- Wine-Devel - list.winehq.org

Re: widl: let widl understand typedef enum
by Huw D M Davies Feb. 8, 2005

Feb. 8, 2005

On Sun, Feb 06, 2005 at 09:31:37AM -0500, Vincent Béron wrote: > Le dim 06/02/2005 à 09:01, Mike Hearn a écrit : > > On Sat, 05 Feb 2005 12:37:08 -0600, Rob Shearman wrote: > > > Yeah, it seems strange, but I have a test program here that shows it is > > > 32-bits application-side for proxies/stubs. > > > > There's all kinds of brokenness going on in their variant marshalling, eg > > some types are mapped seemingly at random to others. I have a list with > > some of the details around here somewhere, but I'd totally believe > > whatever Rob says on this no matter how weird. > > VT_I4 it is then. > > This supercedes my widl3.diff patch. > > Changelog: > A RPC_FC_ENUM16 is a VT_I4. Actually this should be VT_USERDEFINED. Huw. -- Huw Davies huw(a)codeweavers.com

1 0

Wine for Linux PPC emulate Mac OS
by M-Rick Feb. 8, 2005

Feb. 8, 2005

The Wine http://www.winehq.com/ project has been created to emulate Windows to be able to run windows softwares directly under Linux without to have to emulate Windows. It's very great and wonderful idea ! The Darwine http://darwine.opendarwin.org/ project aims to to the same but under Mac OS X with adding also the emulation of an x86 CPU. But, I wanted to know something, or even to make a proposal. Would it be possible to make a special Wine version for Linux PPC that instead of emulating Windows would emulate Mac OS ? Yes there's already the very great Mac On Linux PPC (MOL) http://www.maconlinux.org/ that can let run Mac OS applications, but you have to run Mac OS in parallel ... It is great, but Mac OS takes rooms and CPU and for some usages, maybe it is not completely useful. Would it be possible to make such a software a WinePPC for exemple that would let run Microsoft Office X 2004 Mac or Adobe CS Mac or Macromedia MX Mac directly under Linux PPC without having to install a Mac OS ? To make simple to run Mac OS softwares directly under the Linux desktop environment.

6 6

Re: dsound: Properly implement DllCanUnloadNow
by Robert Reif Feb. 8, 2005

Feb. 8, 2005

James Hawkins wrote: >Hi, > >Changelog >* Properly implement DllCanUnloadNow ref counting. > > > dsound.dll is a strange dll that doesn't really follow proper COM reference counting rules for buffers. This is intentional by Microsoft (don't know why) and was made common by the example code from "Inside Direct X" . You may want to add a test to the dsound.dll regression tests to verify what Microsoft does.

2 2

[oleaut32/typelib.c] TLB_ReadTypeLib question
by Paul Vriens Feb. 7, 2005

Feb. 7, 2005

Hi, while browsing through typelib.c I saw: 287 */ 288 int TLB_ReadTypeLib(LPCWSTR file, INT index, ITypeLib2 **ppTypelib); 289 290 HRESULT WINAPI LoadTypeLib( Is line 288 supposed to be there ? Cheers, Paul.

2 1

Re: widl: support ATTR_READONLY for interface properties
by Huw D M Davies Feb. 7, 2005

Feb. 7, 2005

On Sun, Feb 06, 2005 at 09:50:31AM -0500, Vincent Béron wrote: > There are probably other easy ones like this one, but for now I only > need this one. > > Huw, am I stepping on your toes? Nope, but I do have some more patches left in my tree that I need to tidy up a bit before submitting - notably support for modules. I've almost got a version of stdole2.tlb too... But feel free to carry on. Cheers, Huw. -- Huw Davies huw(a)codeweavers.com

2 1

Re-examining SetThreadContext
by Mike Hearn Feb. 7, 2005

Feb. 7, 2005

Hi Alexandre, A user recently reported another app that doesn't work due to Armadillo copy protection. This scheme requires SetThreadContext/GetThreadContext to work properly and currently they do not. These two functions require the thread to be suspended, therefore they always return a bogus context (blocking on a wineserver wait pipe). I wrote a patch for this a while ago: http://navi.cx/svn/misc/trunk/winekb/patches/thread-context.patch However you felt it had races in it. There were two possible races I could see: - The race between suspending the thread and the sigcontext being uploaded. This was not a problem with my patch as Get/SetThreadContext loop. - A race where you can call SuspendThread again before the SIGUSR1 handler has completed. I believe this was the sticking point last time, no? But I can't see how this can happen because SIGUSR1 is blocked while it's being handled. So in the following sequence: PROCESS A: SuspendThread() PROCESS A: GetThreadContext() PROCESS B: SIGUSR1 PROCESS B: upload sigcontext PROCESS B: block PROCESS A: GetThreadContext() returns PROCESS A: ResumeThread PROCESS B: unblock, download sigcontext, restore, finish I do not see where the races are. Could you please re-examine this patch and let me know if there are any more problems with it? Dealing with the set/thread context operations in SIGUSR1 makes sense because they are only allowed when the target thread is suspended. thanks -mike

2 5

Re: wine/dlls/dbghelp elf_module.c
by Eric Pouech Feb. 7, 2005

Feb. 7, 2005

Alexandre Julliard a écrit : > ChangeSet ID: 15811 > CVSROOT: /opt/cvs-commit > Module name: wine > Changes by: julliard(a)wine.codeweavers.com 2005/01/31 10:24:57 > > Modified files: > dlls/dbghelp : elf_module.c > > Log message: > Silence a noisy FIXME I fear that the issue may be somewhere else than the pure FIXME. Can you describe what actually happened ? A+ -- Eric Pouech

2 1

Question about TrueType rendering
by Eric Frias Feb. 7, 2005

Feb. 7, 2005

I'm trying to get winelib working on a platform whose X server doesn't have native support for TrueType fonts. I've configured wine with freetype, but it never seems to display the TrueType fonts. I don't have xfsft or xfstt installed. Should I expect this to work? I don't understand whether Wine uses the freetype library to render TrueType fonts, or if it just uses it for some other purpose (like getting metrics) and still depends on the font server to render the fonts. I see messages like this when wine starts, so it looks like wine is finding my fonts, but when I Call CreateFont("Arial"), it never maps to a TrueType font: trace:font:ReadFontDir Found "arial.ttf" in "/devel/syncad/.SynaptiCAD/mnt/c/windows/Fonts" trace:font:AddFontFileToList Loading font file "/devel/syncad/.SynaptiCAD/mnt/c/windows/Fonts/arial.ttf" index 0 trace:font:AddFontFileToList fsCsb = 400001ff ffff0000/00007a87 80000000 00000008 00000000 trace:font:AddFontFileToList Added font L"Arial" L"Regular" Is it still necessary to run xfsft to use these fonts, or should wine be able to use them on its own? Thanks, Eric

2 2

Re: wine/ windows/defwnd.c dlls/x11drv/x11drv_main ...
by Michael Stefaniuc Feb. 7, 2005

Feb. 7, 2005

Hi Alexandre, this patch introduced a regression with FreeSolitaire (free download at www.freesolitaire.com) After switching virtual desktops the program windows gets blanked (see attached image). The program dosn't lockThis seems to be the only program with which i could reproduce this, all the others (winemine, notepad, winefile, Flash MX) seem to work just fine. bye michael Alexandre Julliard wrote: > ChangeSet ID: 15496 > CVSROOT: /opt/cvs-commit > Module name: wine > Changes by: julliard(a)wine.codeweavers.com 2005/01/17 13:17:47 > > Modified files: > windows : defwnd.c > dlls/x11drv : x11drv_main.c x11drv.h winpos.c window.c > scroll.c mouse.c event.c clipboard.c > > Log message: > Avoid most references to the internals of the WND structure by passing > around an x11drv_win_data pointer instead. > > Patch: http://cvs.winehq.org/patch.py?id=15496 > > Old revision New revision Changes Path > 1.99 1.100 +0 -3 wine/windows/defwnd.c > 1.97 1.98 +0 -1 wine/dlls/x11drv/x11drv_main.c > 1.42 1.43 +17 -24 wine/dlls/x11drv/x11drv.h > 1.109 1.110 +76 -86 wine/dlls/x11drv/winpos.c > 1.87 1.88 +213 -199 wine/dlls/x11drv/window.c > 1.21 1.22 +0 -1 wine/dlls/x11drv/scroll.c > 1.16 1.17 +3 -6 wine/dlls/x11drv/mouse.c > 1.40 1.41 +5 -4 wine/dlls/x11drv/event.c > 1.30 1.31 +0 -1 wine/dlls/x11drv/clipboard.c > > -- Michael Stefaniuc Tel.: +49-711-96437-199 System Administration Fax.: +49-711-96437-111 Red Hat GmbH Email: mstefani(a)redhat.com Hauptstaetterstr. 58 http://www.redhat.de/ D-70178 Stuttgart

2 2

X11DRV_GetKeyNameText not finding scan code
by tom fogal Feb. 7, 2005

Feb. 7, 2005

Hi all, I'm a newbie to wine coding trying to get a particular application to work. It seems to have trouble getting input correctly, and I've traced it to the aforementioned function in dlls/x11drv/keyboard.c. I'm using the wine CVS. The function seems to search through a particular table (keyc2scan) and I think the problem is that the key I am pressing is not in the table -- I've verified that it is not finding a ton of keypresses, although I can't see a console while the app is running so I can associate a particular value with a particular keypress. The general problem seems common among win32 && Xlib, at least. Application developers want to see which key is pressed, but different keys have different meanings/values depending on the current locale and maybe other settings. wine's GetKeyNameText seems to map that particular winapi call to XKeysymToString(XKeycodeToKeysym) calls, which sounds right. The table gets built in X11DRV_InitKeyboard, AFAICT. I can't see anything wrong with the way it is built -- although there is a huge chunk of code #ifdef 0'd out ("/* this breaks VK_OEM_x VKeys in some layout tables [. . .]"). I'm lost as to what minor detail wine seems to be missing here, it seems consistent with MSDN docs I've found online. Are there known issues with this piece of wine? Is anyone exactly sure what this function should be returning (aka "a\0" in lpBuffer and a retval of 1, for pressing the 'a' key? or some code that means 'a'?) ? Most importantly, whats the next step in debugging this? Thanks, -tom

3 2