November 2008 - Wine-Devel - list.winehq.org

Re: shdocvw: Added Support for the IViewObjectEx interface
by Jacek Caban Nov. 11, 2008

Nov. 11, 2008

Alistair Leslie-Hughes wrote: > Hi, > > Changelog: > shdocvw: Added Support for the IViewObjectEx interface Please check if interface you add is really supposed to be there. It's not the first time someone tries to add an interface to WebBrowser or HTMLDocument that shouldn't be there (this time the patch is already in Git). Please remember that the fact that an app queries for interface doesn't mean it should success! I keep adding tests for such cases, but it really easy to tests it before sending patch and it's something you *should* do. Jacek

2 2

Wine and relocation on not aligned section
by Alessandro Nov. 11, 2008

Nov. 11, 2008

Hi list, while diving into wine sources i've noticed that wine does not apply relocation to section not aligned to page boundary. Is there any reason for this behaviour? If it is wrong i've put up a little patch that solves the issue. Greetings, Alessandro Pignotti -- Vi Veri Veniversum Vivus Vici -Dr. Faustus - Marlowe Public GPG Key ID 0x650B3ED9 on subkeys.gpg.net Key Fingerprint 6243 AAD3 E3EC 52D8 DFAA 2A2F 9FCD 0457 650B 3ED9 Encrypted mails are welcome

1 0

Re: d3dx9_38.dll: Add the dll.
by Tony Wasserka Nov. 11, 2008

Nov. 11, 2008

Hi, > +#include "config.h" > +#include "wine/port.h" > + > +#include <stdarg.h> > + > +#include "windef.h" > +#include "winbase.h" > +#include "wingdi.h" > +#include "winuser.h" > +#include "wine/debug.h" > + > +#include "d3dx9.h" There's no point in including so many unneccessary header files. Also, we have the year 2008, so you may want to update the Copyright ;-)

1 0

Re: winspool.drv/tests: Fix a test on all platforms.
by Nicolas Le Cam Nov. 11, 2008

Nov. 11, 2008

2008/11/6 Nicolas Le Cam <niko.lecam(a)gmail.com>: > Hi, > > DeviceCapabilities DC_FIELDS always sets the DM_FORMNAME flag. > Mark this test as a todo_wine. > > Nicolas Le Cam Is there something wrong with this one ? Should I fix wine too ? I didn't because I don't know where to fix it (in winspool.drv or in gdi32). There's no tests for this part of gdi32 and I really don't know if windows' winspool.drv calls gdi32 DeviceCapabilities function. Could someone familiar with this part can help me ? Nicolas Le Cam

1 0

Re: mlang: Implement IMultiLanguage2::ConvertStringInIStream.
by Dmitry Timoshkov Nov. 10, 2008

Nov. 10, 2008

"Hans Leidekker" <hans(a)codeweavers.com> wrote: > + while (1) > + { > + hr = IStream_Read(pstmIn, src, sizeof(src), &bytes_read); > + if (FAILED(hr)) goto exit; > + > + srclen = bytes_read; > + dstlen = buffer_size; > + ConvertINetString(pdwMode, dwSrcEncoding, dwDstEncoding, src, &srclen, NULL, &dstlen); > + if (dstlen > buffer_size) > + { > + dst = HeapReAlloc(GetProcessHeap(), 0, dst, dstlen); > + if (!dst) return E_OUTOFMEMORY; > + buffer_size = dstlen; > + } > + > + hr = ConvertINetString(pdwMode, dwSrcEncoding, dwDstEncoding, src, &srclen, dst, &dstlen); > + if (FAILED(hr)) goto exit; > + > + hr = IStream_Write(pstmOut, dst, dstlen, NULL); > + if (FAILED(hr)) goto exit; > + > + if (bytes_read != sizeof(src)) break; > + } > + > +exit: 'break' instead of 'goto' would be cleaner IMO. -- Dmitry.

1 0

Re: [ole32/tests] Don't crash on NT4
by Dmitry Timoshkov Nov. 10, 2008

Nov. 10, 2008

"Paul Vriens" <paul.vriens.wine(a)gmail.com> wrote: > static HENHMETAFILE create_emf(void) > { > RECT rect = {0, 0, 100, 100}; > + const RECT rect2 = {0, 0, 100, 100}; > HDC hdc = CreateEnhMetaFile(NULL, NULL, &rect, "HENHMETAFILE Marshaling Test\0Test\0\0"); > - ExtTextOut(hdc, 0, 0, ETO_OPAQUE, NULL, "Test String", strlen("Test String"), NULL); > + ExtTextOut(hdc, 0, 0, ETO_OPAQUE, &rect2, "Test String", strlen("Test String"), NULL); Is there any reason that you introduced rect2 instead of using existing rect? -- Dmitry.

1 0

valgrind warning in "advapi32/tests: Add tests for QueryServiceStatusEx."
by Dan Kegel Nov. 10, 2008

Nov. 10, 2008

Hi Paul, in your change of 21 Aug 2008, http://source.winehq.org/git/wine.git/?a=commit;h=794752200131184506e84939a… valgrind has found something to complain about: ==7145== Conditional jump or move depends on uninitialised value(s) ==7145== at 0x4734514: RtlAllocateHeap (heap.c:1357) ==7145== by 0x496ECF3: test_query_svc (service.c:896) ==7145== by 0x497142C: func_service (service.c:2079) ==7145== by 0x4975C17: run_test (test.h:452) ==7145== by 0x4975E18: main (test.h:502) ==7145== Uninitialised value was created by a stack allocation ==7145== at 0x496E7E1: test_query_svc (service.c:808) Could you have a look? Thanks, Dan

1 0

Please, look at my patch, I'm not sure it's correct
by Vitaly Perov Nov. 10, 2008

Nov. 10, 2008

Hi, This patch add a stub implementation of IMarshal interface. I have no experience at implementing interfaces, so I'm not sure it's implemented correctly. Please, look at this patch. I'll be grateful for any remarks. Thank you. -- Best wishes, Vitaly Perov Russia, Saint-Petersburg. www.etersoft.ru

2 2

Re: Solution for bug 15942 - Send page size with lpr command on printing
by Austin English Nov. 10, 2008

Nov. 10, 2008

On Sun, Nov 9, 2008 at 6:48 PM, Massimo Del Fedele <max(a)veneto.com> wrote: > Send page size with 'lpr' command to CUPS, so it doesn't take always the > default one. > Solves bug 15942. > > Max > > > > +// sprintf(psCmd,"|lpr -P'%s'",pszOutput+4); Don't use C++ comments, use /* comment */ instead. Also, don't comment out code. Remove it if unneeded, or if it should be left as an example, etc., use if(0). -- -Austin

3 2

winehq packages for Debian
by Ben Klein Nov. 10, 2008

Nov. 10, 2008

Hi guys, I'm a regular volunteer in the #winehq IRC channel (man_in_shack). Some of you may know me :) I also maintain a few applications on AppDB and have had a couple of bugs found and fixed via bugzilla posts. I've been building Wine packages for my Debian systems for a while and now I've written myself some helper scripts to make the task trivial :) I build both amd64 and i386 packages. I'd like to request that I take over Scott Ritchie's position as maintainer for the winehq Debian packages as he seems to have all but abandoned Debian package building. I have the following caveats: 1) I don't have a suitable place to upload the packages (in terms of bandwidth restrictions) 2) At the moment, I'm not sure if the packages are suitable for an apt repository, as I don't have proper Changelogs or source packages. I'm not entirely sure what I'd need to do for this, so help would be appreciated 3) I'm fully set up to produce etch packages (Scott's packages aren't etch-compatible any more) as well as packages for lenny and sid 4) I'm not asking to take over Ubuntu package building (Scott seems to be good at this, and I don't have nor want an Ubuntu system) 5) I can be *very* quick to produce packages, as in the day after release (or even building packages from non-release revisions in git if required) Involvement in wine is fun! Regards, Ben Klein

3 4

Another shot at running tests in parallel
by Dan Kegel Nov. 10, 2008

Nov. 10, 2008

On modern CPUs, with a 30 second timeout per test, the wine conformance test suite runs in about 4-5 minutes. That's wonderfully fast, completely sufficient for all purposes... ... except one: running tests under valgrind, which imposes a roughly 10x penalty. 40-50 minutes to run the tests is a bit too long to expect developers to wait for feedback. So I had another try at getting conformance tests to run in parallel, this time without any code modifications. I use WINETEST_WRAPPER="$HOME/alarm 30", where alarm is the output of compiling http://winezeug.googlecode.com/svn/trunk/patchwatcher/alarm.c That file has a hardcoded whitelist of tests that are know to not call CreateWindow (directly or indirectly) or do anything else that might not work well in parallel. alarm takes an exclusive lock when running any test not on that list. Even with this fairly restrictive whitelist, the tests were fairly parallelizable (see below). Because running tests in parallel is tricky, and makes our already somewhat shaky test suite less stable, I will run the tests the normal way first, and only if they pass without regressions will I then do a second run under Valgrind. And I'll ignore any test failures that happen under valgrind; all I care about in those runs are the valgrind warnings. Here are some benchmarks on a quad processor: -j time in seconds 1 255 2 155 3 125 4 105 20 79 With -j4, it looks like valgrind could run the tests in about ten minutes, which would be quick enough. It would be nice if our test suite really ran in parallel some day, but the approximation I have now is probably good enough. - Dan p.s. Out of curiosity, I checked how many tests interfere with themselves if multiple copies are run. To do this, I ran with WINETEST_WRAPPER="$HOME/foowrap.sh" where foowrap.sh was #!/bin/sh # Run same test three times in parallel $@ & sleep 1 $@ & $@ As you can imagine, that broke a lot of tests, many more than just running in parallel.

1 0

Re: Solution for bug 15942 - Send page size with lpr command on printing
by Austin English Nov. 10, 2008

Nov. 10, 2008

On Sun, Nov 9, 2008 at 6:48 PM, Massimo Del Fedele <max(a)veneto.com> wrote: > Send page size with 'lpr' command to CUPS, so it doesn't take always the > default one. > Solves bug 15942. > > Max > > > > +// sprintf(psCmd,"|lpr -P'%s'",pszOutput+4); Don't use C++ comments, use /* comment */ instead. Also, don't comment out code. Remove it if unneeded, or if it should be left as an example, etc., use if(0). -- -Austin

1 0

.ANI patch
by Krzysztof Drewniak Nov. 9, 2008

Nov. 9, 2008

There is a .ANI cursor patch in hacks.git. Would it be possible to integrate it into wine? I know of many applications (especcially games) that would benefit. Thanks Krzys D.

2 3

Wine and relocation on not aligned section
by Alessandro Nov. 9, 2008

Nov. 9, 2008

Hi list, while diving into wine sources i've noticed that wine does not apply relocation to section not aligned to page boundary. Is there any reason for this behaviour? If it is wrong i've put up a little patch that solves the issue. Greetings, Alessandro Pignotti -- Vi Veri Veniversum Vivus Vici -Dr. Faustus - Marlowe Public GPG Key ID 0x650B3ED9 on subkeys.gpg.net Key Fingerprint 6243 AAD3 E3EC 52D8 DFAA 2A2F 9FCD 0457 650B 3ED9 Encrypted mails are welcome

1 0

Re: [PATCH 1/2] Implement GdipGetLogFontA (try2)
by Dmitry Timoshkov Nov. 9, 2008

Nov. 9, 2008

"Adam Petaccia" <adam(a)tpetaccia.com> wrote: > +GpStatus WINGDIPAPI GdipGetLogFontA(GpFont *font, GpGraphics *graphics, > + LOGFONTA *lfa) > +{ > + INT bytesWritten; > + /* FIXME: use graphics */ > + if(!(font && graphics && lfa)) > + return InvalidParameter; > + > + memcpy(lfa, &font->lfw, sizeof(LOGFONTA)); FIELD_OFFSET(LOGFONTA, lfFaceName) instead of sizeof(LOGFONTA) would be better IMO. > + bytesWritten = WideCharToMultiByte(CP_ACP, 0, font->lfw.lfFaceName, -1, > + lfa->lfFaceName, LF_FACESIZE, NULL, NULL); > + > + if (bytesWritten == 0) > + { > + ERR("Could not convert from WideChar to Multibyte with error %d\n", > + GetLastError()); > + return GenericError; > + } The ERR() as well as the bytesWritten check are useless, please remove them. -- Dmitry.

1 0

fixme: IDirect3DDevice9::CreateQuery
by Krzysztof Drewniak Nov. 9, 2008

Nov. 9, 2008

I'm getting a "fixme:d3d:IWineD3DDeviceImpl_EvictManagedResources (0xb550510) : stub fixme:d3d:IWineD3DDeviceImpl_CreateQuery (0xb550510) Unhandled query type 4 fixme:d3d9:IDirect3DDevice9Impl_CreateQuery (0xb543bc8) call to IWineD3DDevice_CreateQuery failed" over and over again and the app isn't working right. Either can someone fix this of drop me a line about what might need to be done? (Not a Wine expert, but I'll gladly help.) Krzys D

2 1

Re: [PATCH] Implement GdipGetLogFontA
by Adam Petaccia Nov. 9, 2008

Nov. 9, 2008

Ignore. Will resend. On Sun, 2008-11-09 at 13:03 -0500, Adam Petaccia wrote: > --- > dlls/gdiplus/font.c | 25 +++++++++++++++++++++++++ > dlls/gdiplus/gdiplus.spec | 2 +- > 2 files changed, 26 insertions(+), 1 deletions(-) > > diff --git a/dlls/gdiplus/font.c b/dlls/gdiplus/font.c > index 32713c7..2477f7b 100644 > --- a/dlls/gdiplus/font.c > +++ b/dlls/gdiplus/font.c > @@ -368,6 +368,31 @@ GpStatus WINGDIPAPI GdipGetFontUnit(GpFont *font, Unit *unit) > } > > /******************************************************************************* > + * GdipGetLogFontA [GDIPLUS.@] > + */ > +GpStatus WINGDIPAPI GdipGetLogFontA(GpFont *font, GpGraphics *graphics, > + LOGFONTA *lfa) > +{ > + INT bytesWritten; > + /* FIXME: use graphics */ > + if(!(font && graphics && lfa)) > + return InvalidParameter; > + > + memcpy(&font->lfw, lfa, sizeof(LOGFONTA)); > + bytesWritten = WideCharToMultiByte(CP_ACP, 0, font->lfw.lfFaceName, -1, > + lfa->lfFaceName, LF_FACESIZE, NULL, NULL); > + > + if (bytesWritten == 0) > + { > + ERR("Could not convert from WideChar to Multibyte with error %d\n", > + GetLastError()); > + return GenericError; > + } > + > + return Ok; > +} > + > +/******************************************************************************* > * GdipGetLogFontW [GDIPLUS.@] > */ > GpStatus WINGDIPAPI GdipGetLogFontW(GpFont *font, GpGraphics *graphics, > diff --git a/dlls/gdiplus/gdiplus.spec b/dlls/gdiplus/gdiplus.spec > index a2790a3..0e7ac5c 100644 > --- a/dlls/gdiplus/gdiplus.spec > +++ b/dlls/gdiplus/gdiplus.spec > @@ -311,7 +311,7 @@ > @ stdcall GdipGetLineSpacing(ptr long ptr) > @ stub GdipGetLineTransform > @ stdcall GdipGetLineWrapMode(ptr ptr) > -@ stub GdipGetLogFontA > +@ stdcall GdipGetLogFontA(ptr ptr ptr) > @ stdcall GdipGetLogFontW(ptr ptr ptr) > @ stdcall GdipGetMatrixElements(ptr ptr) > @ stub GdipGetMetafileDownLevelRasterizationLimit

1 0

Re: mshtml: Implement IHTMLStyle get/put posLeft (try 9)
by Austin English Nov. 8, 2008

Nov. 8, 2008

On Fri, Nov 7, 2008 at 10:53 PM, Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> wrote: > Hi, > Use strtolW correctly. > > Changelog: > mshtml: Implement IHTMLStyle get/put posLeft > > Best Regards > Alistair Leslie-Hughes > > > > > > You forgot to attach the patch :-) -- -Austin

1 0

wine build time: 6 (or 3) minutes with q9400
by Dan Kegel Nov. 8, 2008

Nov. 8, 2008

FWIW: with wine from git from a couple days ago, here are wine build times on various Intel CPUs in seconds on Ubuntu 8.10's gcc-4.3.2 (or 8.04): sec cpu wine notes 676 e7200 June http://kegel.com/new-computer-2008.html ubuntu 8.04 615 e7300 Nov 585 e7400 Nov (estimated, supposedly 5% faster than e7300) 540 e8400 Nov 360 q9400 Nov In all cases, this was with the default optimization level, -O2. Compiling with -O0 was roughly twice as fast (e.g. 196 seconds to build on the q9400). The dual core e7300 and e8400 were tested with -j3; the quad core q9400 with -j5. The current crop of CPUs is that they run cooler than older ones. Replacing the old Athlon 64's in the patchwatcher cluster with new CPUs and 80+ power supplies makes a big difference in the room's temperature! The quad CPUs are coming down in price. Once you get used to fast builds, it's hard to go back, so don't try a q9400 or faster CPU unless you want to keep it :-) Older benchmarks: http://techgage.com/article/intel_core_2_quad_q6600/4 (is gentoo really that fast, or was he building with -O0?) http://www.winehq.org/pipermail/wine-devel/2008-June/066121.html (q6600 386 seconds).

3 5

Re: [1/2] richedit: Created initial tests for windowless richedit controls. (try 2)
by Peter Oberndorfer Nov. 8, 2008

Nov. 8, 2008

On Samstag 08 November 2008, Dylan Smith wrote: > (Ensured compiler independence by modifying byte code templates) > > The initial tests added simply call CreateTextServices, and query for > the ITextServices interface, then cleans up. > > Since CreateTextServices needs an implementation of ITextHost to be > given, this patch needed a stub implementation. This required wrappers > for converting thiscall to stdcall which are now implementing by > adapting a byte code wrapper template. Although byte code isn't portable > across processors, this is only needed for x86 platforms so it is only > meant to be compiler independent. > > In order to investigate what is happening in the tests, it is helpful > to trace the calls to ITextHost methods by the richedit controls. I > realized it would be too verbose to trace all these calls by default, > so I wrapped them all with a check for winedebug > 1, since normally > winedebug is 0 or 1. > --- > dlls/riched20/tests/Makefile.in | 3 +- > dlls/riched20/tests/txtsrv.c | 593 +++++++++++++++++++++++++++++++++++++++ > 2 files changed, 595 insertions(+), 1 deletions(-) > create mode 100644 dlls/riched20/tests/txtsrv.c > Hi, This page [1] seems to suggest memory returned from HeapAlloc is marked as not executable. Other tests (ntdll exception) are using VirtualAlloc() for dynamically created code. patch 1 > +/* The following x86 code string converts between the thiscall and stdcall > + * calling convetions. The thiscall calling convention places the This > + * pointer in ecx on the x86 platform, and the stdcall calling convention > + * pushes the This pointer on the stack as the first argument. > + * > + * The wrapper's code finishes by jumping to the real function. > + * > + * Byte codes are used so that a copy of it can be modified to use > + * for each method in ITextHost. */ patch 2 > -/* The following x86 code string converts between the thiscall and stdcall > +/* The following x86 code strings convert between the thiscall and stdcall > * calling convetions. The thiscall calling convention places the This Here you introduce comments that you seem to change in the second patch. Maybe use the right text in the first patch? Additionally the term code string sounds odd. Maybe replace that with "The following x86 assembler code..." or something similar? > #define ITextServices_QueryInterface(p,a,b) (p)->lpVtbl->QueryInterface(p,a,b) > #define ITextServices_AddRef(p) (p)->lpVtbl->AddRef(p) > #define ITextServices_Release(p) (p)->lpVtbl->Release(p) > -/*** ITextServices methods ***/ > -#define ITextServices_TxSendMessage(p,a,b,c,d) (p)->lpVtbl->TxSendMessage(p,a,b,c,d) > -#define ITextServices_TxDraw(p,a,b,c,d,e,f,g,h,i,j,k,l) (p)->lpVtbl->TxDraw(p,a,b,c,d,e,f,g,h,i,j,k,l) Why do you only move some of those #defines part of ITextServices and not the one part of IUnknown ? I do not know much about those #defines but this seems odd Maybe i am missing something but why do you need those thiscall wrappers? Wouldn't it be possible to use the DEFINE_THISCALL_WRAPPER and THISCALL macros like in txtserv.c? Greetings Peter [1] http://www.unixwiz.net/techtips/xp-sp2.html

2 1

noob req: ddraw/vertexbuffer.c and HW vertex buffers
by FeepingCreature Nov. 8, 2008

Nov. 8, 2008

Hi, I've been playing Gothic2 and getting lots of messages regarding d3d7:IDirect3DVertexBufferImpl_Optimize, printed by ddraw/vertexbuffer.c. Reading the comments for that function seems to indicate that vertex buffers are currently emulated in software, and will be handled in hardware as soon as wined3d is capable of that - however, reading wined3d/vertexbuffer.c, especially CreateVBO, implies that hardware/GL vertex buffers are already supported. Git logs also show that ddraw/vertexbuffer.c has not been changed in a few months. Did I completely mis-interpret the comments, or has this file simply been forgotten? And if so, could some kind soul sacrifice the time to bring it up to speed with the rest of Wine, so that us Gothic players can enjoy better framerates? Thanks for reading. ___________________________________________________________ Telefonate ohne weitere Kosten vom PC zum PC: http://messenger.yahoo.de

2 1

integration of winscard patches: ideas for implementing test cases?
by frechdachs69 Nov. 7, 2008

Nov. 7, 2008

Hi together! I am currently wondering which test cases make sense after an integration of winscard patches. So far most API calls are directly related to pcsc-lite behaviour and therefore we need a result from pcsc-lite which is comparable with what Win/winscard.dll would return to us. But this is almost always depending on the smartcard reader setup at the local machine. As a result I see a problem to automate such tests without having underlying assumptions concerning the attached smartcard readers and maybe inserted (or not) smartcards (i.e. specific hardware is necessary). Any ideas? Does anyone know whether someone has written a virtual card reader which can be accessed using pcsc-lite (i.e. an IFD handler exists which simulates reader behaviour)? Maybe together with the option to insert a virtual card? If not then I am going to subscribe to the pcsc-lite maling list and ask this question there, too. Any other ideas how to automate winscard tests to be able to add such tests to the regression test suite? Regards Heiko

2 1

Re: inetcomm: Add an implementation of IVirtualStream.
by Huw Davies Nov. 7, 2008

Nov. 7, 2008

Hans Leidekker wrote: > This interface is undocumented but fortunately its name was enough > of a hint to create a sensible implementation. Used by Outlook when > retrieving or sending a mail. We already have MimeOleCreateVirtualStream() which is implemented on top of CreateStreamOnHGlobal()... Huw.

2 2

Issues with VMware Workstation 6.5.0, winetest and directx acceleration enabled
by Paul Vriens Nov. 7, 2008

Nov. 7, 2008

Hi, FYI. I had to disable "Accelerate 3D graphics (DirectX 9.0c)" in the VM settings on several of my Vmware boxes. Winetest crashed my W2K and Vista box (ddraw:d3d) and made my Win98 box unresponsive (user32:sysparams when changing to 8bit display). If I can find some spare time I will report this to VMware. I do however run VMware Workstation on Fedora 8 which is not officially supported. -- Cheers, Paul.

2 1

[RFC] winecfg -- approach to making the theme names & colours lists sortable
by Reece Dunn Nov. 7, 2008

Nov. 7, 2008

Hi, At the moment, if you have several themes installed on your system (e.g. by installing StyleXP) winecfg will list those themes in a random (read not sorted) order. If I just add the CBS_SORT flag, the wrong theme is chosen because the order in the combo boxes are assumed to be the same order in a DSA (dynamic array) containing the theme, size or colour data. My idea is to store the additional data in the combo box in the item's item data. This will mean a major rewrite of this area to remove the use of the DSAs, so I am wondering what the best way to approach this is. I am currently thinking something along the lines of: 1. Replace the DSA for the theme colour list, storing the data in the item data of the colour list combo box; 2. Replace the DSA for the theme size list, storing the data in the item data of the size list combo box; 3. Replace the DSA for the theme list, storing the data in the item data of the theme list combo box; 4. Make the theme list and theme colour lists sorted. Does this approach seem reasonable? - Reece

1 0