November 2012 - Wine-devel - list.winehq.org

Re: [PATCH] tools/widl/header.c: Add a way to declare interface data members.
by Michael Stefaniuc 12 Nov '12

12 Nov '12

You can't do that as the COM standard specifies also the ABI. A COM interface is a pointer to a virtual table; it is *not* a struct. That's just an implementation detail in C. On 11/10/2012 08:48 PM, max(a)mtew.isa-geek.net wrote: > From: Max TenEyck Woodbury <max+git(a)mtew.isa-geek.net> > > --- > tools/widl/header.c | 14 ++++++++++++++ > 1 files changed, 14 insertions(+), 0 deletions(-) > > diff --git a/tools/widl/header.c b/tools/widl/header.c > index 2f275c7..965dcbc 100644 > --- a/tools/widl/header.c > +++ b/tools/widl/header.c > @@ -1159,6 +1159,17 @@ static void write_com_interface_start(FILE *header, const type_t *iface) > fprintf(header,"#define __%s_%sINTERFACE_DEFINED__\n\n", iface->name, dispinterface ? "DISP" : ""); > } > > +static void write_interface_data_macro(FILE *header, const type_t *iface) > +{ > + > + if (type_iface_get_inherit(iface)) > + write_interface_data_macro(header, type_iface_get_inherit(iface)); > + > + fprintf(header, "#ifdef %s_IFACE_DATA\n", iface->name); > + fprintf(header, " %s_IFACE_DATA\n", iface->name); > + fprintf(header, "#endif /* defined %s */\n", iface->name); > +} > + > static void write_com_interface_end(FILE *header, type_t *iface) > { > int dispinterface = is_attr(iface->attrs, ATTR_DISPINTERFACE); > @@ -1214,6 +1225,9 @@ static void write_com_interface_end(FILE *header, type_t *iface) > fprintf(header, "} %sVtbl;\n", iface->name); > fprintf(header, "interface %s {\n", iface->name); > fprintf(header, " CONST_VTBL %sVtbl* lpVtbl;\n", iface->name); > + > + write_interface_data_macro(header, iface); > + > fprintf(header, "};\n"); > fprintf(header, "\n"); > fprintf(header, "#ifdef COBJMACROS\n"); bye michael

3 6

How to tell Mac from Linux wine within a Wine app? (steam hardware survey)
by Scott Ritchie 10 Nov '12

10 Nov '12

Is there an easy way for the steam hardware survey to tell whether it's running under Mac or Linux Wine? I remember discussing this issue with an engineer from Valve after asking they make the % of Wine users public again.

3 3

Re: [PATCH] tools/widl/header.c: Add a way to declare interface data members.
by Nikolay Sivov 10 Nov '12

10 Nov '12

On 11/10/2012 22:48, max(a)mtew.isa-geek.net wrote: > From: Max TenEyck Woodbury <max+git(a)mtew.isa-geek.net> > > --- > tools/widl/header.c | 14 ++++++++++++++ > 1 files changed, 14 insertions(+), 0 deletions(-) > > diff --git a/tools/widl/header.c b/tools/widl/header.c > index 2f275c7..965dcbc 100644 > --- a/tools/widl/header.c > +++ b/tools/widl/header.c > @@ -1159,6 +1159,17 @@ static void write_com_interface_start(FILE *header, const type_t *iface) > fprintf(header,"#define __%s_%sINTERFACE_DEFINED__\n\n", iface->name, dispinterface ? "DISP" : ""); > } > > +static void write_interface_data_macro(FILE *header, const type_t *iface) > +{ > + > + if (type_iface_get_inherit(iface)) > + write_interface_data_macro(header, type_iface_get_inherit(iface)); > + > + fprintf(header, "#ifdef %s_IFACE_DATA\n", iface->name); > + fprintf(header, " %s_IFACE_DATA\n", iface->name); > + fprintf(header, "#endif /* defined %s */\n", iface->name); > +} > + > static void write_com_interface_end(FILE *header, type_t *iface) > { > int dispinterface = is_attr(iface->attrs, ATTR_DISPINTERFACE); > @@ -1214,6 +1225,9 @@ static void write_com_interface_end(FILE *header, type_t *iface) > fprintf(header, "} %sVtbl;\n", iface->name); > fprintf(header, "interface %s {\n", iface->name); > fprintf(header, " CONST_VTBL %sVtbl* lpVtbl;\n", iface->name); > + > + write_interface_data_macro(header, iface); > + > fprintf(header, "};\n"); > fprintf(header, "\n"); > fprintf(header, "#ifdef COBJMACROS\n"); What is this?

1 0

Re: ole32: Fix outstanding memory leak (coverity)
by Frédéric Delanoy 10 Nov '12

10 Nov '12

On Sat, Nov 10, 2012 at 3:57 PM, André Hentschel <nerv(a)dawncrow.de> wrote: > --- > dlls/ole32/storage32.c | 5 ++--- > 1 file changed, 2 insertions(+), 3 deletions(-) > > diff --git a/dlls/ole32/storage32.c b/dlls/ole32/storage32.c > index 8498eaf..4099df6 100644 > --- a/dlls/ole32/storage32.c > +++ b/dlls/ole32/storage32.c > @@ -9067,17 +9067,16 @@ static HRESULT OLECONVERT_GetOLE10ProgID(LPSTORAGE pStorage, char *strProgID, DW > else > { > STATSTG stat; > - LPOLESTR wstrProgID; > + LPOLESTR wstrProgID = NULL; > > /* Get the OleType from the registry */ > REFCLSID clsid = &(stat.clsid); > IStorage_Stat(pStorage, &stat, STATFLAG_NONAME); > hRes = ProgIDFromCLSID(clsid, &wstrProgID); > if(hRes == S_OK) > - { > *dwSize = WideCharToMultiByte(CP_ACP, 0, wstrProgID, -1, strProgID, *dwSize, NULL, FALSE); > + if(wstrProgID) > CoTaskMemFree(wstrProgID); > - } CoTaskMemFree is no-op when given NULL argument so "if(wstrProgID)" is not necessary (since you initialize it to NULL) Frédéric

1 0

Re: [PATCH 1/2] msvcp90: Free memory allocated by _get_current_locale function
by Marvin 09 Nov '12

09 Nov '12

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=22858 Your paranoid android. === WVISTAADM (32 bit misc) === misc: unhandled exception c0000005 at 77496739 === W2K8SE (32 bit misc) === misc: unhandled exception c0000005 at 778A59C3 === W7PRO (32 bit misc) === misc: unhandled exception c0000005 at 772D1FFE === W7PROX64 (32 bit misc) === misc: unhandled exception c0000005 at 77DFE23E === TEST64_W7SP1 (32 bit misc) === misc: unhandled exception c0000005 at 7758E39E === W7PROX64 (64 bit misc) === Timeout === TEST64_W7SP1 (64 bit misc) === Timeout

1 0

The EFF is asking for our help explaining why APIs should not be copyrightable
by Scott Ritchie 09 Nov '12

09 Nov '12

https://www.eff.org/deeplinks/2012/11/no-copyrights-apis-help-us-make-case There are court cases pending that are very, very relevant to what we do here. It would help the EFF very much if a few engineers could send them a short email statement explaining how Wine's reimplementation of the API is good for interoperability and competition. Thanks, Scott Ritchie

2 1

[PATCH 1/2] winmm/tests: Add more mmioSeek tests.
by Joerg-Cyril.Hoehle＠t-systems.com 09 Nov '12

09 Nov '12

Sagawa San, it's nice to see somebody contribute to winmm! + return; +} That return is line noise. + hmmio = mmioOpen(NULL, &mmio, MMIO_READ); + if (hmmio == NULL) { + win_skip("mmioOpen error %u\n", mmio.wErrorRet); + return; + } AFAIK mmioOpen has always been present, so there's no reason for a win_skip. An error should be flagged in all cases, Wine or native. Furthermore, the early return prevents execution of the other groups of tests. Hence I suggest: + hmmio = mmioOpen(NULL, &mmio, MMIO_READ); + ok(hmmio != NULL, "mmioOpen error %u\n", mmio.wErrorRet); + if (hmmio) { mmioSeek ... mmioClose } However, the 2nd and 3rd mmioOpen may fail if the test file could not be created, so you may want a different treatment there. Perhaps add a boolean result to create_test_file? >[PATCH 2/2] winmm: Fix SEEK_END direction of mmio files without buffering. It's always good in a patch submission to mention what bug it fixes. Does yours fix bug 19566? I seem to remember that there was an entry in bugzilla about MMIO and seeking... Oh yes, you may look at bug 28982, perhaps your patch fixes that too. Regards, Jörg Höhle

2 2

Re: [PATCH 1/2] gdi32: Ignore contours containing one point in get_glyph_outline function
by Piotr Caban 09 Nov '12

09 Nov '12

Please ignore this patches for now, the tests are broken.

1 0

Re: [PATCH 1/2] user32: Add tests for MapWindowPoints, ClientToScreen and ScreenToClient.
by Alexandre Julliard 09 Nov '12

09 Nov '12

Christian Costa <titan.costa(a)gmail.com> writes: > + SetLastError(0xdeadbeef); > + n = MapWindowPoints(wnd, NULL, NULL, 0); > + err = GetLastError(); > + ok(n == ((window_rect.top << 16) | window_rect.left), "Got %x (%d, %d), expected %x (%d, %d)\n", > + n, (n << 16) >> 16, n >> 16, window_rect.left, window_rect.top, (window_rect.top << 16) | window_rect.left); > + ok(err == 0xdeadbeef, "Got %x, expected %x\n", err, 0xdeadbeef); > + > + SetLastError(0xdeadbeef); > + n = MapWindowPoints(NULL, wnd, NULL, 0); > + err = GetLastError(); > + ok(n == ((-window_rect.top << 16) | (-window_rect.left & 0xffff)), "Got %x (%d, %d), expected %x (%d, %d)\n", > + n, (n << 16) >> 16, n >> 16, (-window_rect.top << 16) | (-window_rect.left & 0xffff), -window_rect.left, -window_rect.top); > + ok(err == 0xdeadbeef, "Got %x, expected %x\n", err, 0xdeadbeef); Testing last error on success is not useful. Also please use MAKELONG and related macros instead of doing it by hand. -- Alexandre Julliard julliard(a)winehq.org

2 2

Re: [PATCH 2/2] gdi32: Added GetGlyphOutlineW tests on glyph that contains empty contour
by Marvin 09 Nov '12

09 Nov '12

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=22852 Your paranoid android. === WVISTAADM (32 bit font) === Failure running script in VM: The specified guest user must be logged in interactively to perform this operation

1 0