September 2010 - Wine-Devel

Re: winedbg: Only add ContextFlags which are defined
by Dmitry Timoshkov Sept. 9, 2010

Sept. 9, 2010

André Hentschel <nerv(a)dawncrow.de> wrote: > CONTROL and INTEGER are standard defines, but ia64 and ARM don't define CONTEXT_FLOATING_POINT ia64 does define CONTEXT_FLOATING_POINT, ARM probably should also do. -- Dmitry.

2 3

Re: user32: Update Italian translation
by Paul Vriens Sept. 9, 2010

Sept. 9, 2010

On 09/09/2010 04:09 PM, Luca Bennati wrote: > Hi Luca, You need to add the UTF-8 pragma: Warning: string "&Sì" seems to be UTF-8 but codepage 1252 is in use. Warning: string "&Più finestre..." seems to be UTF-8 but codepage 1252 is in use. -- Cheers, Paul.

1 0

Re: msxml3: Implement property SelectionNamespaces for XPath
by Alexandre Julliard Sept. 9, 2010

Sept. 9, 2010

Adam Martinson <amartinson(a)codeweavers.com> writes: > +extern inline const struct list* nsList_from_xmlDocPtr(xmlDocPtr doc); Declaring an extern function inline doesn't make much sense. Also it would be more generally useful if it didn't return a const pointer. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [2/7] urlmon/tests: Removed no longer needed todo_wine's
by Paul Vriens Sept. 9, 2010

Sept. 9, 2010

On 09/09/2010 02:46 AM, Thomas Mullaly wrote: > These todo's should have been removed as I implemented each of the > IUriBuilder_{Get/Set}* functions, but, I forgot about them. Hi Thomas, todo_wine's that succeed are marked as failures so that means these are not fixed on (at least) AJ's box as otherwise previous patches would have been rejected. Judging by the test.winehq.org results there are only todo_wine's and no failures on Wine. -- Cheers, Paul.

2 1

Re: msxml3/tests: add lastChild tests
by testbot＠testbot.winehq.org Sept. 9, 2010

Sept. 9, 2010

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=5160 Your paranoid android. === W98SE (32 bit domdoc) === domdoc.c:6040: Test failed: failed to create CLSID_DOMDocument instance: 0x80040154 === WNT4WSSP6 (32 bit domdoc) === domdoc.c:6040: Test failed: failed to create CLSID_DOMDocument instance: 0x80040154 === W2KPROSP4 (32 bit domdoc) === domdoc.c:6040: Test failed: failed to create CLSID_DOMDocument instance: 0x80040154

1 0

Re: d3dx9_36: Implement D3DXCreateMesh and initial ID3DXMesh methods. (try 2)
by Henri Verbeet Sept. 9, 2010

Sept. 9, 2010

On 9 September 2010 02:40, Misha Koshelev <misha680(a)gmail.com> wrote: > + while (count < MAX_FVF_DECL_SIZE && (count == 0 || declaration[count-1].Stream != 0xFF)) > + { > + count++; > + } > + if (count > 1) > + { > + vertex_size = declaration[count-2].Offset + d3dx_decltype_size[declaration[count-2].Type]; > + } You may want to consider using D3DXGetDeclVertexSize() there.

1 0

Re: dlls/rstrtmgr: add stubs for RmGetList and RmRegisterResources
by Paul Vriens Sept. 9, 2010

Sept. 9, 2010

On 09/09/2010 07:41 AM, Austin English wrote: > +@ stdcall RmGetList(long ptr ptr ptr) Hi Austin, This function has 5 parameters. -- Cheers, Paul.

1 0

Re: msxml3: Implement property SelectionNamespaces for XPath
by testbot＠testbot.winehq.org Sept. 9, 2010

Sept. 9, 2010

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=5150 Your paranoid android. === W98SE (32 bit domdoc) === domdoc.c:6002: Test failed: failed to create CLSID_DOMDocument instance: 0x80040154 === WNT4WSSP6 (32 bit domdoc) === domdoc.c:6002: Test failed: failed to create CLSID_DOMDocument instance: 0x80040154 === W2KPROSP4 (32 bit domdoc) === domdoc.c:6002: Test failed: failed to create CLSID_DOMDocument instance: 0x80040154

1 0

Re: shdocvw: Fixed tests on IE <6.
by testbot＠testbot.winehq.org Sept. 9, 2010

Sept. 9, 2010

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=5149 Your paranoid android. === WVISTAADM (32 bit webbrowser) === webbrowser.c:1210: Test failed: unexpected call UpdateUI webbrowser.c:289: Test failed: unexpected nsCmdID 21 webbrowser.c:233: Test failed: unexpected call QueryStatus_STOP

1 0

Re: [try 2] [1/3] dsound: Add support for 32-bit IEEE float buffers
by Chris Robinson Sept. 8, 2010

Sept. 8, 2010

On Wednesday, September 08, 2010 11:02:27 am Owen Rudge wrote: > + *(BYTE *) dst = (clamp(le32(*(const float *)src), -1, 1)) * 127 + > 0.5; Is it safe to call le32 on a float? Especially one that's going to be used more? If the system is big-endian, the float/integer will need to be in big- endian to be processed. I'm also not sure that the conversion math is correct. In this case, -1.0 would get converted to -126.5 (or -32766.5, etc), instead of -128 (or -32768, etc). As well, I don't think the rules for converting a negative float value to an unsigned integer are that straight forward. I seem to recall mention that some systems will actually give 0 in that case. You may have to go the long way and do something like: float v = *(const float*)src; if(v < -1.0f) *(__int8*)dst = -128; if(v < 0.0f) *(__int8*)dst = v*128.0f; if(v > 1.0f) *(__int8*)dst = 127; if(v >= 0.0f) *(__int8*)dst = v*127.0f; ..and similarly for 16/24/32 bit.

2 1

Mistake in : msxml3: Don't use xmlnode's IXMLDOMNode iface in get_[first|last]Node implementations.
by Joris Huizer Sept. 8, 2010

Sept. 8, 2010

Hello, I noticed some copy/paste mistake in this patch; A number of functions for *get_lastChild now call node_get_first_child. HTH, Joris

2 1

daily wine clang tests
by Austin English Sept. 8, 2010

Sept. 8, 2010

Howdy all, I've added a daily test of wine/winetest compiled using clang (tag wine_ae-ub1004-clang). I fixed the first bug it found (http://source.winehq.org/git/wine.git/?a=commitdiff;h=7f30ae6349e01095823e6…) but there are a few more failures that look interesting (interesting = crashes with clang, but not gcc): http://test.winehq.org/data/ec382f43452890354b9edd30ecb7d293045bb8c2/wine_a… http://test.winehq.org/data/ec382f43452890354b9edd30ecb7d293045bb8c2/wine_a… http://test.winehq.org/data/ec382f43452890354b9edd30ecb7d293045bb8c2/wine_a… http://test.winehq.org/data/ec382f43452890354b9edd30ecb7d293045bb8c2/wine_a… http://test.winehq.org/data/ec382f43452890354b9edd30ecb7d293045bb8c2/wine_a… http://test.winehq.org/data/ec382f43452890354b9edd30ecb7d293045bb8c2/wine_a… msi in particular doesn't seem to like clang. Anywho, if you want traces/patches tested, drop me a line. Cheers! -- -Austin

1 0

Re: [1/3] msi: Set more 64-bit properties.
by Alexandre Julliard Sept. 8, 2010

Sept. 8, 2010

Hans Leidekker <hans(a)codeweavers.com> writes: > --- > dlls/msi/package.c | 43 ++++++++++++++++++++++++++++++++----------- > 1 files changed, 32 insertions(+), 11 deletions(-) This breaks the tests on 64-bit: ../../../../wine/tools/runtest -q -P wine -M msi.dll -T ../../.. -p msi_test.exe.so ../../../../wine/dlls/msi/tests/automation.c && touch automation.ok automation.c:2505: Test failed: File not installed automation.c:2506: Test failed: File not installed automation.c:2507: Test failed: File not installed automation.c:2508: Test failed: File not installed automation.c:2509: Test failed: File not installed automation.c:2510: Test failed: File not installed automation.c:2511: Test failed: File not installed automation.c:2512: Test failed: File not installed automation.c:2513: Test failed: File not installed automation.c:2514: Test failed: File not installed automation.c:2515: Test failed: File not installed make: *** [automation.ok] Error 11 -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: user32: Right text placement (Bugfix for #22109)
by Alexandre Julliard Sept. 8, 2010

Sept. 8, 2010

Wolfgang Schwotzer <wolfgang.schwotzer(a)gmx.net> writes: > Much thanks to Konstantin Kondratyuk. He is the brain behind this bugfix. > > --- a/dlls/user32/text.c 2010-06-01 18:14:04.000000000 +0200 > +++ b/dlls/user32/text.c 2010-06-01 18:15:37.000000000 +0200 > @@ -954,8 +954,12 @@ > > if (flags & DT_SINGLELINE) > { > - if (flags & DT_VCENTER) y = rect->top + > - (rect->bottom - rect->top) / 2 - size.cy / 2; > + if (flags & DT_VCENTER) > + { > + int sign = 1; > + if (rect->top < 0 && rect->bottom < 0) sign = -1; > + y = rect->top + (rect->bottom - rect->top) / 2 - sign * size.cy / 2; This doesn't make sense, the sign should depend on the viewport mappings, not on the position being negative. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: usp10: we are unable to shape non TMPF_TRUETYPE fonts
by Alexandre Julliard Sept. 8, 2010

Sept. 8, 2010

Aric Stewart <aric(a)codeweavers.com> writes: > So return a proper error instead of generating unexpected garbage > --- > dlls/usp10/usp10.c | 5 ++++- > 1 files changed, 4 insertions(+), 1 deletions(-) It fails the tests: ../../../tools/runtest -q -P wine -M usp10.dll -T ../../.. -p usp10_test.exe.so usp10.c && touch usp10.ok usp10.c:907: Test failed: ScriptShape should return 0 not (80040200) make[1]: *** [usp10.ok] Error 1 -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: dnsapi: Remove the critical section around res_query on Linux.
by Alexandre Julliard Sept. 8, 2010

Sept. 8, 2010

Hans Leidekker <hans(a)codeweavers.com> writes: > glibc's resolver supports threads since 2.3.2, released in 2003. > Fixes http://bugs.winehq.org/show_bug.cgi?id=24180 It seems to me this would affect all platforms. Why fix it only for Linux? -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [2/3] msi/test: Test more combinations of platforms and installer properties.
by testbot＠testbot.winehq.org Sept. 8, 2010

Sept. 8, 2010

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=5136 Your paranoid android. === W7PROX64 (64 bit package) === package.c:8393: Test failed: Expected "C:\winetest\FileName1", got "" package.c:8404: Test failed: Expected "C:\winetest\", got "" package.c:8415: Test failed: Expected "C:\winetest\", got "" package.c:8465: Test failed: Expected "C:\winetest\FileName3.dll", got "" package.c:8476: Test failed: Expected "C:\winetest\FileName5.dll", got "" package.c:8485: Test failed: Expected "C:\", got "" package.c:8491: Test failed: Expected "C:\winetest\", got "" package.c:8512: Test failed: Expected "C:\winetest\FileName1", got "" package.c:8521: Test failed: Expected "C:\winetest\FileName1", got ""

1 0

Re: [PATCH 2/5] d3d8.h: undefine INTERFACE before defining it
by Alexandre Julliard Sept. 8, 2010

Sept. 8, 2010

Roderick Colenbrander <thunderbird2k(a)gmail.com> writes: > @@ -129,6 +129,7 @@ typedef struct IDirect3DVolumeTexture8 *LPDIRECT3DVOLUMETEXTURE8, *PDIRECT3DVOLU > /***************************************************************************** > * IDirect3D8 interface > */ > +#undef INTERFACE > #define INTERFACE IDirect3D8 > DECLARE_INTERFACE_(IDirect3D8,IUnknown) > { The standard pattern is that it's undefined right after being used. Places that don't do that should be fixed. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: Ongoing Debian package maintenance.
by Ben Klein Sept. 8, 2010

Sept. 8, 2010

Sorry, Octavian, for sending this only to you. On 8 September 2010 06:00, Octavian Voicu <octavian.voicu(a)gmail.com> wrote: > On Tue, Sep 7, 2010 at 9:42 PM, Lei Zhang <thestig(a)google.com> wrote: >> >> It would be helpful if you provide the content of your debian/ >> directory. Mainly you want my control and rules files. I'll get these to you soon :) >> I wrote my own debian/control file and what not and built a >> wine-gecko-1.1.0 .deb over the weekend. I haven't finished working on >> the main Wine packages, but I'll take a look at your suggestions and >> reconsider how to structure the packages. As much as the final structure for the new packages is up to the official maintainer, I'm very interested in that decision and the reasoning behind it. Keep me posted! > It would be cool to also have wine1.3-64, wine1.3-gecko64 for amd64 builds, > besides the standard wine1.3, wine1.3-gecko :) 1) I do not like those package names at all. It should be the Debian-style names, wine-unstable/wine64-unstable. 2) Wine Gecko has its own versions. I had previously built a package called wine-gecko-1.0, IIRC. This naming convention should be continued, IMO, as different Gecko versions could be required at different points for wine (stable) and wine-unstable versions. Package names should be uniform and I suggest they should look like this: wine[64][-unstable][-<module>] where <module> is one of: * dbg * dev * gecko etc. > Octavian >

1 0

Re: [PATCH 2/9] CMD : Correct support for cmd.exe /C string [try 2]
by Alexandre Julliard Sept. 8, 2010

Sept. 8, 2010

Jason Edmeades <jason.edmeades(a)googlemail.com> writes: > Note1: Tested manually on windows with procmon, windows does indeed do > the same as this patch, ie a double path/pathext search. You can't look at the internal behavior of the tool you are implementing, it needs to be treated as a black box. I'm afraid I can't accept that code. -- Alexandre Julliard julliard(a)winehq.org

1 0

Importing strchrnul() from gnulib?
by Wolfram Sang Sept. 8, 2010

Sept. 8, 2010

Hi, taking another approach at Bug 19070, I noticed that dlls/hhctrl.ocx/hhctrl.c has even more problems with commandline-handling. All three invocations of strchr() should better be strchrnul(). (Compare calling hh.exe with '-mapid' and '-mapid ' to see one difference). strchrnul ist protected by _GNU_SOURCE, though. Because I could imagine there are more places throughout the source needing a fix like this, I think about importing strchrnul() directly from gnulib into libs/wine/string.c (most appropriate place I could find). The function is LGPLv2+, of course. Objections? Regards, Wolfram

2 1

Re: [PATCH 2/2] d3dx9: Implement D3DXCreateMesh and initial ID3DXMesh methods.
by Henri Verbeet Sept. 8, 2010

Sept. 8, 2010

You have leaks in your error paths. On 8 September 2010 02:33, Misha Koshelev <misha680(a)gmail.com> wrote: > + /* Create vertex buffer */ > + hr = IDirect3DDevice9_CreateVertexBuffer(device, > + numvertices * sizeof(D3DXVECTOR3) * 2, > + 0, /* Usage */ > + fvf, > + D3DPOOL_MANAGED, > + &vertex_buffer, > + NULL); Tests for that would be nice. E.g., I don't believe for a second the buffer length is really calculated that way, and does it really create an FVF vertex buffer?

1 0

Re: [3/3] msi/tests: Fix some test failures on 64-bit.
by testbot＠testbot.winehq.org Sept. 8, 2010

Sept. 8, 2010

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=5126 Your paranoid android. === W7PROX64 (64 bit package) === package.c:8361: Test failed: Expected "C:\Program Files\VMware\VMware Tools\;C:\Windows\system32;C:\Windows;C:\Windows\System32\Wbem;C:\Windows\System32\WindowsPowerShell\v1.0\", got "C:\Program Files (x86)\VMware\VMware Tools\;C:\Windows\system32;C:\Windows;C:\Windows\System32\Wbem;C:\Windows\System32\WindowsPowerShell\v1.0\"

1 0

Re: [PATCH 4/5] ddraw: Remove some stray tabs.
by testbot＠testbot.winehq.org Sept. 8, 2010

Sept. 8, 2010

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=5122 Your paranoid android. === W98SE (32 bit dsurface) === No test summary line found

2 1

mshtml: Correct test for unknown dispID
by Alistair Leslie-Hughes Sept. 8, 2010

Sept. 8, 2010

Changelog: mshtml: Correct test for unknown dispID Best Regards Alistair Leslie-Hughes

1 0