February 2011 - Wine-Devel - list.winehq.org

Re: [PATCH 01/13] dsound: New sample rate converter core functions.

by Dmitry Timoshkov

Krzysztof Nikiel <knik00(a)gmail.com> wrote: > High quality sample rate converter. > New code seems simplier and faster despite the big sound quality > improvement. > > This patchset only makes sense when applying all 13 parts. It looks like at least some patches got wrapped by your mailer. You can't send Makefile changes separately from added/removed files, a patch should not add dead code. Could you use 4 space indetation for the new code please? -- Dmitry.

Feb. 11, 2011

[rfc] dsound: decrease sound latency

by Maarten Lankhorst

Hi all, Please test this and let me know if it's causing any problems in sound playback, preferably with media applications and games, any feedback would be appreciated. The patch won't fix existing problems, it should just decrease the audio lag. As such I would prefer only to receive feedback on issues INTRODUCED or FIXED by this patch. If it's causing issues, I want to know: 1. Whether pulseaudio is running, and pulseaudio --version 2. What distribution you're using 3. (If you know it) the version of alsa's libraries, packages libasound2 and libasound2-plugins in debian/ubuntu Cheers, Maarten

Feb. 11, 2011

winspool.drv : Implement AbortPrinter() function

by Loïc Maury

Feb. 11, 2011

Re: msxml3: Ignore IActiveScript interface for IXMLHTTPRequest

by Alistair Leslie-Hughes

Hi, On 11/02/2011 8:12 PM, Nikolay Sivov wrote: > On 2/11/2011 10:51, Alistair Leslie-Hughes wrote: >> Hi, >> >> >> Changelog: >> msxml3: Ignore IActiveScript interface for IXMLHTTPRequest > What's a purpose of this? To remove the FIXME from appearing in the logs. Best Regards Alistair Leslie-Hughes

Feb. 11, 2011

Re: msxml3: Ignore IActiveScript interface for IXMLHTTPRequest

by Nikolay Sivov

On 2/11/2011 10:51, Alistair Leslie-Hughes wrote: > Hi, > > > Changelog: > msxml3: Ignore IActiveScript interface for IXMLHTTPRequest What's a purpose of this? > > > Best Regards > Alistair Leslie-Hughes

Feb. 11, 2011

Re: avifil32: Skip tests when we couldn't create or open the AVI file

by Juan Lang

> Nevertheless, I do think the case could happen again (not enough disk space > for instance). IMO, cases like this aren't worth worrying about. They're like out of memory conditions: sure, they can happen, but there's not much we can do about them. The disk space requirements of the tests aren't large, so testers should ensure they have a reasonable amount available. --Juan

Feb. 11, 2011

Re: avifil32: Skip tests when we couldn't create or open the AVI file

by GOUJON Alexandre

On 02/11/2011 12:10 AM, Julius Schwartzenberg wrote: > Is there a specific situation in which the file creation fails? I > expected the temporary directory would always be writable when I > created it like that. Maybe that is not a good assumption when writing > tests though? I wrote the patch because of the crash seen here : http://test.winehq.org/data/020245bba10e15e2a1b415f4a4a08d43f9fb1025/xp_nao… However, I contacted the user who submitted it and it was in fact a "stripped" version of Windows. Nevertheless, I do think the case could happen again (not enough disk space for instance). Moreover, it's basically a "if" "win_skip" and "return" so it won't slow down the test. What do you think ? Does it worth applying my patch ? Well, it's up to you; you know my point of view. Thanks

Feb. 11, 2011

Re: [PATCH 2/2] msvcrt: Implement _strnicoll.

by Dmitry Timoshkov

Eryk Wieliczko <ewdevel(a)gmail.com> wrote: > /********************************************************************* > + * _strnicoll (MSVCRT.@) > + */ > +int CDECL MSVCRT__strnicoll(const char* str1, const char* str2, int digits) > +{ > + int res; > + > + res = strncmp(str1, str2, digits); > + if (res > 0) > + res = 1; > + else if (res < 0) > + res = -1; > + > + return res; > +} The API name implies case insensitive comparison, please add the tests for that. -- Dmitry.

Feb. 11, 2011

.chm files for builtin programs?

by Dan Kegel

Hi all, The native dxdiag uses a .chm file for its help system, so ours might as well, too. I see there's a .chm file checked in to git (in dlls/itss/tests). Any objection to checking a .chm file into programs/dxdiag, along with its source files and script to build it? I think there's an open source .chm compiler at http://sourceforge.net/projects/chmc/ Haven't tried it yet. If that doesn't work, we can use Microsoft's hhc. - Dan

Feb. 11, 2011

Re: ole32/tests: make sure to use a return value (LLVM/Clang) (2/3)

by Marvin

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=9189 Your paranoid android. === WNT4WSSP6 (32 bit ole2) === ole2.c:1660: Test failed: IOleObject_SetExtent failed with error 0x80040005 === W2KPROSP4 (32 bit ole2) === ole2.c:1660: Test failed: IOleObject_SetExtent failed with error 0x80040005 === WXPPROSP3 (32 bit ole2) === ole2.c:1660: Test failed: IOleObject_SetExtent failed with error 0x80040005 === W2K3R2SESP2 (32 bit ole2) === ole2.c:1660: Test failed: IOleObject_SetExtent failed with error 0x80040005 === WVISTAADM (32 bit ole2) === ole2.c:1660: Test failed: IOleObject_SetExtent failed with error 0x80040005 === W2K8SE (32 bit ole2) === ole2.c:1660: Test failed: IOleObject_SetExtent failed with error 0x80040005 === W7PRO (32 bit ole2) === ole2.c:1660: Test failed: IOleObject_SetExtent failed with error 0x80040005 === W7PROX64 (32 bit ole2) === ole2.c:1660: Test failed: IOleObject_SetExtent failed with error 0x80040005 === W7PROX64 (64 bit ole2) === ole2.c:1660: Test failed: IOleObject_SetExtent failed with error 0x80040005

Feb. 10, 2011

Re: avifil32: Skip tests when we couldn't create or open the AVI file

by Julius Schwartzenberg

GOUJON Alexandre wrote: > I know I sent this patch about two weeks ago but I thought it was > pending because of the 1.3.13 release. > So, does anyone has an idea how to make this patch applied in the > current tree ? > > The context is that we don't fail when we couldn't create or open the > AVI file. (I think the title of the patch is explicit) > Thus, we call "WriteFile(hFile" in the first part of the test even if > hFile == INVALID_HANDLE_VALUE. > In the second part, we call "AVIFileGetStream(pFile" even if > "AVIFileOpen(&pFile" failed. Is there a specific situation in which the file creation fails? I expected the temporary directory would always be writable when I created it like that. Maybe that is not a good assumption when writing tests though? Thanks, Julius

Feb. 10, 2011

Re: shell32/tests: make sure return values are used (LLVM/Clang) (4/7)

by Marvin

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=9177 Your paranoid android. === WNT4WSSP6 (32 bit shlfileop) === shlfileop.c:2210: Test failed: SHGetNewLinkInfoA succeeded

Feb. 10, 2011

Re: msi/tests: make sure to use a return value (LLVM/Clang) (2/2)

by Marvin

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=9162 Your paranoid android. === WVISTAADM (32 bit db) === Failure running script in VM: The specified guest user must be logged in interactively to perform this operation

Feb. 10, 2011

Re: user32/tests: Add test for nested default Button in Dialogs

by Marvin

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=9151 Your paranoid android. === WVISTAADM (32 bit dialog) === Failure running script in VM: The specified guest user must be logged in interactively to perform this operation

Feb. 10, 2011

Re: avifil32: Skip tests when we couldn't create or open the AVI file

by GOUJON Alexandre

On 01/30/2011 07:29 PM, Alexandre Goujon wrote: > --- > dlls/avifil32/tests/api.c | 12 ++++++++++++ > 1 files changed, 12 insertions(+), 0 deletions(-) > > diff --git a/dlls/avifil32/tests/api.c b/dlls/avifil32/tests/api.c > index 14ad4b4..c1ba878 100644 > --- a/dlls/avifil32/tests/api.c > +++ b/dlls/avifil32/tests/api.c > @@ -338,6 +338,12 @@ static void create_avi_file(const COMMON_AVI_HEADERS *cah, char *filename) > > ok(hFile != INVALID_HANDLE_VALUE, "Couldn't create file\n"); > > + if(hFile == INVALID_HANDLE_VALUE) > + { > + win_skip("Could not create the AVI file\n"); > + return; > + } > + > WriteFile(hFile,&cah->fh, sizeof(deffh),&written, NULL); > WriteFile(hFile,&cah->mah, sizeof(MainAVIHeader),&written, NULL); > WriteFile(hFile, streamlist, sizeof(streamlist),&written, NULL); > @@ -380,6 +386,12 @@ static void test_default_data(void) > ok(res != REGDB_E_CLASSNOTREG, "Unable to open file: error5=%u\n", REGDB_E_CLASSNOTREG); > ok(res == 0, "Unable to open file: error=%u\n", res); > > + if(res != 0) > + { > + win_skip("Could not open the file.\n"); > + return; > + } > + > res = AVIFileGetStream(pFile,&pStream0, 0, 0); > ok(res == 0, "Unable to open video stream: error=%u\n", res); I know I sent this patch about two weeks ago but I thought it was pending because of the 1.3.13 release. So, does anyone has an idea how to make this patch applied in the current tree ? The context is that we don't fail when we couldn't create or open the AVI file. (I think the title of the patch is explicit) Thus, we call "WriteFile(hFile" in the first part of the test even if hFile == INVALID_HANDLE_VALUE. In the second part, we call "AVIFileGetStream(pFile" even if "AVIFileOpen(&pFile" failed. I can also resend this appending [resent]. Just let me know what to do. Thanks !

Feb. 10, 2011

Re: d3dx9: Implement D3DXFloat16to32Array and D3DXFloat32to16Array.

by Misha Koshelev

On Thu, Feb 10, 2011 at 6:13 AM, Stefan Dösinger <stefandoesinger(a)gmx.at> wrote: >> The brace goes in column 1 so that [[ and ]] work in vi. >> Putting the function name in column 1 lets you search for the >> definition using the anchored regexp ^function_name\> > Code style in Wine doesn't have anything to do with editors. The basic rule is > that when you change existing code or add code to an existing module you > should use the style used by the existing code. If you add a new module you > may pick any style you like as long as it is somewhat sane. Just fyi, I do not mean to speak for Henri, but I believe he had some code preferences with regards to contributing new code to d3dx9 that are stylistically somewhat different than the existing code that was there (at least in mesh.c). Thank you Yours Misha

Feb. 10, 2011

Re: d3dx9: Implement D3DXFloat16to32Array and D3DXFloat32to16Array.

by Stefan Dösinger

> The brace goes in column 1 so that [[ and ]] work in vi. > Putting the function name in column 1 lets you search for the > definition using the anchored regexp ^function_name\> Code style in Wine doesn't have anything to do with editors. The basic rule is that when you change existing code or add code to an existing module you should use the style used by the existing code. If you add a new module you may pick any style you like as long as it is somewhat sane.

Feb. 10, 2011

Re: [3/3] user32: SetParent() needs to map window rectangle to the new parent.

by Marvin

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=9136 Your paranoid android. === W7PROX64 (64 bit msg) === Timeout

Feb. 10, 2011

Re: rpcrt4/tests: remove a dead assignment (LLVM/Clang)

by Austin English

On Wed, Feb 9, 2011 at 15:53, Marvin <testbot(a)testbot.winehq.org> wrote: > 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=9120 > > Your paranoid android. > > > === WNT4WSSP6 (32 bit ndr_marshall) === > ndr_marshall: unhandled exception c0000005 at 004081C6 Compiler bug: https://testbot.winehq.org/JobDetails.pl?Key=9134 patch used to test: diff --git a/dlls/rpcrt4/tests/ndr_marshall.c b/dlls/rpcrt4/tests/ndr_marshall.c index 549d09e..9597bf3 100644 --- a/dlls/rpcrt4/tests/ndr_marshall.c +++ b/dlls/rpcrt4/tests/ndr_marshall.c @@ -17,7 +17,7 @@ * License along with this library; if not, write to the Free Software * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA */ - +// #include <stdarg.h> #define NTDDI_WIN2K 0x05000000 -- -Austin

Feb. 10, 2011

Re: rpcrt4/tests: remove a dead assignment (LLVM/Clang)

by Marvin

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=9120 Your paranoid android. === WNT4WSSP6 (32 bit ndr_marshall) === ndr_marshall: unhandled exception c0000005 at 004081C6

Feb. 9, 2011

Re: winex11.drv: map_window should honor iconic state

by Alexandre Julliard

Alexander Lakhin <exclusion(a)gmail.com> writes: > + data->iconic = (new_style & WS_MINIMIZE) != 0; > + > if (!data->embedded) > { > update_net_wm_states( display, data ); > sync_window_style( display, data ); > wine_tsx11_lock(); > XMapWindow( display, data->whole_window ); > + if (data->iconic) > + XIconifyWindow( display, data->whole_window, DefaultScreen(display) ); This is handled by setting the initial window state in the wm hints. If there's somewhere that the state isn't set correctly, that's what should be fixed. -- Alexandre Julliard julliard(a)winehq.org

Feb. 9, 2011

Re: ntdll: Increase buffer size for debugging functions

by Alexandre Julliard

Maarten Lankhorst <m.b.lankhorst(a)gmail.com> writes: > Since warn+debugstr takes user data increase and basically does > WARN("%s\n", msg); the buffer size is too small, increasing the size > makes the lord of the rings online launcher print messages to WARN > without crashing That's just hiding the bug. -- Alexandre Julliard julliard(a)winehq.org

Feb. 9, 2011

Re: error message for ddraw

by André Hentschel

Am 09.02.2011 08:20, schrieb David Adam: > When I test the following test > > rc = IDirectDraw_SetCooperativeLevel(lpDD, NULL, DDSCL_FULLSCREEN | DDSCL_EXCLUSIVE | DDSCL_CREATEDEVICEWINDOW); > ok(rc==DDERR_INVALIDPARAMS, "Expected DDERR_INVALIDPARAMS, received %x\n", rc); > > in my vista box, I receive the following failing message > > ddrawmodes: test failed: expected DDERR_INVALIDPARAMS, received 8876035a. > > But I don't know what is this message. Does anyone know what means this message? > > Thanks in advance > > A+ > > David > > > > Hi David, it's DXFILEERR_RESOURCENOTFOUND, defined in include/dxfile.h -- Best Regards, André Hentschel

Feb. 9, 2011

Re: d3dx9: Implement D3DXFloat16to32Array and D3DXFloat32to16Array.

by David Laight

On Tue, Feb 08, 2011 at 01:02:36PM +0800, Dmitry Timoshkov wrote: > > +static inline float float_16_to_32(const unsigned short *in) { > > ... Also placing a brace on its own line would follow the style > of other functions. The brace goes in column 1 so that [[ and ]] work in vi. Putting the function name in column 1 lets you search for the definition using the anchored regexp ^function_name\> Putting { and } on the same lines as for, while and else avoids problems like determining whether this is intended to be the bottom or top of a loop: bar = 1; } while (very long conditional expression that is truncated by the ... { int foo; and whether there is an else on the line you can't see below the bottom of the window when you can see: if (...) { ... } The 'else' problem is much worse if you are looking at a printout on fanfold paper, David -- David Laight: david(a)l8s.co.uk

Feb. 9, 2011

Re: Try to Implement my first Stub function - AbortPrinter() - (try 2).

by Loïc Maury

Hello Mr.Eikum, On 09/02/11 16:30, Andrew Eikum wrote: > On 02/09/2011 09:09 AM, Loïc Maury wrote: >>> The indentation looks fine to me now. You've got an extra newline >>> after the "if(printer->doc)" block and before the "else." The >>> formatting on the TRACE statement is still bizarre. Fix the commas and >>> put it on just a couple of lines. No reason for one line and a ton of >>> whitespace for each parameter. See line 1944 for an example. >> Ok, it's correct now ? > > The content looks fine, but there are still formatting problems in > your patch. You're still using commas strangely in your wrapped TRACE > statement. Your editor is broken again, and you are now mixing tabs > and spaces. A few lines still have whitespace at the end. > > These are things you should verify yourself before sending it off for > someone else to look at. Sorry for this, now I have verified with git diff and git diff --check, and I have no trailing whitespace. I think it's correct. Thank you Loïc > > Andrew >

Feb. 9, 2011