October 2013 - Wine-devel - list.winehq.org

Re: mscoree: Add support for ICLRMetaHostPolicy interface
by Dmitry Timoshkov 13 Oct '13

13 Oct '13

Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> wrote: > +static HRESULT WINAPI metahostpolicy_QueryInterface(ICLRMetaHostPolicy *iface, REFIID riid, void **ppvObject) > +{ > + TRACE("%s %p\n", debugstr_guid(riid), ppvObject); > + > + if ( IsEqualGUID( riid, &IID_ICLRMetaHostPolicy ) || > + IsEqualGUID( riid, &IID_IUnknown ) ) > + { > + *ppvObject = iface; > + } > + else > + { > + FIXME("Unsupported interface %s\n", debugstr_guid(riid)); > + return E_NOINTERFACE; > + } > + > + ICLRMetaHostPolicy_AddRef( iface ); > + > + return S_OK; > +} I think that a common COM convention is to set the object pointer to NULL in case of an unsupported interface. Also you're doing pretty good job in avoiding hungarian notation except for ppvObject. Another nitpick is the choice for if/else construct (although I understand that other places also use this sub-optimal one). A more elegant way of implementing QueryInterface would be IMHO: static HRESULT WINAPI xxx_QueryInterface(IFace *iface, REFIID riid, void **object) { if (IsEqualGUID(riid, &IID_IUnknown) || IsEqualGUID(riid, &IID_IAnotherSupported) { xxx_AddRef(iface); *object = iface; return S_OK; } *object = NULL; return E_NOINTERFACE; } -- Dmitry.

1 0

Re: winmm/tests: Use BOOL type where appropriate
by Marvin 11 Oct '13

11 Oct '13

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 https://newtestbot.winehq.org/JobDetails.pl?Key=2721 Your paranoid android. === wxppro (32 bit wave) === wave.c:498: Test failed: waveOutGetPosition(WAVE_MAPPER): returned 13211 bytes, should be 13230 wave.c:498: Test failed: waveOutGetPosition(WAVE_MAPPER): returned 13229 bytes, should be 13230 wave.c:498: Test failed: waveOutGetPosition(WAVE_MAPPER): returned 13075 bytes, should be 13230 wave.c:509: Test failed: waveOutGetPosition(WAVE_MAPPER): returned 13091 samples (13091 bytes), should be 13230 (13230 bytes) wave.c:522: Test failed: waveOutGetPosition(WAVE_MAPPER): returned 594 ms, (13108 bytes), should be 600 (13230 bytes) wave.c:498: Test failed: waveOutGetPosition(WAVE_MAPPER): returned 13220 bytes, should be 13230 wave.c:498: Test failed: waveOutGetPosition(WAVE_MAPPER): returned 17621 bytes, should be 17640 wave.c:509: Test failed: waveOutGetPosition(WAVE_MAPPER): returned 17634 samples (17634 bytes), should be 17640 (17640 bytes) wave.c:498: Test failed: waveOutGetPosition(WAVE_MAPPER): returned 22016 bytes, should be 22050 wave.c:509: Test failed: waveOutGetPosition(WAVE_MAPPER): returned 22027 samples (22027 bytes), should be 22050 (22050 bytes) wave.c:522: Test failed: waveOutGetPosition(WAVE_MAPPER): returned 999 ms, (22044 bytes), should be 1000 (22050 bytes)

1 0

Re: wined3d: Use BOOL type where appropriate
by Henri Verbeet 11 Oct '13

11 Oct '13

On 11 October 2013 22:51, Frédéric Delanoy <frederic.delanoy(a)gmail.com> wrote: > - DWORD onesided_enable = FALSE; > - DWORD twosided_enable = FALSE; > + DWORD onesided_enable = 0; > + DWORD twosided_enable = 0; Actually, all of those initializations are redundant.

1 0

Fwd: Re: kernel32/tests: Add tests for job objects (try 7)
by Andrew Cook 11 Oct '13

11 Oct '13

(no idea why my client sent this to wine-patches) On 10/10/13 15:23, Andrew Cook wrote: > --- > dlls/kernel32/tests/process.c | 159 > +++++++++++++++++++++++++++++++++++++++++- > include/winbase.h | 1 + > include/winnt.h | 90 ++++++++++++++++++++++++ > 3 files changed, 249 insertions(+), 1 deletion(-) > > there appears to be a window between the all processes leaving a job and the job actually being considered empty, and no apparent way to wait for this. I can't reproduce this locally either. Is there an accepted way to wait for some inaccessible kernel-side event? reordering some calls would likely hide the issue but i'd rather avoid that. https://newtestbot.winehq.org/JobDetails.pl?Key=2687 failure on w8 https://newtestbot.winehq.org/JobDetails.pl?Key=2709 failure on w864 https://newtestbot.winehq.org/JobDetails.pl?Key=2685 same patch, success on w864

1 0

Re: xmllite: Use BOOL type where appropriate
by Nikolay Sivov 10 Oct '13

10 Oct '13

On 10/10/2013 00:36, Frédéric Delanoy wrote: > --- > dlls/xmllite/reader.c | 16 ++++++++-------- > 1 file changed, 8 insertions(+), 8 deletions(-) > > diff --git a/dlls/xmllite/reader.c b/dlls/xmllite/reader.c > index 0a4423c..a216951 100644 > --- a/dlls/xmllite/reader.c > +++ b/dlls/xmllite/reader.c > @@ -726,7 +726,7 @@ static void readerinput_grow(xmlreaderinput *readerinput, int length) > } > } > > -static inline int readerinput_is_utf8(xmlreaderinput *readerinput) > +static inline BOOL readerinput_is_utf8(xmlreaderinput *readerinput) > { > static char startA[] = {'<','?'}; > static char commentA[] = {'<','!'}; > @@ -953,7 +953,7 @@ static void reader_skipn(xmlreader *reader, int n) > } > } > > -static inline int is_wchar_space(WCHAR ch) > +static inline BOOL is_wchar_space(WCHAR ch) > { > return ch == ' ' || ch == '\t' || ch == '\r' || ch == '\n'; > } > @@ -1055,7 +1055,7 @@ static HRESULT reader_parse_versioninfo(xmlreader *reader) > } > > /* ([A-Za-z0-9._] | '-') */ > -static inline int is_wchar_encname(WCHAR ch) > +static inline BOOL is_wchar_encname(WCHAR ch) > { > return ((ch >= 'A' && ch <= 'Z') || > (ch >= 'a' && ch <= 'z') || > @@ -1269,7 +1269,7 @@ static HRESULT reader_parse_comment(xmlreader *reader) > } > > /* [2] Char ::= #x9 | #xA | #xD | [#x20-#xD7FF] | [#xE000-#xFFFD] | [#x10000-#x10FFFF] */ > -static inline int is_char(WCHAR ch) > +static inline BOOL is_char(WCHAR ch) > { > return (ch == '\t') || (ch == '\r') || (ch == '\n') || > (ch >= 0x20 && ch <= 0xd7ff) || > @@ -1279,7 +1279,7 @@ static inline int is_char(WCHAR ch) > } > > /* [13] PubidChar ::= #x20 | #xD | #xA | [a-zA-Z0-9] | [-'()+,./:=?;!*#@$_%] */ > -static inline int is_pubchar(WCHAR ch) > +static inline BOOL is_pubchar(WCHAR ch) > { > return (ch == ' ') || > (ch >= 'a' && ch <= 'z') || > @@ -1292,7 +1292,7 @@ static inline int is_pubchar(WCHAR ch) > (ch == '_') || (ch == '\r') || (ch == '\n'); > } > > -static inline int is_namestartchar(WCHAR ch) > +static inline BOOL is_namestartchar(WCHAR ch) > { > return (ch == ':') || (ch >= 'A' && ch <= 'Z') || > (ch == '_') || (ch >= 'a' && ch <= 'z') || > @@ -1312,7 +1312,7 @@ static inline int is_namestartchar(WCHAR ch) > } > > /* [4 NS] NCName ::= Name - (Char* ':' Char*) */ > -static inline int is_ncnamechar(WCHAR ch) > +static inline BOOL is_ncnamechar(WCHAR ch) > { > return (ch >= 'A' && ch <= 'Z') || > (ch == '_') || (ch >= 'a' && ch <= 'z') || > @@ -1336,7 +1336,7 @@ static inline int is_ncnamechar(WCHAR ch) > (ch >= 0xfdf0 && ch <= 0xfffd); > } > > -static inline int is_namechar(WCHAR ch) > +static inline BOOL is_namechar(WCHAR ch) > { > return (ch == ':') || is_ncnamechar(ch); > } I don't actually see what this will achieve, but I see such patches are accepted. Is it a new style rule?

3 2

Re: mshtml: Compile tests with __WINESRC__ define.
by Marvin 10 Oct '13

10 Oct '13

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 https://newtestbot.winehq.org/JobDetails.pl?Key=2689 Your paranoid android. === w864 (64 bit activex) === activex.c:1346: Test failed: unexpected call SetObjectRects === wxppro (32 bit htmldoc) === htmldoc.c:2898: Test failed: mon(0026BFD0) != exmon(00269670) htmldoc.c:7601: Test failed: mon(0026BFD0) != exmon(00269670) htmldoc.c:5978: Test failed: mon(0026BFD0) != exmon(00269670) htmldoc.c:2898: Test failed: mon(0029CA28) != exmon(00286FF8) htmldoc.c:7601: Test failed: mon(0029CA28) != exmon(00286FF8) htmldoc.c:5978: Test failed: mon(0029CA28) != exmon(00286FF8) === wvista (32 bit htmldoc) === No test summary line found === w7pro64 (32 bit htmldoc) === No test summary line found

1 0

Re: ieframe: Compile tests with __WINESRC__ define.
by Marvin 10 Oct '13

10 Oct '13

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 https://newtestbot.winehq.org/JobDetails.pl?Key=2688 Your paranoid android. === wxppro (32 bit webbrowser) === webbrowser.c:792: Test failed: ReadyState = 1, expected 4

1 0

Re: wmvcore: Stub implementation of IWMMetadataEditor interface
by Nikolay Sivov 10 Oct '13

10 Oct '13

On 10/10/2013 14:58, Jeff Latimer wrote: > --- > dlls/wmvcore/Makefile.in | 2 +- > dlls/wmvcore/wmvcore_main.c | 100 > +++++++++++++++++++++++++++++++++++++++++++- > include/wmsdkidl.idl | 11 ++--- > 3 files changed, 105 insertions(+), 8 deletions(-) > +typedef struct MetadataEditorImpl { > + IWMMetadataEditor MetadataEditor_iface; > + LONG ref; > +} MetadataEditorImpl; > +static inline MetadataEditorImpl *impl_from_MetadataEditor(IWMMetadataEditor *iface) Please follow naming convetions for interfaces. > +HRESULT WINAPI WMCreateEditor_close(IWMMetadataEditor *iface) > +{ > + FIXME("iface=%p\n", iface); > + HeapFree(GetProcessHeap(), 0, iface); > + return S_OK; > +} > +HRESULT WINAPI WMCreateEditor_flush(IWMMetadataEditor *iface) > +{ > + FIXME("iface=%p\n", iface); > + HeapFree(GetProcessHeap(), 0, iface); > + return S_OK; > +} This looks totally wrong. Also please keep implementation functions order and interface methods order the same.

1 0

Re: [PATCH 5/5] user32: Implement better stub of OpenInputDesktop.
by Qian Hong 10 Oct '13

10 Oct '13

Hello, this patch is in pending status, is there any way I can improve it? IMO there is no way to 'correctly' implement OpenInputDesktop before implementing SwitchDesktop, as far as SwitchDesktop is a stub, it is safe to assume that OpenInputDesktop will always return either NULL or Winsta0/Default, that is what the patch does and what the tests in [PATCH 3/5] shows.. It is not trivial to implement SwitchDesktop, also I don't know any real world app needing SwitchDesktop except some virtual desktop manager, so I'm not sure it is worth to spend time on implementing SwitchDesktop. However, OpenInputDesktop is needed by multiple apps (TeamViewer, QQ International, Inspect tool from Windows Platform SDK as bug 12067), is it acceptable to submit such a 'better stub' to Wine and leaving SwitchDesktop as a stub? Any comment is great appreciated! On Tue, Oct 8, 2013 at 11:41 AM, Qian Hong <qhong(a)codeweavers.com> wrote: > Fixed http://bugs.winehq.org/show_bug.cgi?id=12067 , let QQ users happy :) > > --- > dlls/user32/tests/winstation.c | 22 ---------------------- > dlls/user32/winstation.c | 22 +++++++++++++++++++--- > 2 files changed, 19 insertions(+), 25 deletions(-) > > -- Regards, Qian Hong - http://www.codeweavers.com

2 2

Re: Fix text rotation problem in GM_ADVANCED graphics mode caused by incorrect implementation of GetTextExtentPointW().
by Alexandre Julliard 10 Oct '13

10 Oct '13

Ralf Habacker <ralf.habacker(a)freenet.de> writes: > With other patches i have been told to implement such stuff in the dib > driver. Unfortunally this do not works in this case, because in the top > level function it looks like having driver specific stuff using display > coordinates. It would still most likely have to be in the driver, though maybe the driver would not be calling that exact entry point. In any case, you can't change the DC transform like this. and you'll need test cases. -- Alexandre Julliard julliard(a)winehq.org

3 5