Re: Correctly unicodify winmm/mmio.c

2005-10-17 Thread Andreas Mohr
Hi, On Tue, Oct 18, 2005 at 08:21:55AM +0200, Eric Pouech wrote: > Vincent Béron wrote: > >Changelog: > >Correctly unicodify winmm/mmio.c. > don't apply this patch, it's plain wrong. > the A<->W message mapping isn't implemented yet, and all our internal > handlers are still ANSI. I was just thin

Re: Correctly unicodify winmm/mmio.c

2005-10-17 Thread Eric Pouech
Vincent Béron wrote: Changelog: Correctly unicodify winmm/mmio.c. don't apply this patch, it's plain wrong. the A<->W message mapping isn't implemented yet, and all our internal handlers are still ANSI. The steps to go are (in order): - implement A<->W message mapping for MMIO - translate our

Re: Unicodify hhctrl.ocx HtmlHelpW

2005-10-17 Thread James Hawkins
On 17 Oct 2005 20:02:16 -0400, Vincent Béron <[EMAIL PROTECTED]> wrote: > Changelog: > Unicodify HtmlHelpW in hhctrl.ocx > We should leave the message out completely at this point. The integral components of HTML Help have been implemented, and at least HH_DISPLAY_TOPIC needs to call doWinMain.

Fwd: Re: [kernel] correct OpenFile test

2005-10-17 Thread Oliver Stieber
-- Maxime Bellengé <[EMAIL PROTECTED]> wrote: > > Hello, > > You can try this application installer : > http://jatennis.free.fr/jatinst32.php > I've had a look and the installer is failing because of the result returned by : 0009:Call version.VerFindFileA(,0040b4ec "JaTennis32.exe",004

Re: [kernel] correct OpenFile test

2005-10-17 Thread Oliver Stieber
--- Maxime Bellengé <[EMAIL PROTECTED]> wrote: > For me returning HFILE_ERROR makes one of my application installer > fails. If the function returns FALSE, it behaves normaly and exactly > like on windows. The application installer works correctly. > Which version of window are you using and do

RE: Cannot compile 20050830 on Solaris 9

2005-10-17 Thread rddone
Ok, Im almost there (or at least alot closer). After editing a source file and a few Makefiles and manually applying several diffs from Robert Lunnons patchkit from different Wine versions (none applicable to 20050830). I got Wine to compile on Solaris 9! Now when running wine, it sits for

Re: [kernel] correct OpenFile test

2005-10-17 Thread Maxime Bellengé
For me returning HFILE_ERROR makes one of my application installer fails. If the function returns FALSE, it behaves normaly and exactly like on windows. The application installer works correctly. Maxime On Mon, 2005-10-17 at 15:20 +0100, Oliver Stieber wrote: > --- Dmitry Timoshkov <[EMAIL PROTE

Re: [kernel] correct OpenFile test

2005-10-17 Thread Oliver Stieber
--- Dmitry Timoshkov <[EMAIL PROTECTED]> wrote: > "Oliver Stieber" <[EMAIL PROTECTED]> wrote: > > > > file.c:1413: Test failed: hFile != FALSE : -1 > > > > -1 is HFILE_ERROR and FALSE is 0 (defined in windef.h), is this running > > the old or the > patched > > one? > > It's the result of ru

Re: [kernel] correct OpenFile test

2005-10-17 Thread Dmitry Timoshkov
"Oliver Stieber" <[EMAIL PROTECTED]> wrote: > > file.c:1413: Test failed: hFile != FALSE : -1 > > -1 is HFILE_ERROR and FALSE is 0 (defined in windef.h), is this running the > old or the patched > one? It's the result of running existing tests. I just wanted to point out that the second chunk

Re: [kernel] correct OpenFile test

2005-10-17 Thread Oliver Stieber
--- Dmitry Timoshkov <[EMAIL PROTECTED]> wrote: > "Dmitry Timoshkov" <[EMAIL PROTECTED]> wrote: > > > >This patch corrects the test for OpenFile with a style of OF_EXIST. > > > The test should fail > if the > > > function returns anything other than HFILE_ERROR as a failure. This is > > >

Re: [kernel] correct OpenFile test

2005-10-17 Thread Dmitry Timoshkov
"Dmitry Timoshkov" <[EMAIL PROTECTED]> wrote: > >This patch corrects the test for OpenFile with a style of OF_EXIST. The > > test should fail if the > > function returns anything other than HFILE_ERROR as a failure. This is > > based on > > http://msdn.microsoft.com/library/default.asp?url=/

Re: [kernel] correct OpenFile test

2005-10-17 Thread Oliver Stieber
--- Dmitry Timoshkov <[EMAIL PROTECTED]> wrote: > "Oliver Stieber" <[EMAIL PROTECTED]> wrote: > > >This patch corrects the test for OpenFile with a style of OF_EXIST. The > > test should fail if > the > > function returns anything other than HFILE_ERROR as a failure. This is > > based on >

Help debugging a SEH

2005-10-17 Thread Oliver Stieber
Hi, I've been trying to locate an odd bug in the game Perimeter. The crash happens when execution changes to x41 bytes into the ProcessHeap structure, so I assume that there's some stack corruption. The last bit of the log looks like this: (with relay exclude of ntdll.*;kernel32.*;gdi32.*) 000

Re: [kernel] correct OpenFile test

2005-10-17 Thread Dmitry Timoshkov
"Oliver Stieber" <[EMAIL PROTECTED]> wrote: >This patch corrects the test for OpenFile with a style of OF_EXIST. The > test should fail if the > function returns anything other than HFILE_ERROR as a failure. This is based > on > http://msdn.microsoft.com/library/default.asp?url=/library/en-u

Re: NTDLL - breaks MDAC Installation

2005-10-17 Thread Mike McCormack
Ray Jones wrote: This patch: breaks the MDAC Installation, no matter which version you attempt to install. The patch changes the way dll overrides work, and is unlikely to be the direct cause of the regression, you probably only need to adjust your configuration a little. If anyone lik

NTDLL - breaks MDAC Installation

2005-10-17 Thread Ray Jones
This patch: ___ === RCS file: /home/wine/wine/dlls/ntdll/loadorder.c,v retrieving revision 1.16 retrieving revision 1.17 diff -u -p -r1.16 -r1.17 --- wine/dlls/ntdll/loadorder.c 2005/06/21 09:52:41 1.16 ++

Re: regression in kernel OpenFile

2005-10-17 Thread Dmitry Timoshkov
"Oliver Stieber" <[EMAIL PROTECTED]> wrote: > MSDN Notes: > http://msdn.microsoft.com/library/default.asp?url=/library/en-us/fileio/fs/openfile.asp > If the function fails, the return value is HFILE_ERROR > > There's no reference to returning FALSE is OF_EXIST is passed as a parameter. dlls/ker

Re: bug 3578

2005-10-17 Thread Stephen Torri
On Mon, 2005-10-17 at 10:33 +0200, [EMAIL PROTECTED] wrote: > Hi all, > > i posted a bug report last friday (10/14) about a nice shoot'em up called > Warblade, complaining at startup : > > "err:seh:EXC_DefaultHandling Exception frame is not in stack limits => unable > to > dispatch exception." >

Re: user/test: Tests for broken behavior of ShowWindow for maximizedpopup windows

2005-10-17 Thread Dmitry Timoshkov
"Vitaliy Margolen" [EMAIL PROTECTED] wrote: > changelog: > user/test: > - Tests for broken behavior of ShowWindow for maximized popup windows. > + todo_wine ok_sequence(WmCreateInvisibleMaxPopupSeq, > "CreateWindow(WS_MAXIMIZED):popup", TRUE); You don't need to use todo_wine for a failing o

Re: ntdll: Append the file_part for \\?\ long file names too [TRY 2]

2005-10-17 Thread Alexandre Julliard
James Hawkins <[EMAIL PROTECTED]> writes: > diff -u -p -r1.35 path.c > --- dlls/ntdll/path.c 11 Aug 2005 10:41:27 - 1.35 > +++ dlls/ntdll/path.c 15 Oct 2005 16:24:23 - > @@ -421,6 +421,9 @@ BOOLEAN WINAPI RtlDosPathNameToNtPathNa > if (!ntpath->Buffer) return FALSE; >

Re: SHELL32: Use ShellLink_QueryInterface to return the right interface in IShellLink_Constructor.

2005-10-17 Thread Alexandre Julliard
Mike McCormack <[EMAIL PROTECTED]> writes: > @@ -1176,18 +1176,7 @@ HRESULT WINAPI IShellLink_Constructor( I > > TRACE("(%p)->()\n",sl); > > - if (IsEqualIID(riid, &IID_IUnknown) || > - IsEqualIID(riid, &IID_IShellLinkA)) > - *ppv = sl; > - else if (IsEqualIID(rii

bug 3578

2005-10-17 Thread sebastien . fievet
Hi all, i posted a bug report last friday (10/14) about a nice shoot'em up called Warblade, complaining at startup : "err:seh:EXC_DefaultHandling Exception frame is not in stack limits => unable to dispatch exception." I searched the lists for other entries about this game, and found another bug

Re: kernel: Fix [Bug:1885] - LoadLibraryA("name ")

2005-10-17 Thread Alexandre Julliard
Vitaliy Margolen <[EMAIL PROTECTED]> writes: > Any reason why this patch didn't go in? It does fix the reported bug and tests > show the correct behavior. You should only make a copy of the name when really necessary, to avoid penalizing the common case. -- Alexandre Julliard [EMAIL PROTECTED]