July 2004 - Wine-Devel - list.winehq.org

Re: Wine and locales

by Dmitry Timoshkov

"Alexandre Julliard" <julliard(a)winehq.org> wrote: > > There is no need to test every API, but there are APIs we need to require > > mandatory testing: > > > > GetSystemDefaultLCID > > GetUserDefaultLCID > > GetACP > > I don't see what you want to test here. I want to test whether locale has really changed and what subsequences happened. I want to see system/user locale before the switch and after. I'd like also add GetSystemDefaultLangID, GetSystemDefaultUILanguage, GetUserDefaultLangID and GetUserDefaultUILanguage to the list. Probably also GetThreadLocale. > > FindResourceA > > FindResourceExA > > These ones have been well tested already. I don't see the tests for them in the CVS, especially part about LANGID preference order. > > CompareStringA > > GetLocaleInfoA > > These take an explicit locale parameter so I don't see why changing > the system locale would suddenly break them. You are right. > > Right, that's why I'm asking to revert patch and commit it simultaneously > > with a test case and possible fixes for bugs revealed by the tests. > > I see no reason to revert the patch. It's going to make no difference > for 99% of our users, and for the remaining 1% it's about as likely to > fix things as it is to break them. OK, in the worst case perhaps I will need to send a custom version of kernel32 with the patch reverted to the affected cxoffice users. -- Dmitry.

July 30, 2004

formal standard for wine devel style

by James Hawkins

Is there a written formal standard specifying the style in which wine code is written? I remember a message a while back about how a patch couldn't be committed because it didn't meet the standards of wine coding style and that someone who knew the style could fix it. -- James Hawkins

July 30, 2004

Re: Wine and locales

by Alexandre Julliard

"Dmitry Timoshkov" <dmitry(a)baikal.ru> writes: > There is no need to test every API, but there are APIs we need to require > mandatory testing: > > GetSystemDefaultLCID > GetUserDefaultLCID > GetACP I don't see what you want to test here. > FindResourceA > FindResourceExA These ones have been well tested already. > CompareStringA > GetLocaleInfoA These take an explicit locale parameter so I don't see why changing the system locale would suddenly break them. > Right, that's why I'm asking to revert patch and commit it simultaneously > with a test case and possible fixes for bugs revealed by the tests. I see no reason to revert the patch. It's going to make no difference for 99% of our users, and for the remaining 1% it's about as likely to fix things as it is to break them. -- Alexandre Julliard julliard(a)winehq.org

July 30, 2004

Re: RPCRT4_NdrClientCall2 stub (fixme)?

by Filip Navara

Alexandre Julliard wrote: > Mike Hearn <mike(a)navi.cx> writes: > >>> Yes, I have an implementation of NdrClientCall2 in my local tree, but >>> I'm a bit busy with other COM things at the moment. I'll send an >>> updated >>> patch to the list either tonight or sometime over the weekend. >>> >> >> Alexandre, why was this not merged when it was originally sent? >> > > No idea, please resubmit. > IIRC Robert said that the patch isn't cleaned up yet and so it shouldn't be commited. He also said that he hasn't tested it much except for simple tests and the complex types weren't tested at all. - Filip

July 30, 2004

Re: Wine and locales

by Dmitry Timoshkov

"Alexandre Julliard" <julliard(a)winehq.org> wrote: > The system locale defines the current codepages, so it affects > MultiByteToWideChar and similar APIs. The user locale affects resource > loading. It's likely that there are places in Wine that don't handle > the distinction properly, but I don't think it's reasonable to demand > tests for every single API before we can consider the patch. There is no need to test every API, but there are APIs we need to require mandatory testing: GetSystemDefaultLCID GetUserDefaultLCID GetACP FindResourceA FindResourceExA CompareStringA GetLocaleInfoA > Particularly since without the patch there is no way to make user and > system locales different, so there is no way to test anything at all. Right, that's why I'm asking to revert patch and commit it simultaneously with a test case and possible fixes for bugs revealed by the tests. -- Dmitry.

July 30, 2004

Re: change the nt service pack version from 6 to 6a, correct win2k3 name

by Ivan Leo Puoti

>Easy, take several W2K3 system DLLs and kick them through winedump with what command line?

July 30, 2004

Re: add autodetect audio driver capability to winmm, second try

by Alexandre Julliard

James Hawkins <truiken(a)gmail.com> writes: > + /* try to detect arts */ > + argv_new[2] = "ps awx|grep artsd|grep -v grep|grep artsd > /dev/null"; > + if(!spawnvp(_P_WAIT, "/bin/sh", argv_new)) > + { > + driversFound[numFound] = "winearts.drv"; > + name[numFound] = "aRts"; > + numFound++; > + } This code does not belong in winmm. Autodetection in winmm should be done by having the drivers themselves fail to load if the corresponding support is missing. -- Alexandre Julliard julliard(a)winehq.org

July 29, 2004

Re: RPCRT4_NdrClientCall2 stub (fixme)?

by Alexandre Julliard

Mike Hearn <mike(a)navi.cx> writes: >> Yes, I have an implementation of NdrClientCall2 in my local tree, but >> I'm a bit busy with other COM things at the moment. I'll send an updated >> patch to the list either tonight or sometime over the weekend. > > Alexandre, why was this not merged when it was originally sent? No idea, please resubmit. -- Alexandre Julliard julliard(a)winehq.org

July 29, 2004

Re: Test drive error

by Dimitrie O. Paun

On Thu, Jul 29, 2004 at 10:05:09PM +0300, sergio ojalvo wrote: > I get the tar files with the sources from sourceforge and didn't change > this file. Yes, but unfortumately, winemaker is not perfect, and in this very case it's buggy. En.rc is not meant to be compiled independently, but rather be included in another file. That's why you need to manually fix the Makefile, but replacing En.rc with rsrc.rc. > > Using the std C++ lib is still a problem if you need to use > > msvcrt. Try to use wineg++ (instead of winegcc) *without* > > the -mno-cygwin flag. > Can you explain exactly what to do? I'm newbie on this project. > I just do: > $ winemaker --mfc -Imydir . > $ make Once again, winemaker is far from perfect. You need to manually adjust and fix the generated Makefile. In this case, as I said, 1. replace winegcc with wineg++ 2. get rid of the -mno-cygwin flag -- Dimi.

July 29, 2004

Re: RPCRT4_NdrClientCall2 stub (fixme)?

by Mike Hearn

On Thu, 29 Jul 2004 19:46:18 +0100, Robert Shearman wrote: > Wow. I think that's the first time I've ever been called a guru :) Heh, I think Ove is the true COM guru around here ... maybe me and Rob are catching up though. > Yes, I have an implementation of NdrClientCall2 in my local tree, but > I'm a bit busy with other COM things at the moment. I'll send an updated > patch to the list either tonight or sometime over the weekend. Alexandre, why was this not merged when it was originally sent? thanks -mike

July 29, 2004