September 2006 - Wine-devel

How to deal with both git trees
by Paul Vriens 04 Sep '06

04 Sep '06

Hi, I've sent in a patch that got some remarks by Juan Lang. The patch is already applied to mmbranch. Most patches sent to wine-patches that have (valid) comments will not be applied by AJ. Should I now resend to make sure it gets applied when AJ is back AND sent a patch to fixup mmbranch. Or will the previous one be reverted and the next applied to mmbranch? Cheers, Paul.

2 1

Build problems with latest mmbranch
by Paul Vriens 04 Sep '06

04 Sep '06

Hi, 'make depend && make' gives: make[2]: Entering directory `/wine/wine-git/dlls/opengl32' gcc -c -I. -I. -I../../include -I../../include -D__WINESRC__ -D_REENTRANT -fPIC -Wall -pipe -fno-strict-aliasing -gstabs+ -Wdeclaration-after-statement -Wwrite-strings -Wpointer-arith -g -O2 -o wgl.o wgl.c wgl.c:104: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘p_glXGetProcAddressARB’ wgl.c: In function ‘wglGetProcAddress’: wgl.c:325: error: ‘p_glXGetProcAddressARB’ undeclared (first use in this function) wgl.c:325: error: (Each undeclared identifier is reported only once wgl.c:325: error: for each function it appears in.) wgl.c:341: warning: implicit declaration of function ‘p_glXGetProcAddressARB’ wgl.c:341: warning: assignment makes pointer from integer without a cast wgl.c: At top level: wgl.c:912: error: expected declaration specifiers or ‘...’ before ‘glXGetProcAddressARB_t’ wgl.c: In function ‘wgl_initialize_glx’: wgl.c:964: warning: implicit declaration of function ‘proc’ wgl.c:964: warning: assignment makes pointer from integer without a cast wgl.c:965: warning: assignment makes pointer from integer without a cast wgl.c:966: warning: assignment makes pointer from integer without a cast wgl.c:971: warning: assignment makes pointer from integer without a cast wgl.c:972: warning: assignment makes pointer from integer without a cast wgl.c:973: warning: assignment makes pointer from integer without a cast wgl.c:985: warning: assignment makes pointer from integer without a cast wgl.c:988: warning: assignment makes pointer from integer without a cast wgl.c:989: warning: assignment makes pointer from integer without a cast wgl.c: In function ‘process_attach’: wgl.c:1066: error: ‘p_glXGetProcAddressARB’ undeclared (first use in this function) wgl.c:1086: error: too many arguments to function ‘wgl_initialize_glx’ make[2]: *** [wgl.o] Error 1 make[2]: Leaving directory `/wine/wine-git/dlls/opengl32' make[1]: *** [opengl32] Error 2 make[1]: Leaving directory `/wine/wine-git/dlls' make: *** [dlls] Error 2 Any idea? Cheers, Paul.

2 1

Setupapi [0/6] : ReactOS setupapi patches
by Hervé Poussineau 03 Sep '06

03 Sep '06

Hello, As setupapi merge from ReactOS has been a recurrent demand [1], I'm doing this set of patches. Note that I'm not especially interested by this merge, but I do it only to please people. I've already spent one day to do them, I don't want to loose more of my time [2]. Note also that I'm not the only one who can provide better patches if those ones don't fullfill requirements. You're welcome (yes, you! the reader of Wine-patches/Wine-devel ML) to improve them if needed. The goal of this set of patches is to set a solid ground to implement lots of missing functions in setupapi. I've tried to cut them in smallest possible parts. Additionnaly, this patchset implements SetupDiEnumDeviceInfo and (mostly) SetupDiGetClassDevsExW Patch 1/6 Move Class related fonctions to their own file devclass.c No code change. Patch 2/6 Forward SetupDiGetClassDevsA/W and SetupDiGetClassDevsExA to SetupDiGetClassDevsExW This patch depends of patch 1 Patch 3/6 Add in-memory structures needed to implement SetupDiGetClassDevsExW in a following patch. Basically, we have one list container (structure DeviceInfoSet) which will contain multiple devices (structure DeviceInfoElement) in a chained list Patch 4/6 Update SetupDiCreateDeviceInfoList and SetupDiCreateDeviceInfoListA/W to initialize all fields of in-memory structure This patch depends of patch 3 Patch 5/6 Check parameters in SetupDiGetClassDevsExW stub, and create device set if needed. Now, only the real work needs to be done. This patch depends of patches 1, 3 and 4 Patch 6/6 Implement SetupDiEnumDeviceInfo This patch depends of patches 3 and 4 Regards, Hervé, ReactOS developper [1] http://marc.theaimsgroup.com/?l=wine-devel&m=115721353431733&w=2 http://marc.theaimsgroup.com/?l=wine-devel&m=113584611128655&w=2 [2] Because I'm not a Wine user, nor have Linux on my computer

1 0

Re: [wintrust] Register WINTRUST_ACTION_GENERIC_VERIFY_V2
by Juan Lang 03 Sep '06

03 Sep '06

Hi Paul, +static LONG WINTRUST_WriteSingleUsageEntry(CHAR* OID, Any reason that can't be LPCSTR OID instead? That will remove some cast-qual warnings I think. + WCHAR UsageKey[MAX_PATH]; (snip) + /* Turn OID into a wide-character string */ + Len = MultiByteToWideChar( CP_ACP, 0, OID, -1, NULL, 0 ); + OIDW = HeapAlloc( GetProcessHeap(), 0, Len * sizeof(WCHAR) ); + MultiByteToWideChar( CP_ACP, 0, OID, -1, OIDW, Len ); + lstrcatW(UsageKey, OIDW); This could overflow UsageKey. It would be better to allocate UsageKey as sizeof(Trust) + sizeof(Usages) + Len * sizeof(WCHAR), then concatenate the strings into it. I'm not sure whether there's a maximum length for a registry key, but it's certainly longer than MAX_PATH if it exists. + if (Res != ERROR_SUCCESS) goto error_close_key; + +error_close_key: + RegCloseKey(Key); This goto is unnecessary, the target of the branch is the next instruction anyway. Thanks, --Juan __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

1 0

Could I get some permissions for Bugzilla?
by Matt Finnicum 03 Sep '06

03 Sep '06

Hello, I'm not really sure what levels of permissions Bugzilla has, but could I get the ability to modify bugs? (things like assign them to myself, mark them as fixed, etc). If it's possibile, I suppose only having such access to the "wine-richedit" component would be fine, too. --Matt Finnicum

2 1

How do I automate the registering of a DLL
by Paul Vriens 03 Sep '06

03 Sep '06

Hi, while I'm working on DllRegisterServer for wintrust, I wanted to make it 'autorun' when running on a clean system (through wineprefixcreate). My thought was to add an entry to the RegisterDllsSection part in tools/wine.inf but that doesn't seem to work. (regsvr32 wintrust.dll does work). What am I missing? Cheers, Paul.

1 1

Re: Wine release 0.9.20
by Roland Schama 03 Sep '06

03 Sep '06

Hello Alexandre, Is your group affiliated with "CodeWeavers"? They claim to have a "wine" product called Crossover Mac that will have a final release soon. If you are not the same group, what are the differences? Thanks, Roland Schama On Aug 24, 2006, at 11:05 AM, Alexandre Julliard wrote: > This is release 0.9.20 of Wine, a free implementation of Windows on > Unix. > > What's new in this release: > - XEmbed system tray support. > - Many improvements to NTLM support. > - Many messages made localizable instead of hardcoded to English. > - Improved support for various OpenGL platforms. > - More improvements to the IDL compiler. > - Lots of bug fixes. > > Because of lags created by using mirrors, this message may reach you > before the release is available at the public sites. The sources will > be available from the following locations: > > http://ibiblio.org/pub/linux/system/emulators/wine/ > wine-0.9.20.tar.bz2 > http://prdownloads.sourceforge.net/wine/wine-0.9.20.tar.bz2 > > Binary packages for various distributions will be available from: > > http://www.winehq.org/site/download > > You will find documentation on > > http://www.winehq.org/site/documentation > > You can also get the current source directly from the git or CVS > repositories. Check respectively http://www.winehq.org/site/git or > http://www.winehq.org/site/cvs for details. > > Patches should be submitted to "wine-patches(a)winehq.org". Please > don't forget to include a ChangeLog entry. If you submitted a patch, > please check to make sure it has been included in the new release. > > Wine is available thanks to the work of many people. See the file > AUTHORS in the distribution for the complete list. > > -- > Alexandre Julliard > julliard(a)winehq.org >

5 4

FC4 issue with 0.9.20
by David Anderson 03 Sep '06

03 Sep '06

It's not anything the list would concern itself with, and I've not had a chance to look at it yet... I'm just sending this as an FYI. Any suggestions would of course be welcome. the Fedora Core 4 install of the (new to FC4 rpms) wine-core-0.9.20-1.fc4 gets [various messages deleted] fixme:ole:OleLoadPictureEx (0x94fbac,167,1,{7bf80980-bf32-101a-8bbb-00aa00300cab},x=0,y=0,f=0,0x33fab8), partially implemented. fixme:ole:OLEPictureImpl_LoadGif Trying to load GIF, but no support for libgif/libungif compiled in. fixme:ole:OLEPictureImpl_get_Attributes Unknown pictype -1 fixme:ole:OleLoadPictureEx (0x94fbac,153,1,{7bf80980-bf32-101a-8bbb-00aa00300cab},x=0,y=0,f=0,0x33fab8), partially implemented. fixme:ole:OLEPictureImpl_LoadGif Trying to load GIF, but no support for libgif/libungif compiled in. fixme:ole:OLEPictureImpl_get_Attributes Unknown pictype -1 fixme:ole:OleLoadPictureEx (0x94fbac,153,1,{7bf80980-bf32-101a-8bbb-00aa00300cab},x=0,y=0,f=0,0x33fab8), partially implemented. fixme:ole:OLEPictureImpl_LoadGif Trying to load GIF, but no support for libgif/libungif compiled in. fixme:ole:OleLoadPictureEx (0x94fbac,128,1,{7bf80980-bf32-101a-8bbb-00aa00300cab},x=0,y=0,f=0,0x33fab8), partially implemented. fixme:ole:OLEPictureImpl_LoadGif Trying to load GIF, but no support for libgif/libungif compiled in. fixme:ole:OLEPictureImpl_SaveAsFile (0x15f4b0)->(0x1783c0, 0, (nil)), hacked stub. err:shdocvw:get_typeinfo LoadRegTypeLib failed: 8002801d fixme:ole:OLEPictureImpl_FindConnectionPoint no connection point for {33ad4ed2-6699-11cf-b70c-00aa0060d393} [and the fixmes like this repeat several times] This on a windows app that worked (barely) before. The other two windows apps I have installed seem to still work. I get a little window with Run-time error '-2147319779 (8002801d)': Automation error Application-defined or object-defined error and the app is dead . I'm supposing this is a problem with the rpm somehow (rather than wine itself), but that is a wild guess. The previous FC4 wine install (also from a normal FC4 binary rpm) did not have this problem. The app was only barely usable, some little buttons showed no text labels, so all was not totally well before either. David Anderson (I'll also try this app building wine with latest wine source on an AMD athlon 64bit with wine built -m32 to see what happens).

2 1

Re: riched20: ME_SetSelection -1 to -1 behavior fix
by Matt Finnicum 02 Sep '06

02 Sep '06

Mark, This patch, like the first one you submitted, is linewrapped (meaning it won't apply cleanly). I'd suggest just attaching the patch to your e-mail, so you don't have to worry about these things. Also, it's always nice to have a conformance test for things like this. Perhaps show that it's the correct behaviour for both EM_EXSETSEL and EM_SETSEL? If you've got any questions on writing one (or just want me to do it for you), just send me an e-mail. Thanks for contributing, --Matt On 9/2/06, Mark Lu <excelblue(a)gmail.com> wrote: > ChangeLog: > * Corrects the behavior of ME_SetSelection when selecting from -1 to -1 > --- > dlls/riched20/caret.c | 8 ++++++++ > 1 files changed, 8 insertions(+), 0 deletions(-) > > diff --git a/dlls/riched20/caret.c b/dlls/riched20/caret.c > index bdab179..06f021c 100644 > --- a/dlls/riched20/caret.c > +++ b/dlls/riched20/caret.c > @@ -90,6 +90,14 @@ void ME_SetSelection(ME_TextEditor *edit > ME_ClearTempStyle(editor); > return; > } > + if (from == -1 && to == -1) /*-1,-1 means put the selection at the > end of the text */ > + { > + editor->pCursors[1].pRun = editor->pCursors[0].pRun = > ME_FindItemBack(editor->pBuffer->pLast, diRun); > + editor->pCursors[1].nOffset = editor->pCursors[0].nOffset = 0; > + ME_InvalidateSelection(editor); > + ME_ClearTempStyle(editor); > + return; > + } > if (from == -1) > { > editor->pCursors[1] = editor->pCursors[0]; > -- > 1.4.1.1 > > > >

1 0

Re: usp10: Resend - usp10: Implement and test Script{Apply, Record}DigitSubstitution.
by Mike McCormack 02 Sep '06

02 Sep '06

Albert Lee wrote: > + TRACE("%p, %p, %p\n", sds, sc, ss); > + > + SCRIPT_DIGITSUBSTITUTE psds; > + This causes a warning: gcc -m32 -c -I. -I. -I../../include -I../../include -D__WINESRC__ -D_REENTRANT -fPIC -Wall -pipe -fno-strict-aliasing -gstabs+ -Wdeclaration-after-statement -Wwrite-strings -Wpointer-arith -g -O2 -o usp10.o usp10.c usp10.c: In function ‘ScriptApplyDigitSubstitution’: usp10.c:256: warning: ISO C90 forbids mixed declarations and code Mike

2 1