December 2015 - Wine-Devel - list.winehq.org

Re: [PATCH v3] user32: Add MOUSEHOOKSTRUCTEX to fix mouse wheel support for JA2 1.13 and other apps which use it
by André Hentschel Dec. 22, 2015

Dec. 22, 2015

Am 22.12.2015 um 16:34 schrieb Kira Backes: > Hi, Welcome to Wine development! Some suggestions: I would maybe call the inner struct "mouse". What about "user32: Use MOUSEHOOKSTRUCTEX to improve mouse wheel support" as subject/title?

1 0

Help with configure.ac rule wanted
by Uwe Bonnes Dec. 22, 2015

Dec. 22, 2015

Hello, for some special hardware, e.g the STLink from ST Microelectronic for debugging an programming Arm microprocessors, I have translation dlls that translates the hardware access to libusb calls. To detect libusb, until recently I added the libusb detection code from ftp://ftp.etersoft.ru/pub/people/amorozov/usb/ However recenty Alexandre did some work around the configuration area and things no longer work. I would appreciate if anybody can help me with coming up with a libusb detection rule for wine. Thanks -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 1623569 ------- Fax. 06151 1623305 ---------

3 3

Re: ws2_32: Implement automatic broadcast for UDP sockets on sendto()
by Hans Leidekker Dec. 22, 2015

Dec. 22, 2015

Hi Bruno, >+ if (wsa->addr->sa_family == WS_AF_INET) >+ { >+ /* When the target IPv4 address ends in 255 we must always send it as >+ * a broadcast. Trying to send the packet without setting SO_BROADCAST >+ * results in EACCES, to avoid that we will manually enable the flag >+ * and send the packet, after that we will restore the disabled flag >+ * behavior. This is the most common estimate as the broadcast address >+ * should actually be calculated using the netmask for the interface. */ It should really be calculated using the netmask, sending a broadcast packet here is wrong if the user is not in a /24 network.

2 2

Re: ws2_32: Add WSAPoll() implementation (resend)
by Marvin Dec. 22, 2015

Dec. 22, 2015

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://testbot.winehq.org/JobDetails.pl?Key=19092 Your paranoid android. === w7u (32 bit sock) === sock.c:640: Test failed: oob_server (ae0): not at the OOB mark: 1

1 0

Re: [PATCH 1/2] winhlp32: fix table formatting
by Nikolay Sivov Dec. 22, 2015

Dec. 22, 2015

On 21.12.2015 22:59, Jean-Christophe Cardot wrote: > Fixes https://bugs.winehq.org/show_bug.cgi?id=20977 > Fixes https://bugs.winehq.org/show_bug.cgi?id=14288 > > Fix bugs in the handling of the hlp values in order to generate correct > rtf tables; Read the DPI from the registry (instead of hardcoded 72); > Use font scaling (depends on the version) to compute the rtf values > so that the columns width are correct and precise (compared with MS > winhlp32.exe); Add some comments and DEFINES tu make the code easier > to understand. > This is a first step into making wine's winhlp much better at handling > hlp files and - when time permits - mvp files. > > Tested on Linux (Mageia 5) > Hi, Jean-Christophe. Thanks for working on this. Please try to split this one to several patches, for example one to introduce symbols for magic numbers, one for HLPFILE_ReadFont changes, one for DPI changes etc. It will make it much easier to read. > +#define DEFDPI 96 > +static const WCHAR logpixels_reg[] = > {'S','y','s','t','e','m','\\','C','u','r','r','e','n','t','C','o','n','t','r','o','l','S','e','t','\\','H','a','r','d','w','a','r','e',' > ','P','r','o','f','i','l','e','s','\\','C','u','r','r','e','n','t','\\','S','o','f','t','w','a','r','e','\\','F','o','n','t','s',0}; > +static const WCHAR logpixels[] = {'L','o','g','P','i','x','e','l','s',0}; You sure you need this? Won't it work with GetDevicesCaps() on desktop HDC? Also your mail client wrapped your patch, so it can't be applied. You need to fix the way it sends inline patches, send it as attachment, or use git-send-email to do the job for you.

1 0

Re: [PATCH v3] msvcrt: Rename __libm_sse2_* functions to fix compilation, when using Intel compiler.
by Piotr Caban Dec. 22, 2015

Dec. 22, 2015

> diff --git a/dlls/msvcrt/math.c b/dlls/msvcrt/math.c > index 2459bf5..270bcda 100644 > --- a/dlls/msvcrt/math.c > +++ b/dlls/msvcrt/math.c > @@ -1980,7 +1980,7 @@ void _safe_fprem1(void) > /*********************************************************************** > * __libm_sse2_acos (MSVCRT.@) > */ The patch is still not applying cleanly. For some reason in your diff there are 8 spaces before __libm_sse2_acos while in math.c there are 2 tabs. Thanks, Piotr

2 1

Re: [PATCH v2] msvcrt: Rename __libm_sse2_* functions to fix compilation when using Intel compiler.
by Piotr Caban Dec. 22, 2015

Dec. 22, 2015

On 12/22/15 11:23, Paul Gofman wrote: > +void __cdecl MSVCRT___libm_sse2_atan2(void) > { > double d1, d2; > __asm__ __volatile__( "movq %%xmm0,%0; movq %%xmm1,%1 " : "=m" > (d1), "=m" (d2) ); > @@ -2046,7 +2046,7 @@ void __cdecl __libm_sse2_atan2(void) Your mail client wraps the lines: Applying: msvcrt: Rename __libm_sse2_* functions to fix compilation when using Intel compiler. fatal: patch fragment without header at line 394: @@ -2046,7 +2046,7 @@ void __cdecl __libm_sse2_atan2(void) Thanks, Piotr

2 1

Re: [PATCH] msvcrt: change __libm_sse2_... functions internal names (exported names preserved) msvcr80, msvcr90, msvcr100, msvcr110, msvcr120: update spec file
by Piotr Caban Dec. 22, 2015

Dec. 22, 2015

On 12/22/15 10:54, Paul Gofman wrote: > -void __cdecl __libm_sse2_acos(void) > +void __cdecl l__libm_sse2_acos(void) In other places where we need to rename functions "MSVCRT_" prefix is added to function name. Could you please change the function names this way? Also it will be nicer to change the commit message to something like: msvcrt: Rename __libm_sse2_* functions to fix compilation when using Intel compiler. Thanks, Piotr

2 1

Re: [PATCH 2/2] oledb32: GetConversionSize shouldn't fail for null variants. (try 2)
by Huw Davies Dec. 22, 2015

Dec. 22, 2015

On Tue, Dec 22, 2015 at 09:09:46PM +1100, Alistair Leslie-Hughes wrote: > diff --git a/dlls/oledb32/tests/convert.c b/dlls/oledb32/tests/convert.c > index a785457..ac2f6be 100644 > --- a/dlls/oledb32/tests/convert.c > +++ b/dlls/oledb32/tests/convert.c > @@ -2627,6 +2627,33 @@ static void test_getconversionsize(void) > ok(hr == S_OK, "got 0x%08x\n", hr); > VariantClear(&var); > > + dst_len = 78; > + V_VT(&var) = VT_NULL; > + hr = IDataConvert_GetConversionSize(convert, DBTYPE_VARIANT, DBTYPE_WSTR, NULL, &dst_len, &var); > + ok(hr == S_OK, "got 0x%08x\n", hr); > + > + dst_len = 0; > + src_len = 20; > + V_VT(&var) = VT_BSTR; > + V_BSTR(&var) = SysAllocString(strW); > + hr = IDataConvert_GetConversionSize(convert, DBTYPE_VARIANT, DBTYPE_STR, &src_len, &dst_len, &var); > + ok(hr == S_OK, "got 0x%08x\n", hr); > + ok(dst_len == 5, "%ld\n", dst_len); > + VariantClear(&var); > + > + dst_len = 0; > + src_len = 20; > + V_VT(&var) = VT_I4; > + V_I4(&var) = 4; > + hr = IDataConvert_GetConversionSize(convert, DBTYPE_VARIANT, DBTYPE_STR, &src_len, &dst_len, &var); > + ok(hr == S_OK, "got 0x%08x\n", hr); > + VariantClear(&var); > + > + /* On Windows dst_len will get a value of 110 but we aren't testing this parameter. */ > + V_VT(&var) = VT_NULL; > + hr = IDataConvert_GetConversionSize(convert, DBTYPE_VARIANT, DBTYPE_STR, NULL, &dst_len, &var); > + ok(hr == S_OK, "got 0x%08x\n", hr); > + Why did you put the comment on the 2nd VT_NULL conversion rather than the first? Same goes for the unneeded dst_len initiailization (which also applies in the VT_I4 case). If would probably make more sense to group these three together, i.e. move the VT_BSTR case to the top, then have the comment, then the three tests (all of which do not need the dst_len init). Huw.

1 0

Re: [PATCH 2/2] oledb32: GetConversionSize shouldn't fail for null variants.
by Huw Davies Dec. 22, 2015

Dec. 22, 2015

On Tue, Dec 22, 2015 at 06:32:04PM +1100, Alistair Leslie-Hughes wrote: > diff --git a/dlls/oledb32/tests/convert.c b/dlls/oledb32/tests/convert.c > index a785457..55539f9 100644 > --- a/dlls/oledb32/tests/convert.c > +++ b/dlls/oledb32/tests/convert.c > @@ -2627,6 +2627,33 @@ static void test_getconversionsize(void) > ok(hr == S_OK, "got 0x%08x\n", hr); > VariantClear(&var); > > + dst_len = 78; > + V_VT(&var) = VT_NULL; > + hr = IDataConvert_GetConversionSize(convert, DBTYPE_VARIANT, DBTYPE_WSTR, NULL, &dst_len, &var); > + ok(hr == S_OK, "got 0x%08x\n", hr); You should test dst_len after return. Likewise below. Huw.

4 7