August 2007 - Wine-Devel - list.winehq.org

Re: [oleaut32/tests] Error when running tests on Wine, anyone?

by Detlef Riekenberg

On So, 2007-08-12 at 10:50 +0200, Stefan Leichter wrote: > Am Sunday 12 August 2007 09:17 schrieb Detlef Riekenberg: > > On Fr, 2007-08-10 at 13:34 +0200, Paul Vriens wrote: > > > Hi, > > > > > > Can somebody confirm if all oleaut32 tests run correctly on Wine. I'm > > > trying to change LoadLibrary to GetModuleHandle and though it was my > > > failure, but current GIT fails for me: > > > > > > ../../../tools/runtest -v -P wine -M oleaut32.dll -T ../../.. -p > > > oleaut32_test.exe.so vartest.c && touch vartest.ok > > > fixme:variant:VarMod Could not convert left type 4096 to 20? rc == > > > 0x80020008 fixme:variant:VarMod Could not convert left type 8192 to 20? > > > rc == 0x80020008 fixme:variant:VarMod Could not convert left type 16384 > > > to 20? rc == 0x80020008 vartest.c:3128: Test failed: got VT_DATE(-1.5) > > > expected VT_DATE(-1.4) vartest.c:3129: Test failed: got VT_R8(-1.5) > > > expected VT_R8(-1.4) vartest: 1031879 tests executed (0 marked as todo, 2 > > > failures), 0 skipped. make: *** [vartest.ok] Error 2 > > > > detlef(a)p4:~/wine.cvs/bin$ wine dlls/oleaut32/tests/oleaut32_test.exe.so > > vartest > > vartest.c:1844:Skipping VarAbs(BSTR) as decimal separator is ',' > > fixme:variant:VarMod Could not convert left type 4096 to 20? rc == > > 0x80020008 > > fixme:variant:VarMod Could not convert left type 8192 to 20? rc == > > 0x80020008 > > fixme:variant:VarMod Could not convert left type 16384 to 20? rc == > > 0x80020008 > > vartest.c:3129: Test failed: got VT_R8(-145) expected VT_R8(-1.4) > > vartest: 1031878 tests executed (0 marked as todo, 1 failure), 0 > > skipped. > > > > > > Intel P4 2GHz / Ubuntu 6.06 > > Should we think about localisation? That is it! > I expect Detlef to run the same local than > me on his box and we are getting the same failing test > > sle(a)sle3:/usr/src/wine/wine-git$ env | grep LANG > LANG=de_DE.UTF-8 > LANGUAGE=de_DE:de:en_GB:en Same LANG here, but I can switch to "0 failures": >detlef(a)p4:~/wine.cvs/bin$ LANG="en_GB" wine dlls/oleaut32/tests/oleaut32_test.exe.so vartest >... >vartest: 1031879 tests executed (0 marked as todo, 0 failures), 0 skipped. I get the same "0 failures" with "en_GB.UTF-8", "es_ES", "it_IT", "fr_FR" > Of course we are running both debian base distributions. This might be another > item to look on. But: w2k_sp4 (german): 0 faulures -- By by ... Detlef

Aug. 13, 2007

Re: Allow building of ddraw.dll / ddrawex.dll on MingW

by Roderick Colenbrander

Hi, I think I got my mingw from the debian packages. Others also suggested that the dxguid version in mingw was outdated. Roderick On Sunday 12 August 2007 22:01, Stefan Leichter wrote: > Am Sunday 12 August 2007 15:17 schrieb Roderick Colenbrander: > > Allow building of ddraw/ddrawex on mingw. By default the dxguid from > > mingw is used which doesn't work for us as it is outdated. > > > > Regards, > > Roderick Colenbrander > > Can you please name the outdated dxguid and the location where you got the > mingw libs. > > Thanks Stefan

Aug. 12, 2007

Re: Allow building of ddraw.dll / ddrawex.dll on MingW

by Stefan Leichter

Am Sunday 12 August 2007 15:17 schrieb Roderick Colenbrander: > Allow building of ddraw/ddrawex on mingw. By default the dxguid from mingw > is used which doesn't work for us as it is outdated. > > Regards, > Roderick Colenbrander Can you please name the outdated dxguid and the location where you got the mingw libs. Thanks Stefan

Aug. 12, 2007

Re: WGL: proper stereo / double buffering support for ChoosePixelFormat

by Chris Robinson

On Sunday 12 August 2007 05:44:31 am Roderick Colenbrander wrote: > pglXGetFBConfigAttrib(gdi_display, fmt->fbconfig, > GLX_DOUBLEBUFFER, &value); > - if (value) dwFlags |= PFD_DOUBLEBUFFER; > - if (!(ppfd->dwFlags & PFD_DOUBLEBUFFER_DONTCARE) && > ((ppfd->dwFlags^dwFlags) & PFD_DOUBLEBUFFER)) > - { > - TRACE("dbl buffer mismatch for iPixelFormat=%d\n", i+1); > + if( (value & PFD_DOUBLEBUFFER) && (ppfd->dwFlags && > PFD_DOUBLEBUFFER) ) { This change doesn't look quite right. 'value' will be set to true or false, it won't have the PFD_DOUBLEBUFFER flag set coming from a GLX function. And checking the ppfd->dwFlags should use a single &, not a double &&. Also, even with those fixes, it'll only attempt to match when the doublebuffer flag is set on both the fbconfig and the wanted pixel format, not if it's unset on both, or if it's supposed to be ignored. As well, the check will preempt the other fields if a doublebuffer match was already found. Same with the stereo flag.

Aug. 12, 2007

Re: schannel status and crypt32 blockage

by Juan Lang

Hi Yuval, > [Should I repost on-list?] Yeah, let's talk about this in the open. > I've created a test that connects to https://wiki.winehq.org, and > found the following: Excellent! Glad it's getting some attention. > - It fails in rsaenh due to OPAQUEKEYBLOB not supported in CPExportKey > and CPImportKey, with CALG_RC4 and CALG_SCHANNEL_MAC_KEY. I've > implemented some hack (can't pass through AJ), and it continues. Interesting. I haven't tried this case, but it doesn't look (from looking at rsaenh) like this should be too tough. Send the hack, and I'll try to take a look when I get a chance. > - I can't get a testcase for EncryptMessage/DecryptMessage that works > on Windows, so I can't tell how far have I got. I don't know what I do > wrong. Okay. You might ask again on-list with some code, there are a couple people here (e.g. Kai) that are more knowledgeable about secur32. > I've also tried googletalk. > - It fails immediately after CertGetCertificateChain reports it is a > stub. Right. I have a partially done implementation of that lying around somewhere that I can send you if you like. It may not work right away, I'm not positive CertGetIssuerCertificateFromStore is right just yet. I'm planning to work on that very soon, most likely next week, so if you can wait that long I should have something. > I'd like to note that native crypt32 is not an option (not even > temporary, such as schannel), as it tries to use keysvc. So basically > I'm blocked. Okay. > Are you planning on implementing Cert*CertificateChain soon? Yes :) > What about rsaenh, is it in the scope of your work? Not exactly. I'm just trying to get iTunes to start. But I'd probably take a look anyway ;) Cheers, --Juan ____________________________________________________________________________________ Pinpoint customers who are looking for what you sell. http://searchmarketing.yahoo.com/

Aug. 12, 2007

Allow building of ddraw.dll / ddrawex.dll on MingW

by Roderick Colenbrander

Allow building of ddraw/ddrawex on mingw. By default the dxguid from mingw is used which doesn't work for us as it is outdated. Regards, Roderick Colenbrander

Aug. 12, 2007

Re: [oleaut32/tests] Error when running tests on Wine, anyone?

by Stefan Leichter

Am Sunday 12 August 2007 09:17 schrieb Detlef Riekenberg: > On Fr, 2007-08-10 at 13:34 +0200, Paul Vriens wrote: > > Hi, > > > > Can somebody confirm if all oleaut32 tests run correctly on Wine. I'm > > trying to change LoadLibrary to GetModuleHandle and though it was my > > failure, but current GIT fails for me: > > > > ../../../tools/runtest -v -P wine -M oleaut32.dll -T ../../.. -p > > oleaut32_test.exe.so vartest.c && touch vartest.ok > > fixme:variant:VarMod Could not convert left type 4096 to 20? rc == > > 0x80020008 fixme:variant:VarMod Could not convert left type 8192 to 20? > > rc == 0x80020008 fixme:variant:VarMod Could not convert left type 16384 > > to 20? rc == 0x80020008 vartest.c:3128: Test failed: got VT_DATE(-1.5) > > expected VT_DATE(-1.4) vartest.c:3129: Test failed: got VT_R8(-1.5) > > expected VT_R8(-1.4) vartest: 1031879 tests executed (0 marked as todo, 2 > > failures), 0 skipped. make: *** [vartest.ok] Error 2 > > detlef(a)p4:~/wine.cvs/bin$ wine dlls/oleaut32/tests/oleaut32_test.exe.so > vartest > vartest.c:1844:Skipping VarAbs(BSTR) as decimal separator is ',' > fixme:variant:VarMod Could not convert left type 4096 to 20? rc == > 0x80020008 > fixme:variant:VarMod Could not convert left type 8192 to 20? rc == > 0x80020008 > fixme:variant:VarMod Could not convert left type 16384 to 20? rc == > 0x80020008 > vartest.c:3129: Test failed: got VT_R8(-145) expected VT_R8(-1.4) > vartest: 1031878 tests executed (0 marked as todo, 1 failure), 0 > skipped. > > > Intel P4 2GHz / Ubuntu 6.06 Should we think about localisation? I expect Detlef to run the same local than me on his box and we are getting the same failing test sle(a)sle3:/usr/src/wine/wine-git$ env | grep LANG LANG=de_DE.UTF-8 LANGUAGE=de_DE:de:en_GB:en sle(a)sle3:/usr/src/wine/wine-git$ Of course we are running both debian base distributions. This might be another item to look on. Bye Stefan

Aug. 12, 2007

Re: [oleaut32/tests] Error when running tests on Wine, anyone?

by Detlef Riekenberg

On Fr, 2007-08-10 at 13:34 +0200, Paul Vriens wrote: > Hi, > > Can somebody confirm if all oleaut32 tests run correctly on Wine. I'm trying to > change LoadLibrary to GetModuleHandle and though it was my failure, but current > GIT fails for me: > > ../../../tools/runtest -v -P wine -M oleaut32.dll -T ../../.. -p > oleaut32_test.exe.so vartest.c && touch vartest.ok > fixme:variant:VarMod Could not convert left type 4096 to 20? rc == 0x80020008 > fixme:variant:VarMod Could not convert left type 8192 to 20? rc == 0x80020008 > fixme:variant:VarMod Could not convert left type 16384 to 20? rc == 0x80020008 > vartest.c:3128: Test failed: got VT_DATE(-1.5) expected VT_DATE(-1.4) > vartest.c:3129: Test failed: got VT_R8(-1.5) expected VT_R8(-1.4) > vartest: 1031879 tests executed (0 marked as todo, 2 failures), 0 skipped. > make: *** [vartest.ok] Error 2 > detlef(a)p4:~/wine.cvs/bin$ wine dlls/oleaut32/tests/oleaut32_test.exe.so vartest vartest.c:1844:Skipping VarAbs(BSTR) as decimal separator is ',' fixme:variant:VarMod Could not convert left type 4096 to 20? rc == 0x80020008 fixme:variant:VarMod Could not convert left type 8192 to 20? rc == 0x80020008 fixme:variant:VarMod Could not convert left type 16384 to 20? rc == 0x80020008 vartest.c:3129: Test failed: got VT_R8(-145) expected VT_R8(-1.4) vartest: 1031878 tests executed (0 marked as todo, 1 failure), 0 skipped. Intel P4 2GHz / Ubuntu 6.06 -- By by ... Detlef

Aug. 12, 2007

wininet: Don't peek data in InternetQueryDataAvailable if there is no more to be expected.

by Misha Koshelev

I am going to send a more complete try2 patchset tomorrow (the patch as given is correct but there are some more problems with InternetQueryDataAvailable that need to be fixed, like the fact that it actually doesn't return ERROR_NO_MORE_FILES when all HTTP data has been retrieved). Misha

Aug. 12, 2007

Re: landfill.bugzilla.org?

by Jan Zerebecki

On Sat, Aug 11, 2007 at 01:10:58PM +0200, Roderick Colenbrander wrote: > On Saturday 11 August 2007 13:07, Dan Kegel wrote: > > Why are all our new bugs cc'd to bugbot(a)landfill.bugzilla.org? > > I suspect that's a default that needs to be removed... > > In the #winehackers channel there's now a bot which notifies of bugzilla > changes. It could be related to that. Yes, "buggbot" is fed through that address. This is not a default, but was explicitly set. (It is the same new globalwatchers setting used for our bug mailinglist. That means notification on everything, where as before we could miss changes, e.g. when the bug was reassigned without cc-ing the mailinglist.) Jan

Aug. 11, 2007