '10

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=2276 Your paranoid android. === WINEBUILD (build) === Patch failed

1 0

software built and worked under wine but not in vista?
by Hin-Tak Leung 24 May '10

24 May '10

I have encountered a slightly strange issue with building the windows self-extracting installer release of ghostscript (http://www.ghostscript.com/) - the self-extracting installer is created with the winzip self-extractor package, which can be manually unzipped. The curious thing is that I can build the self-extracting installer under 32-bit wine with the win 7 SDK (on 64-bit linux). but when I tried to run the self-installer under Vista, it says %1 is not a valid win32 application. Manually unpacking the selft-extracting installer and run 'setupgs.exe' (the setup program) also gives 'setupgs.ese is not a valid win32 application' message. But the main ghostcript executable seems operational. (the setup program mostly only just copy the extracted files to "Program Files" and set up a few registry entries, but ghostscript doesn't not require those to work). So it seems that setupgs.exe is mis-compiled under wine with win7 sdk but just happened to also work under wine. Any idea how it might happen? I have checked things like end-of-line, but it is curious why the setupgs program works under wine but not in vista, the reverse situation compared to most.

5 11

too much dynamic loading?
by Marcus Meissner 24 May '10

24 May '10

Hi, As libjpeg and libpng bumped their major versions in the last months I had to adjust some of my library requires in my wine.spec file. This caused me thinking if it is really necessary to dynamically load nearly every library. Can we make some of those direct linking? libjpeg and libpng and some other lowlevel libraries might be useful. Opinions? Ciao, Marcus

6 10

Re: comctl32/listview: Fix vertical position offset for subitem rectangle
by testbot＠testbot.winehq.org 23 May '10

23 May '10

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=2274 Your paranoid android. === W98SE (32 bit listview) === listview.c:2214: Test failed: got 1 === WNT4WSSP6 (32 bit listview) === listview.c:2214: Test failed: got 1 === W2KPROSP4 (32 bit listview) === listview.c:2214: Test failed: got 1 === WXPPROSP3 (32 bit listview) === listview.c:2214: Test failed: got 1 === W2K3R2SESP2 (32 bit listview) === listview.c:2214: Test failed: got 1 === WVISTAADM (32 bit listview) === listview.c:2214: Test failed: got 1 === W2K8SE (32 bit listview) === listview.c:2214: Test failed: got 1 === W7PRO (32 bit listview) === listview.c:2214: Test failed: got 1 === W7PROX64 (32 bit listview) === listview.c:2214: Test failed: got 1 === W7PROX64 (64 bit listview) === listview.c:2214: Test failed: got 1

1 0

Re: [1/2] shell32: Export lowlevel PIDL manipulation calls by name too
by testbot＠testbot.winehq.org 23 May '10

23 May '10

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=2273 Your paranoid android. === WNT4WSSP6 (32 bit shlfileop) === shlfileop.c:2202: Test failed: SHGetNewLinkInfoA succeeded

1 0

Re: [PATCH] shlwapi: initial implement of SHAutoComplete
by Nikolay Sivov 23 May '10

23 May '10

On 5/23/2010 05:09, Hirofumi Katayama wrote: > See attachment. > Hi, some comments here. > +/****************************************************************************/ > + > +static const WCHAR szMRUList[] = L"MRUList"; > +static const WCHAR szRunMRU[] = > + L"Software\\Microsoft\\Windows\\CurrentVersion\\Explorer\\RunMRU"; > + > Don't use this L"" specifier for strings, use array initialization style instead. > +void *my_malloc(UINT size) > +{ > + return HeapAlloc(GetProcessHeap(), 0, size); > +} > + > +void *my_realloc(void *p, UINT size) > +{ > + if (p == NULL) > + return my_malloc(size); > + return HeapReAlloc(GetProcessHeap(), 0, p, size); > +} > + > +void my_free(void *p) > +{ > + HeapFree(GetProcessHeap(), 0, p); > +} > + > This isn't very useful, I think it's not a pain to use direct calls for this file. > +#define malloc my_malloc > +#define realloc my_realloc > +#define free my_free > +#define _wcsdup my__wcsdup > Don't do that. > + > +STDMETHODIMP IEnumString_fnQueryInterface(IEnumString *pEnumString, REFIID riid, void** ppvObject); > +STDMETHODIMP_(ULONG) IEnumString_fnAddRef(IEnumString *pEnumString); > +STDMETHODIMP_(ULONG) IEnumString_fnRelease(IEnumString *pEnumString); > +STDMETHODIMP IEnumString_fnNext(IEnumString *pEnumString, ULONG celt, LPOLESTR *rgelt, ULONG *pceltFetched); > +STDMETHODIMP IEnumString_fnSkip(IEnumString *pEnumString, ULONG celt); > +STDMETHODIMP IEnumString_fnReset(IEnumString *pEnumString); > +STDMETHODIMP IEnumString_fnClone(IEnumString *pEnumString, IEnumString **ppenum); > STDMETHODIMP isn't used in Wine code, use explicit WINAPI please. Also forward declaration for methods isn't used, usually a vtable is initialized just after all methods bodies. > +typedef struct > +{ > + IEnumStringVtbl *lpVtbl; > + LONG m_lRef; > + IMalloc * m_pMalloc; > + INT m_i; > + INT m_c; > + LPWSTR * m_strings; > +} EnumStringImpl; > Please avoid this MFC style like m_* for members, and use more descriptive names. > +EnumStringImpl *IEnumString_construct(VOID) > +{ > + EnumStringImpl *pEnumString; > + > + pEnumString = (EnumStringImpl *) malloc(sizeof(EnumStringImpl)); > + if (pEnumString != NULL) > + { > + pEnumString->lpVtbl =&EnumStringVtbl; > + pEnumString->m_lRef = 0; > + pEnumString->m_pMalloc = NULL; > + SHGetMalloc(&pEnumString->m_pMalloc); > + pEnumString->m_i = 0; > + pEnumString->m_strings = GetMRUStrings(&pEnumString->m_c); > + } > + return pEnumString; > +} > This is wrong. Initial refcount should be 1. > +EnumStringImpl *IEnumString_construct(VOID) > > + > +VOID IEnumString_destruct(EnumStringImpl *pEnumString) > Should be static both, and methods too of course. > + pEnumString->lpVtbl->AddRef(pEnumString); > No need for that, use defined macros to call methods from vtables. > +STDMETHODIMP_(ULONG) IEnumString_fnAddRef(IEnumString *pEnumString) > +{ > + EnumStringImpl *this = (EnumStringImpl *) pEnumString; > + this->m_lRef++; > + return (ULONG) this->m_lRef; > +} > You should use interlocked increment here (same for _Release()). > + pProp->fnOldWndProc = (WNDPROC) SetWindowLongPtr(hwndEdit, GWLP_WNDPROC, (LONG_PTR) AutoCompleteEditWndProc); > Did you ever build it? I think we have a blocking things for using implicit A/W calls.

1 0

make "bisected" a keyword in bugzilla?
by Wolfram Sang 23 May '10

23 May '10

Hi, I noticed that Wylda uses "--private keyword: bisected" when appropriate. IMHO this could be useful as a real keyword, e.g. if you search for bugs you'd like to try tackling. Has this been considered already? Regards, Wolfram

8 10

Re: kernel32/tests: check return value of alarmThread in test_NamedPipe_2() (try 2)
by Andrew Nguyen 22 May '10

22 May '10

On Sat, May 22, 2010 at 5:02 AM, Austin English <austinenglish(a)gmail.com> wrote: > Changes from try 1: > Fix another typo. > Set LastError before checking it > Print LastError as a decimal > Don't check for INVALID_HANDLE_VALUE as a return value of > CreateThread, instead make sure serverThread is not NULL. > > -- > -Austin > > > > You're calling SetLastError(0xdeadbeef) at the wrong time in the pipe server subtests. If CreateThread fails, you will unconditionally print the decimal representation of 0xdeadbeef in the failure messages. SetLastError has to be called before CreateThread to ensure that the failure messages print a last error that the preceding CreateThread call sets.

2 1

Re: [5/5] winemenubuilder: remove legacy PNG and XPM conversion code
by Vincent Povirk 22 May '10

22 May '10

Fools! With this patch, the WIC empire shall finally be complete. Er, I mean, this is pretty cool. And thanks for the ICO decoder fix. On Sat, May 22, 2010 at 11:01 AM, Damjan Jovanovic <damjan.jov(a)gmail.com> wrote: > Changelog: > * winemenubuilder: remove legacy PNG and XPM conversion code > > The patch may look frighteningly long but the majority is just > harmless deletions :-). > > Damjan Jovanovic > > > >

1 0

Re: crypt32: use ret in compareStore() (try 2)
by Juan Lang 22 May '10

22 May '10

Hi Austin, + SetLastError(0xdeadbeef); ret = CertSaveStore(store, X509_ASN_ENCODING, CERT_STORE_SAVE_AS_STORE, CERT_STORE_SAVE_TO_MEMORY, &blob, 0); + todo_wine ok(ret && GetLastError() == CRYPT_E_NOT_FOUND, "CertSaveStore failed. Expected CRYPT_E_NOT_FOUND, got %d\n", GetLastError()); In general we don't check last error when a function succeeds. I think an ok(ret, ...) would be fine here. Also, I think it's neater to print crypt errors as %08x, as the errors are generally 0x8009.... Thanks for fixing this, I was just sloppy not to have checked ret in the first place! --Juan

1 0