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

Re: COMCTL32: Minor corrections

by Dmitry Timoshkov

"Filip Navara" <xnavara(a)volny.cz> wrote: > >> - DrawEdge(NULL, &(infoPtr->rcDraw), EDGE_SUNKEN, BF_RECT | BF_ADJUST); > >> + /* subract the size of the edge drawn by DrawEdge */ > >> + InflateRect(&infoPtr->rcDraw, -2, -2); > > > > Please use GetSystemMetrics(SM_CXEDGE) and GetSystemMetrics(SM_CYEDGE). > > I wanted to do that, but the DrawEdge code uses "1" and "-1" so I > modified the code to match it. Then DrawEdge needs to be fixed to use GetSystemMetrics as well. -- Dmitry.

Dec. 4, 2004

Re: COMCTL32: Minor corrections

by Filip Navara

Robert Shearman wrote: > Filip Navara wrote: > >> - /* use DrawEdge to adjust the size of rcEdge to get rcDraw */ >> memcpy((&infoPtr->rcDraw), (&infoPtr->rcClient), >> sizeof(infoPtr->rcDraw)); >> >> - DrawEdge(NULL, &(infoPtr->rcDraw), EDGE_SUNKEN, BF_RECT | BF_ADJUST); >> + /* subract the size of the edge drawn by DrawEdge */ >> + InflateRect(&infoPtr->rcDraw, -2, -2); > > Please use GetSystemMetrics(SM_CXEDGE) and GetSystemMetrics(SM_CYEDGE). I wanted to do that, but the DrawEdge code uses "1" and "-1" so I modified the code to match it. >> --- lib/comctl32/tooltips.c 20 Oct 2004 08:36:55 -0000 1.9 >> +++ lib/comctl32/tooltips.c 4 Dec 2004 23:18:56 -0000 >> @@ -2818,7 +2818,7 @@ >> TOOLTIPS_Unregister (void) >> { >> int i; >> - for (i = 0; i < TTI_ERROR+1; i++) >> - DeleteObject(hTooltipIcons[i]); >> + for (i = TTI_INFO; i <= TTI_ERROR; i++) >> + DestroyIcon(hTooltipIcons[i]); >> UnregisterClassW (TOOLTIPS_CLASSW, NULL); >> } > > Someone already sent a patch for this (and it is my buggy code, > sorry). Not sure why it didn't go in, but it should. My bad, I missed it. Regards, Filip

Dec. 4, 2004

Re: RFC: Use editline in winedbg

by Robert Shearman

Mike Hearn wrote: >On Sat, 04 Dec 2004 17:17:28 -0600, Robert Shearman wrote: > > >>I think this should be available via Win32 calls without the licensing >>issues. See dlls/kernel/editline.c >> >> > >Yeah, maybe, though I'm not sure how easy it'd be to integrate readline >with that stuff. I *really* *really* want this feature, it'd make my life >approximately 110.4% better, so at the risk of annoying Alexandre >I'll vote +1 for putting it in even if the code is in the wrong place and >UNIX specific. Because I want it that much :) > > It *should* be working at the moment, but it isn't for some reason. If you *really* *really* want this feature right now, then you can use Jeremy's patch, but we should find out why the console functions aren't doing what we expect. Rob

Dec. 4, 2004

Re: COMCTL32: Minor corrections

by Robert Shearman

Filip Navara wrote: > Changelog: > - Tooltip icons must be destroyed with DestroyIcon and not with > DeleteObject. > - Don't use DrawEdge with NULL device context, it's invalid call and > sets last error. > >------------------------------------------------------------------------ > >Index: lib/comctl32/datetime.c >=================================================================== >RCS file: /CVS/ReactOS/reactos/lib/comctl32/datetime.c,v >retrieving revision 1.5 >diff -u -r1.5 datetime.c >--- lib/comctl32/datetime.c 19 Sep 2004 09:50:06 -0000 1.5 >+++ lib/comctl32/datetime.c 4 Dec 2004 22:59:42 -0000 >@@ -1170,10 +1170,10 @@ > > TRACE("Height=%ld, Width=%ld\n", infoPtr->rcClient.bottom, infoPtr->rcClient.right); > >- /* use DrawEdge to adjust the size of rcEdge to get rcDraw */ > memcpy((&infoPtr->rcDraw), (&infoPtr->rcClient), sizeof(infoPtr->rcDraw)); > >- DrawEdge(NULL, &(infoPtr->rcDraw), EDGE_SUNKEN, BF_RECT | BF_ADJUST); >+ /* subract the size of the edge drawn by DrawEdge */ >+ InflateRect(&infoPtr->rcDraw, -2, -2); > > Please use GetSystemMetrics(SM_CXEDGE) and GetSystemMetrics(SM_CYEDGE). > > /* set the size of the button that drops the calendar down */ > /* FIXME: account for style that allows button on left side */ >Index: lib/comctl32/tooltips.c >=================================================================== >RCS file: /CVS/ReactOS/reactos/lib/comctl32/tooltips.c,v >retrieving revision 1.9 >diff -u -r1.9 tooltips.c >--- lib/comctl32/tooltips.c 20 Oct 2004 08:36:55 -0000 1.9 >+++ lib/comctl32/tooltips.c 4 Dec 2004 23:18:56 -0000 >@@ -2818,7 +2818,7 @@ > TOOLTIPS_Unregister (void) > { > int i; >- for (i = 0; i < TTI_ERROR+1; i++) >- DeleteObject(hTooltipIcons[i]); >+ for (i = TTI_INFO; i <= TTI_ERROR; i++) >+ DestroyIcon(hTooltipIcons[i]); > UnregisterClassW (TOOLTIPS_CLASSW, NULL); > } > > Someone already sent a patch for this (and it is my buggy code, sorry). Not sure why it didn't go in, but it should. Rob

Dec. 4, 2004

Re: RFC: Use editline in winedbg

by Mike Hearn

On Sat, 04 Dec 2004 17:17:28 -0600, Robert Shearman wrote: > I think this should be available via Win32 calls without the licensing > issues. See dlls/kernel/editline.c Yeah, maybe, though I'm not sure how easy it'd be to integrate readline with that stuff. I *really* *really* want this feature, it'd make my life approximately 110.4% better, so at the risk of annoying Alexandre I'll vote +1 for putting it in even if the code is in the wrong place and UNIX specific. Because I want it that much :)

Dec. 4, 2004

Re: RFC: Use editline in winedbg

by Andreas Mohr

Hi, On Sat, Dec 04, 2004 at 04:54:14PM -0600, Jeremy White wrote: > Hi folks, > > The attached patch dynamically detects and uses the BSD licensed editline > library (a pretty close replacement for readline), if it's available. > This gives winedbg command line recall. Very nice, finally! Now where's the warning message that says: "editline library not detected: command line history won't be available..." ? ;-) Andreas Mohr

Dec. 4, 2004

Re: RFC: Use editline in winedbg

by Robert Shearman

Jeremy White wrote: > Hi folks, > > The attached patch dynamically detects and uses the BSD licensed editline > library (a pretty close replacement for readline), if it's available. > This gives winedbg command line recall. I think this should be available via Win32 calls without the licensing issues. See dlls/kernel/editline.c I'm not sure if winedbg is currently supposed to support it and there is a bug, or if it just hasn't enabled the feature, or if it is even able to use it. Rob

Dec. 4, 2004

Re: implement DllCanUnloadNow for dmusic

by Robert Shearman

James Hawkins wrote: >On Sat, 04 Dec 2004 15:13:47 -0600, Robert Shearman <rob(a)codeweavers.com> wrote: > > >>Ok, think about this sequence of events that should leave the DLL ref >>count at 0: >>1. Create an IDirectMusicBufferImpl object. >>1a. Calls LockModule. >>1b. Calls QueryInterface which calls LockModule. >>2. Call Release, which calls LockModule. >> >>The solution is to either remove the LockModule from the constructor, or >>to remove the LockModule from the AddRef and only call UnlockModule in >>the destructor part of Release. This issue is also present in the other >>files in your patch (not quoted here). >> >>I should have thought about this before when I commented on your last >>patch. I guess I need to put some tests together for this feature. >> >>Rob >> >> >> > >MSDN about QueryInterface: >"Returns a pointer to a specified interface on an object to which a >client currently holds an interface pointer. This function must call >IUnknown::AddRef on the pointer it returns." > >QueryInterface should be calling AddRef, so if any constructor calls >QueryInterface (I'm guessing without looking that they all do) then >there shouldn't be a call to LockModule in that constructor. Is that >about right? > > That is correct. Rob

Dec. 4, 2004

Re: implement DllCanUnloadNow for dmusic

by James Hawkins

On Sat, 04 Dec 2004 15:13:47 -0600, Robert Shearman <rob(a)codeweavers.com> wrote: > James Hawkins wrote: > > >Changelog > > * properly implement DllCanUnloadNow ref counting > > > > > > > > Sorry, still isn't correct. Please see my comments below. > > >------------------------------------------------------------------------ > > > >Index: dlls/dmusic/buffer.c > >=================================================================== > >RCS file: /home/wine/wine/dlls/dmusic/buffer.c,v > >retrieving revision 1.8 > >diff -u -r1.8 buffer.c > >--- dlls/dmusic/buffer.c 6 Sep 2004 21:34:25 -0000 1.8 > >+++ dlls/dmusic/buffer.c 4 Dec 2004 20:52:16 -0000 > >@@ -38,6 +38,9 @@ > > ULONG WINAPI IDirectMusicBufferImpl_AddRef (LPDIRECTMUSICBUFFER iface) { > > IDirectMusicBufferImpl *This = (IDirectMusicBufferImpl *)iface; > > TRACE("(%p): AddRef from %ld\n", This, This->ref); > >+ > >+ DMUSIC_LockModule(); > >+ > > return ++(This->ref); > > } > > > >@@ -45,9 +48,13 @@ > > IDirectMusicBufferImpl *This = (IDirectMusicBufferImpl *)iface; > > ULONG ref = --This->ref; > > TRACE("(%p): ReleaseRef to %ld\n", This, This->ref); > >+ > > if (ref == 0) { > > HeapFree(GetProcessHeap(), 0, This); > > } > >+ > >+ DMUSIC_UnlockModule(); > >+ > > return ref; > > } > > > >@@ -160,6 +167,8 @@ > > } > > dmbuff->lpVtbl = &DirectMusicBuffer_Vtbl; > > dmbuff->ref = 0; /* will be inited by QueryInterface */ > >+ > >+ DMUSIC_LockModule(); > > > > return IDirectMusicBufferImpl_QueryInterface ((LPDIRECTMUSICBUFFER)dmbuff, lpcGUID, ppobj); > > } > > > > > > Ok, think about this sequence of events that should leave the DLL ref > count at 0: > 1. Create an IDirectMusicBufferImpl object. > 1a. Calls LockModule. > 1b. Calls QueryInterface which calls LockModule. > 2. Call Release, which calls LockModule. > > The solution is to either remove the LockModule from the constructor, or > to remove the LockModule from the AddRef and only call UnlockModule in > the destructor part of Release. This issue is also present in the other > files in your patch (not quoted here). > > I should have thought about this before when I commented on your last > patch. I guess I need to put some tests together for this feature. > > Rob > MSDN about QueryInterface: "Returns a pointer to a specified interface on an object to which a client currently holds an interface pointer. This function must call IUnknown::AddRef on the pointer it returns." QueryInterface should be calling AddRef, so if any constructor calls QueryInterface (I'm guessing without looking that they all do) then there shouldn't be a call to LockModule in that constructor. Is that about right? -- James Hawkins

Dec. 4, 2004

RFC: Use editline in winedbg

by Jeremy White

Hi folks, The attached patch dynamically detects and uses the BSD licensed editline library (a pretty close replacement for readline), if it's available. This gives winedbg command line recall. This gets around readline's license requirements. I'm nervous because I wasn't sure if that was the only issue with readline, or if there were further issues. Wack me with a clue bat if this is wrong, otw, I'll submit it. Cheers, Jeremy

Dec. 4, 2004

Re: implement DllCanUnloadNow for dmusic

by Robert Shearman

James Hawkins wrote: >Changelog > * properly implement DllCanUnloadNow ref counting > > > Sorry, still isn't correct. Please see my comments below. >------------------------------------------------------------------------ > >Index: dlls/dmusic/buffer.c >=================================================================== >RCS file: /home/wine/wine/dlls/dmusic/buffer.c,v >retrieving revision 1.8 >diff -u -r1.8 buffer.c >--- dlls/dmusic/buffer.c 6 Sep 2004 21:34:25 -0000 1.8 >+++ dlls/dmusic/buffer.c 4 Dec 2004 20:52:16 -0000 >@@ -38,6 +38,9 @@ > ULONG WINAPI IDirectMusicBufferImpl_AddRef (LPDIRECTMUSICBUFFER iface) { > IDirectMusicBufferImpl *This = (IDirectMusicBufferImpl *)iface; > TRACE("(%p): AddRef from %ld\n", This, This->ref); >+ >+ DMUSIC_LockModule(); >+ > return ++(This->ref); > } > >@@ -45,9 +48,13 @@ > IDirectMusicBufferImpl *This = (IDirectMusicBufferImpl *)iface; > ULONG ref = --This->ref; > TRACE("(%p): ReleaseRef to %ld\n", This, This->ref); >+ > if (ref == 0) { > HeapFree(GetProcessHeap(), 0, This); > } >+ >+ DMUSIC_UnlockModule(); >+ > return ref; > } > >@@ -160,6 +167,8 @@ > } > dmbuff->lpVtbl = &DirectMusicBuffer_Vtbl; > dmbuff->ref = 0; /* will be inited by QueryInterface */ >+ >+ DMUSIC_LockModule(); > > return IDirectMusicBufferImpl_QueryInterface ((LPDIRECTMUSICBUFFER)dmbuff, lpcGUID, ppobj); > } > > Ok, think about this sequence of events that should leave the DLL ref count at 0: 1. Create an IDirectMusicBufferImpl object. 1a. Calls LockModule. 1b. Calls QueryInterface which calls LockModule. 2. Call Release, which calls LockModule. The solution is to either remove the LockModule from the constructor, or to remove the LockModule from the AddRef and only call UnlockModule in the destructor part of Release. This issue is also present in the other files in your patch (not quoted here). I should have thought about this before when I commented on your last patch. I guess I need to put some tests together for this feature. Rob

Dec. 4, 2004

DInput has been broken for about 6 months now

by Scott Ritchie

Well, one of my favorite games, Fallout 2, didn't work when I tried it. This is strange though, as it was working about 3 years ago. Looking through the AppDB page, http://appdb.winehq.org/appview.php?appId=194&versionId=319 - it looks like it's been broken since June. Specifically, here's the symptoms: "DXGRAB" = "Y" will cause the app to completely takeover and lockup the system, except no input into the app (mouse or keyboard) works. I have to ctrl-alt-f2 and killall -9 wine-preloader to get back to work. Turning off DXGRAB causes the app to respond to keyboard input, but the mouse still doesn't work. Some very strange things happened in gnome when I brought my email window in front of the app (the mouse would skip from the edge of the gnome window past the app, making it impossible to click on, and other wierdness. However, none of this translated into the in-game cursor moving) I'm running the app in Desktop=640x480 mode. Furthermore, it appears that some of the old problems listed in the AppsDB page are still there - the game is very slow at certain points (well, the loading and unloading I can do with the keyboard...), and I'm running a computer roughly 10 times as fast as was designed for the game. After asking about on the channel a bit, the finger was pointed at DINPUT. I think some of the problems with DINPUT would make a fairly good candidate for something to tackle on Sunday, although doing a >6 month regression test could be annoying. I personally know about 3 people I could get to switch to Wine entirely if we could get this working :) I haven't filed a bugzilla report yet, but if it helps I will. Thoughts? -Scott Ritchie

Dec. 4, 2004

Re: New Canadian keyboard layout

by Jean-Michel Dault

Le sam 04/12/2004 à 12:59, Vincent Béron a écrit : > > The keyboard that the Government of Quebec uses is the CAN/CSA > > Z243.200-92, and is *not defined* in Wine. > If it is the one I think it is (AKA as Canadian multi-lingual, with the > "è" in place of the deadkey "`"), nobody I know except the government > uses it. There's a reason: it's not a good layout to use on a regular > basis. It looks a bit like the old Canadian multi-lingual, but it's not quite the same. Not a lot of people use it because these new keyboards are not produced in the same quantities, so they're more expensive ($99 vs $29). However, as for the usage on a regular basis, latest studies show 71% of corporate users prefer to have a layout with dedicated éèà keys. See http://www.autoroute.gouv.qc.ca/publica/clavier/faq.htm > Use ca_enhanced, as everybody else. There's a reason the name has > "enhanced" in it :) There was the old "ca", after that there was the "cf", then the "ca_enhanced", which is not based on "ca" at all, but on the "cf". Then, they came up with the CAN/CSA Z243.200-92, which has been included in XFree86 under the name "ca". Add to that the new revision with the Euro sign. At one point in time, there were 16 slightly different keyboards used at the Government! (Yes, I know it doesn't help you for your project) Just try to explain to the government that if they want to use Wine, that they have to buy over 80,000 keyboards, and that's not even counting Federal and Municipal governments. And that would be impossible since there is a law that says govt agencies have to use the CAN/CSA keyboard, and Approvisionnements Québec only carries them. Besides, it's only a patch, someone already did the dirty work of making the XFree86 symbol file, plus the console layout for iso-8859-1, plus the utf-8 console layout, plus the Wine layout ;-) > > I added the proper keyboard layout, and, apart from the fact that this > > keyboard uses up to 5 symbols per key and wine only supports 4, I get a > > 85% score, and every key works, every accent, every deadkey. > Which one has 5 symbols? I can only count 4 (in your patch getting a > score of 127). The CAN/CSA Z243.200-92 has up to 5 symbols per key. See the link below for the graphic. However, the Wine keyboard definition only supports 4 symbols maximum. The only important thing is that the keys are mapped to the right place. > > If you need more information on this keyboard, look into the XFree86 > > sources: http://cvsweb.xfree86.org/cvsweb/xc/programs/xkbcomp/symbols/ca > > Or for a picture of this keyboard: > > http://www.oqlf.gouv.qc.ca/images/ti/clavier.gif Jean-Michel Dault CTO, Revolution Linux inc.

Dec. 4, 2004

Re: Implemented CreateAsyncBindCtx

by Mike Hearn

On Fri, 03 Dec 2004 23:01:36 -0500, Jacek Caban wrote: > + hres = IBindCtx_RegisterObjectParam(bctx, (LPOLESTR)BSCBHolder, (IUnknown*)callback); > + > + *pbind = bctx; > + > + return S_OK; > } You don't do anything with hres here, is that intentional? Can RegisterObjectParam fail? thanks -mike

Dec. 4, 2004

Re: New Canadian keyboard layout

by Vincent Béron

Le sam 04/12/2004 à 11:15, Jean-Michel Dault a écrit : > Le sam 04/12/2004 à 07:20, Dmitry Timoshkov a écrit : > > "Jean-Michel Dault" <jmdault(a)revolutionlinux.com> wrote: > > > I'm doing a pilot project for the Quebec government, and we have one > > > major problem: they have a keyboard layout that's unsupported by Wine, > > > though it was commited to XFree about 3 years ago. > > What do you mean by "unsupported by Wine", is that just a warning that > > you see? What exactly doesn't work without this patch? > > The keyboard that the Government of Quebec uses is the CAN/CSA > Z243.200-92, and is *not defined* in Wine. If it is the one I think it is (AKA as Canadian multi-lingual, with the "è" in place of the deadkey "`"), nobody I know except the government uses it. There's a reason: it's not a good layout to use on a regular basis. > > This keyboard is a *legal requirement* for all computers used at the > Quebec government and the Bilingual areas of the Canadian Government. > > Problem: When X11DRV_KEYBOARD_DetectLayout tries to detect the layout, > it goes through every layout it knows, analyses and scores them. > Unfortunately, the highest score returned is for the Brazilian ABNT-2 > keyboard, and the score is only 42%, so it's *way* different. > > What happens is that none of the accents work, and that's very bad for > people who have to write in french! > > To reproduce, assuming you have a standard US keyboard, do a "setxkbmap > ca", and try the "?/" key in any iso-8859-1 aware X application (ie: > OpenOffice). You should see an eacute "é" sign. Then try with Wine, > there is no key. Same thing with the "}]", it should return ccedilla "ç" > sign. Works in X, doesn't work in Wine. Use ca_enhanced, as everybody else. There's a reason the name has "enhanced" in it :) (Yes, I know it doesn't help you for your project) > > I added the proper keyboard layout, and, apart from the fact that this > keyboard uses up to 5 symbols per key and wine only supports 4, I get a > 85% score, and every key works, every accent, every deadkey. Which one has 5 symbols? I can only count 4 (in your patch getting a score of 127). > > If you need more information on this keyboard, look into the XFree86 > sources: http://cvsweb.xfree86.org/cvsweb/xc/programs/xkbcomp/symbols/ca > Or for a picture of this keyboard: > http://www.oqlf.gouv.qc.ca/images/ti/clavier.gif Vincent

Dec. 4, 2004

Re: New Canadian keyboard layout

by Jean-Michel Dault

Please use this new patch instead. There were a few mismatches in my previous patch, I tweaked it a bit, and now I get a perfect score of 127. trace:keyboard:X11DRV_KEYBOARD_DetectLayout Attempting to match against "Canadian CAN/CSA-Z243.200-92 keyboard layout" trace:keyboard:X11DRV_KEYBOARD_DetectLayout matches=48, mismatches=0, seq=47, score=127 Le sam 04/12/2004 à 11:15, Jean-Michel Dault a écrit : > Le sam 04/12/2004 à 07:20, Dmitry Timoshkov a écrit : > > "Jean-Michel Dault" <jmdault(a)revolutionlinux.com> wrote: > > > I'm doing a pilot project for the Quebec government, and we have one > > > major problem: they have a keyboard layout that's unsupported by Wine, > > > though it was commited to XFree about 3 years ago. > > What do you mean by "unsupported by Wine", is that just a warning that > > you see? What exactly doesn't work without this patch? > > The keyboard that the Government of Quebec uses is the CAN/CSA > Z243.200-92, and is *not defined* in Wine. > > This keyboard is a *legal requirement* for all computers used at the > Quebec government and the Bilingual areas of the Canadian Government. > > Problem: When X11DRV_KEYBOARD_DetectLayout tries to detect the layout, > it goes through every layout it knows, analyses and scores them. > Unfortunately, the highest score returned is for the Brazilian ABNT-2 > keyboard, and the score is only 42%, so it's *way* different. > > What happens is that none of the accents work, and that's very bad for > people who have to write in french! > > To reproduce, assuming you have a standard US keyboard, do a "setxkbmap > ca", and try the "?/" key in any iso-8859-1 aware X application (ie: > OpenOffice). You should see an eacute "é" sign. Then try with Wine, > there is no key. Same thing with the "}]", it should return ccedilla "ç" > sign. Works in X, doesn't work in Wine. > > I added the proper keyboard layout, and, apart from the fact that this > keyboard uses up to 5 symbols per key and wine only supports 4, I get a > 85% score, and every key works, every accent, every deadkey. > > If you need more information on this keyboard, look into the XFree86 > sources: http://cvsweb.xfree86.org/cvsweb/xc/programs/xkbcomp/symbols/ca > Or for a picture of this keyboard: > http://www.oqlf.gouv.qc.ca/images/ti/clavier.gif > > Regards, > > Jean-Michel Dault jmdault(a)mandrakesoft.com -- Jean-Michel Dault <jmdault(a)mandrakesoft.com>

Dec. 4, 2004

Re: Making the Documentation into help files

by Mike Hearn

On Sun, 28 Nov 2004 20:08:41 -0800, Scott Ritchie wrote: > So, how is this done? I presume there's a freedesktop.org standard for > these things, as both Gnome and KDE have them in the same way, but > currently we're not converting the documentation sgml files into these > help files. Nope there isn't, the two desktops have similar but different systems for doing it. For GNOME look for "scrollkeeper". thanks -mike

Dec. 4, 2004

Re: New Canadian keyboard layout

by Jean-Michel Dault

Le sam 04/12/2004 à 07:20, Dmitry Timoshkov a écrit : > "Jean-Michel Dault" <jmdault(a)revolutionlinux.com> wrote: > > I'm doing a pilot project for the Quebec government, and we have one > > major problem: they have a keyboard layout that's unsupported by Wine, > > though it was commited to XFree about 3 years ago. > What do you mean by "unsupported by Wine", is that just a warning that > you see? What exactly doesn't work without this patch? The keyboard that the Government of Quebec uses is the CAN/CSA Z243.200-92, and is *not defined* in Wine. This keyboard is a *legal requirement* for all computers used at the Quebec government and the Bilingual areas of the Canadian Government. Problem: When X11DRV_KEYBOARD_DetectLayout tries to detect the layout, it goes through every layout it knows, analyses and scores them. Unfortunately, the highest score returned is for the Brazilian ABNT-2 keyboard, and the score is only 42%, so it's *way* different. What happens is that none of the accents work, and that's very bad for people who have to write in french! To reproduce, assuming you have a standard US keyboard, do a "setxkbmap ca", and try the "?/" key in any iso-8859-1 aware X application (ie: OpenOffice). You should see an eacute "é" sign. Then try with Wine, there is no key. Same thing with the "}]", it should return ccedilla "ç" sign. Works in X, doesn't work in Wine. I added the proper keyboard layout, and, apart from the fact that this keyboard uses up to 5 symbols per key and wine only supports 4, I get a 85% score, and every key works, every accent, every deadkey. If you need more information on this keyboard, look into the XFree86 sources: http://cvsweb.xfree86.org/cvsweb/xc/programs/xkbcomp/symbols/ca Or for a picture of this keyboard: http://www.oqlf.gouv.qc.ca/images/ti/clavier.gif Regards, Jean-Michel Dault jmdault(a)mandrakesoft.com

Dec. 4, 2004

Re: Wine book

by Christian Costa

Lionel Ulmer wrote: >>Just kidding. I remember Tom saying last year that native DirectX >>actually works under Wine. Maybe it's worth covering? Am I way off >>base? >> >> > >Well, native DInput somewhat works (for which I am pretty surprised, I would >need to trace it to see what Win32 calls it uses) > I got the Y axis inverted last time with native DInput. I don't remenber which game though... If can find it again if you plan to decide to trace what calls the dll does. Christian

Dec. 4, 2004

Re: New Canadian keyboard layout

by Dmitry Timoshkov

"Jean-Michel Dault" <jmdault(a)revolutionlinux.com> wrote: > I'm doing a pilot project for the Quebec government, and we have one > major problem: they have a keyboard layout that's unsupported by Wine, > though it was commited to XFree about 3 years ago. What do you mean by "unsupported by Wine", is that just a warning that you see? What exactly doesn't work without this patch? -- Dmitry.

Dec. 4, 2004

Re: Problem to draw a bitmap

by Dmitry Timoshkov

"Vincent Béron" <vberon(a)mecano.gme.usherb.ca> wrote: > > > > /* Make the window visible on the screen */ > > > > ShowWindow (hwnd, nFunsterStil); > > > > Insert UpdateWindow(hwnd); here. > > > > > > PrintBmp(hwnd, "BOARD.bmp"); > > Still, it works on XP and not on Wine, so we should fix Wine. > It's probably a message-related difference between the two. It works on XP by a pure luck. I bet it won't work if you run the app from a full screen text mode console. Window painting is asynchronous by its nature, but the app thinks it's not. -- Dmitry.

Dec. 4, 2004

Re: Problem to draw a bitmap

by Vincent Béron

Le ven 03/12/2004 à 21:48, Dmitry Timoshkov a écrit : > "Rémi Assailly" <remi.assailly(a)free.fr> wrote: > > > I tested a program ( an other one, with source code this time ) > > On my Windows XP system I have no problem. > > When I use Wine I just get a window. ( where is my bitmap ? ) > > Then I tried this : "WINEDEBUG=gdi wine progtest" and as Wine is slower > > I could see my bitmap, which disappeared as fast as it appeared... > > > > /* Make the window visible on the screen */ > > > ShowWindow (hwnd, nFunsterStil); > > Insert UpdateWindow(hwnd); here. > > > > PrintBmp(hwnd, "BOARD.bmp"); Still, it works on XP and not on Wine, so we should fix Wine. It's probably a message-related difference between the two. Vincent

Dec. 3, 2004

Re: Problem to draw a bitmap

by Dmitry Timoshkov

"Rémi Assailly" <remi.assailly(a)free.fr> wrote: > I tested a program ( an other one, with source code this time ) > On my Windows XP system I have no problem. > When I use Wine I just get a window. ( where is my bitmap ? ) > Then I tried this : "WINEDEBUG=gdi wine progtest" and as Wine is slower > I could see my bitmap, which disappeared as fast as it appeared... > > /* Make the window visible on the screen */ > > ShowWindow (hwnd, nFunsterStil); Insert UpdateWindow(hwnd); here. > > PrintBmp(hwnd, "BOARD.bmp"); -- Dmitry.

Dec. 3, 2004

Re: [resend] shell32 tests

by Rolf Kalbermatter

>+static HMODULE hshell32; >+static int (WINAPI *pSHCreateDirectoryExA)(HWND, LPCSTR, LPSECURITY_ATTRIBUTES); >+ >+static void InitFunctionPointers(void) >+{ >+ hshell32 = GetModuleHandleA("shell32.dll"); >+ >+ if(hshell32) >+ pSHCreateDirectoryExA = (void*)GetProcAddress(hshell32, "SHCreateDirectoryExA"); >+} >+ > /* creates a file with the specified name for tests */ > void createTestFile(CHAR *name) > { >@@ -398,17 +409,19 @@ > int ret; > > set_curr_dir_path(path, "testdir2\\test4.txt\0"); >- ret = SHCreateDirectoryExA(NULL, path, NULL); >+ ret = pSHCreateDirectoryExA(NULL, path, NULL); I think this is not right. Implementing dynamic loading of functions in tests is supposedly because that function may not be available on some systems. But here you invoke the (maybe uninitialized, or NULL assigned) function pointer nevertheless without protecting for this. As such your patch does only allow loading the test on such systems but will create a nice crash when run. Rolf Kalbermatter

Dec. 3, 2004

Re: 2nd attemp at DllCanUnloadNow

by Robert Shearman

James Hawkins wrote: >This time I respect heap-based objects, leaving there internal ref >counts and respective HeapFree's. I also took Alexandre's suggestions >and changed Un/LockServer to DMCOMPOS_LockServer etc and change cLocks >to refCount. Let me know if this is any closer to what it needs to >be. Thanks. > > Looking better. A few comments though: >------------------------------------------------------------------------ > >Index: dlls/dmcompos/chordmap.c >=================================================================== >RCS file: /home/wine/wine/dlls/dmcompos/chordmap.c,v >retrieving revision 1.7 >diff -u -r1.7 chordmap.c >--- dlls/dmcompos/chordmap.c 23 Aug 2004 19:39:57 -0000 1.7 >+++ dlls/dmcompos/chordmap.c 2 Dec 2004 22:28:45 -0000 >@@ -55,6 +55,9 @@ > ULONG WINAPI IDirectMusicChordMapImpl_IUnknown_AddRef (LPUNKNOWN iface) { > ICOM_THIS_MULTI(IDirectMusicChordMapImpl, UnknownVtbl, iface); > TRACE("(%p): AddRef from %ld\n", This, This->ref); >+ >+ DMCOMPOS_LockModule(); >+ > return ++(This->ref); > } > >@@ -62,9 +65,13 @@ > ICOM_THIS_MULTI(IDirectMusicChordMapImpl, UnknownVtbl, iface); > ULONG ref = --This->ref; > TRACE("(%p): ReleaseRef to %ld\n", This, This->ref); >+ > if (ref == 0) { > HeapFree(GetProcessHeap(), 0, This); > } >+ >+ DMCOMPOS_UnlockModule(); >+ > return ref; > } > >Index: dlls/dmcompos/chordmaptrack.c >=================================================================== >RCS file: /home/wine/wine/dlls/dmcompos/chordmaptrack.c,v >retrieving revision 1.6 >diff -u -r1.6 chordmaptrack.c >--- dlls/dmcompos/chordmaptrack.c 23 Aug 2004 19:39:57 -0000 1.6 >+++ dlls/dmcompos/chordmaptrack.c 2 Dec 2004 22:28:45 -0000 >@@ -52,6 +52,9 @@ > ULONG WINAPI IDirectMusicChordMapTrack_IUnknown_AddRef (LPUNKNOWN iface) { > ICOM_THIS_MULTI(IDirectMusicChordMapTrack, UnknownVtbl, iface); > TRACE("(%p): AddRef from %ld\n", This, This->ref); >+ >+ DMCOMPOS_LockModule(); >+ > return ++(This->ref); > } > >@@ -59,9 +62,13 @@ > ICOM_THIS_MULTI(IDirectMusicChordMapTrack, UnknownVtbl, iface); > ULONG ref = --This->ref; > TRACE("(%p): ReleaseRef to %ld\n", This, This->ref); >+ > if (ref == 0) { > HeapFree(GetProcessHeap(), 0, This); > } >+ >+ DMCOMPOS_UnlockModule(); >+ > return ref; > } > >Index: dlls/dmcompos/composer.c >=================================================================== >RCS file: /home/wine/wine/dlls/dmcompos/composer.c,v >retrieving revision 1.8 >diff -u -r1.8 composer.c >--- dlls/dmcompos/composer.c 6 Sep 2004 21:34:26 -0000 1.8 >+++ dlls/dmcompos/composer.c 2 Dec 2004 22:28:45 -0000 >@@ -39,6 +39,9 @@ > ULONG WINAPI IDirectMusicComposerImpl_AddRef (LPDIRECTMUSICCOMPOSER iface) { > IDirectMusicComposerImpl *This = (IDirectMusicComposerImpl *)iface; > TRACE("(%p): AddRef from %ld\n", This, This->ref); >+ >+ DMCOMPOS_LockModule(); >+ > return ++(This->ref); > } > >@@ -46,9 +49,13 @@ > IDirectMusicComposerImpl *This = (IDirectMusicComposerImpl *)iface; > ULONG ref = --This->ref; > TRACE("(%p): ReleaseRef to %ld\n", This, This->ref); >+ > if (ref == 0) { > HeapFree(GetProcessHeap(), 0, This); > } >+ >+ DMCOMPOS_UnlockModule(); >+ > return ref; > } > >Index: dlls/dmcompos/dmcompos_main.c >=================================================================== >RCS file: /home/wine/wine/dlls/dmcompos/dmcompos_main.c,v >retrieving revision 1.9 >diff -u -r1.9 dmcompos_main.c >--- dlls/dmcompos/dmcompos_main.c 6 Sep 2004 21:34:26 -0000 1.9 >+++ dlls/dmcompos/dmcompos_main.c 2 Dec 2004 22:28:45 -0000 >@@ -21,41 +21,54 @@ > > WINE_DEFAULT_DEBUG_CHANNEL(dmcompos); > >+LONG refCount; >+ > > This can be be made static since it is not used outside of this file and tells anyone looking it that this is the case too. > typedef struct { >- /* IUnknown fields */ > IClassFactoryVtbl *lpVtbl; >- DWORD ref; > } IClassFactoryImpl; > > /****************************************************************** > * DirectMusicChordMap ClassFactory > */ > static HRESULT WINAPI ChordMapCF_QueryInterface(LPCLASSFACTORY iface,REFIID riid,LPVOID *ppobj) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- FIXME("(%p, %s, %p): stub\n", This, debugstr_dmguid(riid), ppobj); >+ FIXME("- no interface\n\tIID:\t%s\n", debugstr_guid(riid)); >+ >+ if (ppobj == NULL) return E_POINTER; >+ > return E_NOINTERFACE; > } > > static ULONG WINAPI ChordMapCF_AddRef(LPCLASSFACTORY iface) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- return ++(This->ref); >+ DMCOMPOS_LockModule(); >+ >+ return 2; /* non-heap based object */ > } > > static ULONG WINAPI ChordMapCF_Release(LPCLASSFACTORY iface) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- /* static class, won't be freed */ >- return --(This->ref); >+ DMCOMPOS_UnlockModule(); >+ >+ return 1; /* non-heap based object */ > } > > static HRESULT WINAPI ChordMapCF_CreateInstance(LPCLASSFACTORY iface, LPUNKNOWN pOuter, REFIID riid, LPVOID *ppobj) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- TRACE ("(%p, %p, %s, %p)\n", This, pOuter, debugstr_dmguid(riid), ppobj); >- return DMUSIC_CreateDirectMusicChordMapImpl (riid, ppobj, pOuter); >+ HRESULT result; >+ >+ TRACE ("(%p, %s, %p)\n", pOuter, debugstr_dmguid(riid), ppobj); >+ >+ result = DMUSIC_CreateDirectMusicChordMapImpl (riid, ppobj, pOuter); >+ DMCOMPOS_LockModule(); >+ >+ return result; > } > > LockModule should only be called if the creation succeeded. In this instance it should be moved inside of the DMUSIC_CreateDirectMusicChordMapImpl into the success case (I presume this function returns an appropriate error if it cannot allocate any memory). > static HRESULT WINAPI ChordMapCF_LockServer(LPCLASSFACTORY iface,BOOL dolock) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- FIXME("(%p, %d): stub\n", This, dolock); >+ TRACE("(%d)\n", dolock); >+ >+ if (dolock) >+ DMCOMPOS_LockModule(); >+ else >+ DMCOMPOS_UnlockModule(); >+ > return S_OK; > } > >@@ -67,37 +80,49 @@ > ChordMapCF_LockServer > }; > >-static IClassFactoryImpl ChordMap_CF = {&ChordMapCF_Vtbl, 1 }; >+static IClassFactoryImpl ChordMap_CF = {&ChordMapCF_Vtbl}; > > /****************************************************************** > * DirectMusicComposer ClassFactory > */ > static HRESULT WINAPI ComposerCF_QueryInterface(LPCLASSFACTORY iface,REFIID riid,LPVOID *ppobj) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- FIXME("(%p, %s, %p): stub\n",This,debugstr_dmguid(riid),ppobj); >+ FIXME("- no interface\n\tIID:\t%s\n", debugstr_guid(riid)); >+ >+ if (ppobj == NULL) return E_POINTER; >+ > return E_NOINTERFACE; > } > > static ULONG WINAPI ComposerCF_AddRef(LPCLASSFACTORY iface) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- return ++(This->ref); >+ DMCOMPOS_LockModule(); >+ >+ return 2; /* non-heap based object */ > } > > static ULONG WINAPI ComposerCF_Release(LPCLASSFACTORY iface) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- /* static class, won't be freed */ >- return --(This->ref); >+ DMCOMPOS_UnlockModule(); >+ >+ return 1; /* non-heap based object */ > } > > static HRESULT WINAPI ComposerCF_CreateInstance(LPCLASSFACTORY iface, LPUNKNOWN pOuter, REFIID riid, LPVOID *ppobj) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- TRACE ("(%p, %p, %s, %p)\n", This, pOuter, debugstr_dmguid(riid), ppobj); >- return DMUSIC_CreateDirectMusicComposerImpl (riid, ppobj, pOuter); >+ HRESULT result; >+ TRACE ("(%p, %s, %p)\n", pOuter, debugstr_dmguid(riid), ppobj); >+ >+ result = DMUSIC_CreateDirectMusicComposerImpl (riid, ppobj, pOuter); >+ DMCOMPOS_LockModule(); >+ >+ return result; > } > LockModule should only be called if the creation succeeded. In this instance it should be moved inside of the DMUSIC_CreateDirectMusicComposerImpl into the success case (I presume this function returns an appropriate error if it cannot allocate any memory). > > static HRESULT WINAPI ComposerCF_LockServer(LPCLASSFACTORY iface,BOOL dolock) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- FIXME("(%p, %d): stub\n", This, dolock); >+ TRACE("(%d)\n", dolock); >+ >+ if (dolock) >+ DMCOMPOS_LockModule(); >+ else >+ DMCOMPOS_UnlockModule(); >+ > return S_OK; > } > >@@ -109,37 +134,49 @@ > ComposerCF_LockServer > }; > >-static IClassFactoryImpl Composer_CF = {&ComposerCF_Vtbl, 1 }; >+static IClassFactoryImpl Composer_CF = {&ComposerCF_Vtbl}; > > /****************************************************************** > * DirectMusicChordMapTrack ClassFactory > */ > static HRESULT WINAPI ChordMapTrackCF_QueryInterface(LPCLASSFACTORY iface,REFIID riid,LPVOID *ppobj) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- FIXME("(%p, %s, %p): stub\n",This,debugstr_dmguid(riid),ppobj); >+ FIXME("- no interface\n\tIID:\t%s\n", debugstr_guid(riid)); >+ >+ if (ppobj == NULL) return E_POINTER; >+ > return E_NOINTERFACE; > } > > static ULONG WINAPI ChordMapTrackCF_AddRef(LPCLASSFACTORY iface) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- return ++(This->ref); >+ DMCOMPOS_LockModule(); >+ >+ return 2; /* non-heap based object */ > } > > static ULONG WINAPI ChordMapTrackCF_Release(LPCLASSFACTORY iface) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- /* static class, won't be freed */ >- return --(This->ref); >+ DMCOMPOS_UnlockModule(); >+ >+ return 1; /* non-heap based object */ > } > > static HRESULT WINAPI ChordMapTrackCF_CreateInstance(LPCLASSFACTORY iface, LPUNKNOWN pOuter, REFIID riid, LPVOID *ppobj) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- TRACE ("(%p, %p, %s, %p)\n", This, pOuter, debugstr_dmguid(riid), ppobj); >- return DMUSIC_CreateDirectMusicChordMapTrack (riid, ppobj, pOuter); >+ HRESULT result; >+ TRACE ("(%p, %s, %p)\n", pOuter, debugstr_dmguid(riid), ppobj); >+ >+ result = DMUSIC_CreateDirectMusicChordMapTrack (riid, ppobj, pOuter); >+ DMCOMPOS_LockModule(); >+ >+ return result; > } > > Same again here. > > static HRESULT WINAPI ChordMapTrackCF_LockServer(LPCLASSFACTORY iface,BOOL dolock) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- FIXME("(%p, %d): stub\n", This, dolock); >+ TRACE("(%d)\n", dolock); >+ >+ if (dolock) >+ DMCOMPOS_LockModule(); >+ else >+ DMCOMPOS_UnlockModule(); >+ > return S_OK; > } > >@@ -151,39 +188,49 @@ > ChordMapTrackCF_LockServer > }; > >-static IClassFactoryImpl ChordMapTrack_CF = {&ChordMapTrackCF_Vtbl, 1 }; >+static IClassFactoryImpl ChordMapTrack_CF = {&ChordMapTrackCF_Vtbl}; > > /****************************************************************** > * DirectMusicTemplate ClassFactory > */ > static HRESULT WINAPI TemplateCF_QueryInterface(LPCLASSFACTORY iface,REFIID riid,LPVOID *ppobj) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- FIXME("(%p, %s, %p): stub\n", This, debugstr_dmguid(riid), ppobj); >+ FIXME("- no interface\n\tIID:\t%s\n", debugstr_guid(riid)); >+ >+ if (ppobj == NULL) return E_POINTER; >+ > return E_NOINTERFACE; > } > > static ULONG WINAPI TemplateCF_AddRef(LPCLASSFACTORY iface) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- return ++(This->ref); >+ DMCOMPOS_LockModule(); >+ >+ return 2; /* non-heap based object */ > } > > static ULONG WINAPI TemplateCF_Release(LPCLASSFACTORY iface) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- /* static class, won't be freed */ >- return --(This->ref); >+ DMCOMPOS_UnlockModule(); >+ >+ return 1; /* non-heap based object */ > } > > static HRESULT WINAPI TemplateCF_CreateInstance(LPCLASSFACTORY iface, LPUNKNOWN pOuter, REFIID riid, LPVOID *ppobj) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- TRACE ("(%p, %p, %s, %p)\n", This, pOuter, debugstr_dmguid(riid), ppobj); >+ TRACE ("(%p, %s, %p)\n", pOuter, debugstr_dmguid(riid), ppobj); > /* nothing yet */ >- WARN("(%p, %s,%p): not found\n", This, debugstr_dmguid(riid), ppobj); >+ WARN("(%s,%p): not found\n", debugstr_dmguid(riid), ppobj); >+ >+ DMCOMPOS_LockModule(); >+ > return E_NOINTERFACE; > } > > Remove the DMCOMPOS_LockModule call, otherwise it will leak a ref count if (and its a pretty big if) the app handles the error returned by this function correctly. > > static HRESULT WINAPI TemplateCF_LockServer(LPCLASSFACTORY iface,BOOL dolock) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- FIXME("(%p, %d): stub\n", This, dolock); >+ TRACE("(%d)\n", dolock); >+ >+ if (dolock) >+ DMCOMPOS_LockModule(); >+ else >+ DMCOMPOS_UnlockModule(); >+ > return S_OK; > } > >@@ -195,37 +242,49 @@ > TemplateCF_LockServer > }; > >-static IClassFactoryImpl Template_CF = {&TemplateCF_Vtbl, 1 }; >+static IClassFactoryImpl Template_CF = {&TemplateCF_Vtbl}; > > /****************************************************************** > * DirectMusicSignPostTrack ClassFactory > */ > static HRESULT WINAPI SignPostTrackCF_QueryInterface(LPCLASSFACTORY iface,REFIID riid,LPVOID *ppobj) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- FIXME("(%p, %s, %p): stub\n",This,debugstr_dmguid(riid),ppobj); >+ FIXME("- no interface\n\tIID:\t%s\n", debugstr_guid(riid)); >+ >+ if (ppobj == NULL) return E_POINTER; >+ > return E_NOINTERFACE; > } > > static ULONG WINAPI SignPostTrackCF_AddRef(LPCLASSFACTORY iface) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- return ++(This->ref); >+ DMCOMPOS_LockModule(); >+ >+ return 2; /* non-heap based object */ > } > > static ULONG WINAPI SignPostTrackCF_Release(LPCLASSFACTORY iface) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- /* static class, won't be freed */ >- return --(This->ref); >+ DMCOMPOS_UnlockModule(); >+ >+ return 1; /* non-heap based object */ > } > > static HRESULT WINAPI SignPostTrackCF_CreateInstance(LPCLASSFACTORY iface, LPUNKNOWN pOuter, REFIID riid, LPVOID *ppobj) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- TRACE ("(%p, %p, %s, %p)\n", This, pOuter, debugstr_dmguid(riid), ppobj); >- return DMUSIC_CreateDirectMusicSignPostTrack (riid, ppobj, pOuter); >+ HRESULT result; >+ TRACE ("(%p, %s, %p)\n", pOuter, debugstr_dmguid(riid), ppobj); >+ >+ result = DMUSIC_CreateDirectMusicSignPostTrack (riid, ppobj, pOuter); >+ DMCOMPOS_LockModule(); >+ >+ return result; > } > > > Same as the two other instances of this problem. > static HRESULT WINAPI SignPostTrackCF_LockServer(LPCLASSFACTORY iface,BOOL dolock) { >- IClassFactoryImpl *This = (IClassFactoryImpl *)iface; >- FIXME("(%p, %d): stub\n", This, dolock); >+ TRACE("(%d)\n", dolock); >+ >+ if (dolock) >+ DMCOMPOS_LockModule(); >+ else >+ DMCOMPOS_UnlockModule(); >+ > return S_OK; > } > >@@ -237,7 +296,7 @@ > SignPostTrackCF_LockServer > }; > >-static IClassFactoryImpl SignPostTrack_CF = {&SignPostTrackCF_Vtbl, 1 }; >+static IClassFactoryImpl SignPostTrack_CF = {&SignPostTrackCF_Vtbl}; > > /****************************************************************** > * DllMain >@@ -256,6 +315,18 @@ > return TRUE; > } > >+/********************************************************************** >+ * Dll lifetime tracking declaration >+ */ >+void DMCOMPOS_LockModule() >+{ >+ InterlockedIncrement(&refCount); >+} >+ >+void DMCOMPOS_UnlockModule() >+{ >+ InterlockedDecrement(&refCount); >+} > > /****************************************************************** > * DllCanUnloadNow (DMCOMPOS.1) >@@ -263,8 +334,7 @@ > * > */ > HRESULT WINAPI DMCOMPOS_DllCanUnloadNow(void) { >- FIXME("(void): stub\n"); >- return S_FALSE; >+ return refCount != 0 ? S_FALSE : S_OK; > } > > >Index: dlls/dmcompos/dmcompos_private.h >=================================================================== >RCS file: /home/wine/wine/dlls/dmcompos/dmcompos_private.h,v >retrieving revision 1.10 >diff -u -r1.10 dmcompos_private.h >--- dlls/dmcompos/dmcompos_private.h 22 Aug 2004 21:38:47 -0000 1.10 >+++ dlls/dmcompos/dmcompos_private.h 2 Dec 2004 22:28:45 -0000 >@@ -134,11 +134,11 @@ > extern ULONG WINAPI IDirectMusicComposerImpl_AddRef (LPDIRECTMUSICCOMPOSER iface); > extern ULONG WINAPI IDirectMusicComposerImpl_Release (LPDIRECTMUSICCOMPOSER iface); > /* IDirectMusicComposer: */ >-extern HRESULT WINAPI IDirectMusicComposerImpl_ComposeSegmentFromTemplate (LPDIRECTMUSICCOMPOSER iface, IDirectMusicStyle* pStyle, IDirectMusicSegment* pTemplate, WORD wActivity, IDirectMusicChordMap* pChordMap, IDirectMusicSegment** ppSegment); >-extern HRESULT WINAPI IDirectMusicComposerImpl_ComposeSegmentFromShape (LPDIRECTMUSICCOMPOSER iface, IDirectMusicStyle* pStyle, WORD wNumMeasures, WORD wShape, WORD wActivity, BOOL fIntro, BOOL fEnd, IDirectMusicChordMap* pChordMap, IDirectMusicSegment** ppSegment); >-extern HRESULT WINAPI IDirectMusicComposerImpl_ComposeTransition (LPDIRECTMUSICCOMPOSER iface, IDirectMusicSegment* pFromSeg, IDirectMusicSegment* pToSeg, MUSIC_TIME mtTime, WORD wCommand, DWORD dwFlags, IDirectMusicChordMap* pChordMap, IDirectMusicSegment** ppTransSeg); >-extern HRESULT WINAPI IDirectMusicComposerImpl_AutoTransition (LPDIRECTMUSICCOMPOSER iface, IDirectMusicPerformance* pPerformance, IDirectMusicSegment* pToSeg, WORD wCommand, DWORD dwFlags, IDirectMusicChordMap* pChordMap, IDirectMusicSegment** ppTransSeg, IDirectMusicSegmentState** ppToSegState, IDirectMusicSegmentState** ppTransSegState); >-extern HRESULT WINAPI IDirectMusicComposerImpl_ComposeTemplateFromShape (LPDIRECTMUSICCOMPOSER iface, WORD wNumMeasures, WORD wShape, BOOL fIntro, BOOL fEnd, WORD wEndLength, IDirectMusicSegment** ppTemplate); >+extern HRESULT WINAPI IDirectMusicComposerImpl_ComposeSegmentFromTemplate (LPDIRECTMUSICCOMPOSER iface, IDirectMusicStyle* pStyle, IDirectMusicSegment* pTemplate, WORD wActivity, IDirectMusicChordMap* pChordMap, IDirectMusicSegment** ppSegment); >+extern HRESULT WINAPI IDirectMusicComposerImpl_ComposeSegmentFromShape (LPDIRECTMUSICCOMPOSER iface, IDirectMusicStyle* pStyle, WORD wNumMeasures, WORD wShape, WORD wActivity, BOOL fIntro, BOOL fEnd, IDirectMusicChordMap* pChordMap, IDirectMusicSegment** ppSegment); >+extern HRESULT WINAPI IDirectMusicComposerImpl_ComposeTransition (LPDIRECTMUSICCOMPOSER iface, IDirectMusicSegment* pFromSeg, IDirectMusicSegment* pToSeg, MUSIC_TIME mtTime, WORD wCommand, DWORD dwFlags, IDirectMusicChordMap* pChordMap, IDirectMusicSegment** ppTransSeg); >+extern HRESULT WINAPI IDirectMusicComposerImpl_AutoTransition (LPDIRECTMUSICCOMPOSER iface, IDirectMusicPerformance* pPerformance, IDirectMusicSegment* pToSeg, WORD wCommand, DWORD dwFlags, IDirectMusicChordMap* pChordMap, IDirectMusicSegment** ppTransSeg, IDirectMusicSegmentState** ppToSegState, IDirectMusicSegmentState** ppTransSegState); >+extern HRESULT WINAPI IDirectMusicComposerImpl_ComposeTemplateFromShape (LPDIRECTMUSICCOMPOSER iface, WORD wNumMeasures, WORD wShape, BOOL fIntro, BOOL fEnd, WORD wEndLength, IDirectMusicSegment** ppTemplate); > extern HRESULT WINAPI IDirectMusicComposerImpl_ChangeChordMap (LPDIRECTMUSICCOMPOSER iface, IDirectMusicSegment* pSegment, BOOL fTrackScale, IDirectMusicChordMap* pChordMap); > > >@@ -237,6 +237,11 @@ > extern HRESULT WINAPI IDirectMusicSignPostTrack_IPersistStream_Save (LPPERSISTSTREAM iface, IStream* pStm, BOOL fClearDirty); > extern HRESULT WINAPI IDirectMusicSignPostTrack_IPersistStream_GetSizeMax (LPPERSISTSTREAM iface, ULARGE_INTEGER* pcbSize); > >+/********************************************************************** >+ * Dll lifetime tracking declaration for devenum.dll >+ */ >+void DMCOMPOS_LockModule(); >+void DMCOMPOS_UnlockModule(); > > /***************************************************************************** > * Misc. >Index: dlls/dmcompos/signposttrack.c >=================================================================== >RCS file: /home/wine/wine/dlls/dmcompos/signposttrack.c,v >retrieving revision 1.7 >diff -u -r1.7 signposttrack.c >--- dlls/dmcompos/signposttrack.c 6 Sep 2004 21:34:26 -0000 1.7 >+++ dlls/dmcompos/signposttrack.c 2 Dec 2004 22:28:45 -0000 >@@ -52,6 +52,9 @@ > ULONG WINAPI IDirectMusicSignPostTrack_IUnknown_AddRef (LPUNKNOWN iface) { > ICOM_THIS_MULTI(IDirectMusicSignPostTrack, UnknownVtbl, iface); > TRACE("(%p): AddRef from %ld\n", This, This->ref); >+ >+ DMCOMPOS_LockModule(); >+ > return ++(This->ref); > } > >@@ -59,9 +62,13 @@ > ICOM_THIS_MULTI(IDirectMusicSignPostTrack, UnknownVtbl, iface); > ULONG ref = --This->ref; > TRACE("(%p): ReleaseRef to %ld\n", This, This->ref); >+ > if (ref == 0) { > HeapFree(GetProcessHeap(), 0, This); > } >+ >+ DMCOMPOS_UnlockModule(); >+ > return ref; > } > > > Rob

Dec. 3, 2004

Re: Wine book

by Lionel Ulmer

On Fri, Dec 03, 2004 at 06:45:47PM +0100, Ivan Leo Puoti wrote: > > The 'high level' parts (like d3drm) should work too, > No program or sdk sample I know of and uses this dll currently works on wine. Well, I should have said then 'there are no technical reasons for D3DRM to not work on Wine' :-) Lionel (who will let someone else debug this if he feels the need) -- Lionel Ulmer - http://www.bbrox.org/

Dec. 3, 2004

Register IME class at startup?

by Ivan Leo Puoti

On windows xp/2k the IME window class appears to be always registered, even if a program doesn't link to nor loads imm32, the class is registered. How should this be handled in wine, should the class be registered at startup? Ivan. ____________________________________________________________ Libero ADSL: 3 mesi gratis e navighi a 1.2 Mega, senza costi di attivazione. Abbonati subito su http://www.libero.it

Dec. 3, 2004

Re: wine-preloader defunct

by Bill Medland

On December 3, 2004 08:17 am, Vincent Béron wrote: > Le ven 03/12/2004 à 10:15, Bill Medland a écrit : > > On December 2, 2004 10:43 pm, Mike McCormack wrote: > > > Kenneth Porter wrote: > > > >> The bug present in Linux 2.6.[0-7] and is fixed in Linux 2.6.8 and > > > >> later. > > > > > > > > Running 2.6.8-1.521smp on Fedora Core 2. > > > > > > When I say Linux 2.6.8 fixes the bug, I mean the standard kernel > > > shipped by Linus. I've got no idea what Redhat does, and they tend to > > > apply alot of patches to their kernel, so it wouldn't suprise me if > > > it's still broken. > > > > > > Mike > > > > Yep. RedHat Enterprise Linux 3 and RedHat Desktop 3 2.4.21-20 (heavily > > modified, as you said!!!) is still broken > > The fixed version is 2.6.8, earlier versions were problematic. I don't > think a 2.4.21 variant can be called "still broken" since it predates > the fix. > > Vincent Yes, but that's the problem with Red Hat; just 'cause it says it's 2.4 doesn't really mean a thing. RHEL3 basically contains, as far as I understant it, a 2.4 kernel with a load of 2.6 stuff hacked into it. (I think RHEL4 is supposed to have a 2.6 kernel when it comes out) -- Bill Medland mailto:billmedland(a)mercuryspeed.com http://webhome.idirect.com/~kbmed

Dec. 3, 2004

Re: Wine book

by Ivan Leo Puoti

> The 'high level' parts (like d3drm) should work too, No program or sdk sample I know of and uses this dll currently works on wine. Ivan. ____________________________________________________________ Libero ADSL: 3 mesi gratis e navighi a 1.2 Mega, senza costi di attivazione. Abbonati subito su http://www.libero.it

Dec. 3, 2004

Re: Implement GetLayout

by Rémi Assailly

Rémi Assailly wrote: >Sorry, I prefer that ;) >What I wanted to know was if my implementation of GetLayout was correct >before trying to implement SetLayout ( and add mirroring in all >functions which need to check it ... ) or what was incorrect to fix it ! >Okay, I will do what you advise me to do. >I found that SetProcessDefaultLayout is not implemented, I think it will >something like > >BOOL WINAPI SetProcessDefaultLayout( DWORD ); in winuser.h > I correct, SetProcessDefaultLayout, and GetProcessDefaultLayout are implemented. I do not know why I did not found them the first time... Sorry...

Dec. 3, 2004

Problem to draw a bitmap

by Rémi Assailly

I tested a program ( an other one, with source code this time ) On my Windows XP system I have no problem. When I use Wine I just get a window. ( where is my bitmap ? ) Then I tried this : "WINEDEBUG=gdi wine progtest" and as Wine is slower I could see my bitmap, which disappeared as fast as it appeared... Here the source code : > #include <windows.h> > #include <wingdi.h> > > void PrintBmp(HWND hWnd, char *filename) > { > HBITMAP hBmp; > HDC hDC; > hBmp=(HBITMAP)LoadImage(NULL,filename,IMAGE_BITMAP,0,0,LR_LOADFROMFILE); > hDC=GetDC(hWnd); > DrawState(hDC,NULL,NULL,(LPARAM)hBmp,NULL,0,0,0,0,DST_BITMAP); > DeleteObject(hBmp); > ReleaseDC(hWnd,hDC); > } > > /* Declare Windows procedure */ > LRESULT CALLBACK WindowProcedure (HWND, UINT, WPARAM, LPARAM); > > /* Make the class name into a global variable */ > char szClassName[ ] = "WindowsApp"; > > > int WINAPI > WinMain (HINSTANCE hThisInstance, > HINSTANCE hPrevInstance, > LPSTR lpszArgument, > int nFunsterStil) > > { > HWND hwnd; /* This is the handle for our window */ > MSG messages; /* Here messages to the application are saved */ > WNDCLASSEX wincl; /* Data structure for the windowclass */ > > /* The Window structure */ > wincl.hInstance = hThisInstance; > wincl.lpszClassName = szClassName; > wincl.lpfnWndProc = WindowProcedure; /* This function is called by windows */ > wincl.style = CS_DBLCLKS; /* Catch double-clicks */ > wincl.cbSize = sizeof (WNDCLASSEX); > > /* Use default icon and mouse-pointer */ > wincl.hIcon = LoadIcon (NULL, IDI_APPLICATION); > wincl.hIconSm = LoadIcon (NULL, IDI_APPLICATION); > wincl.hCursor = LoadCursor (NULL, IDC_ARROW); > wincl.lpszMenuName = NULL; /* No menu */ > wincl.cbClsExtra = 0; /* No extra bytes after the window class */ > wincl.cbWndExtra = 0; /* structure or the window instance */ > /* Use Windows's default color as the background of the window */ > wincl.hbrBackground = (HBRUSH) COLOR_BACKGROUND; > > /* Register the window class, and if it fails quit the program */ > if (!RegisterClassEx (&wincl)) > return 0; > > /* The class is registered, let's create the program*/ > hwnd = CreateWindowEx ( > 0, /* Extended possibilites for variation */ > szClassName, /* Classname */ > "Windows App", /* Title Text */ > WS_OVERLAPPEDWINDOW, /* default window */ > CW_USEDEFAULT, /* Windows decides the position */ > CW_USEDEFAULT, /* where the window ends up on the screen */ > 544, /* The programs width */ > 375, /* and height in pixels */ > HWND_DESKTOP, /* The window is a child-window to desktop */ > NULL, /* No menu */ > hThisInstance, /* Program Instance handler */ > NULL /* No Window Creation data */ > ); > > > /* Make the window visible on the screen */ > ShowWindow (hwnd, nFunsterStil); > PrintBmp(hwnd, "BOARD.bmp"); > /* Run the message loop. It will run until GetMessage() returns 0 */ > while (GetMessage (&messages, NULL, 0, 0)) > { > /* Translate virtual-key messages into character messages */ > TranslateMessage(&messages); > /* Send message to WindowProcedure */ > DispatchMessage(&messages); > } > > /* The program return-value is 0 - The value that PostQuitMessage() gave */ > return messages.wParam; > } > > > /* This function is called by the Windows function DispatchMessage() */ > > LRESULT CALLBACK > WindowProcedure (HWND hwnd, UINT message, WPARAM wParam, LPARAM lParam) > { > switch (message) /* handle the messages */ > { > case WM_DESTROY: > PostQuitMessage (0); /* send a WM_QUIT to the message queue */ > break; > default: /* for messages that we don't deal with */ > return DefWindowProc (hwnd, message, wParam, lParam); > } > > return 0; > }

Dec. 3, 2004

Re: wine-preloader defunct

by Vincent Béron

Le ven 03/12/2004 à 10:15, Bill Medland a écrit : > On December 2, 2004 10:43 pm, Mike McCormack wrote: > > Kenneth Porter wrote: > > >> The bug present in Linux 2.6.[0-7] and is fixed in Linux 2.6.8 and > > >> later. > > > > > > Running 2.6.8-1.521smp on Fedora Core 2. > > > > When I say Linux 2.6.8 fixes the bug, I mean the standard kernel shipped > > by Linus. I've got no idea what Redhat does, and they tend to apply > > alot of patches to their kernel, so it wouldn't suprise me if it's still > > broken. > > > > Mike > > Yep. RedHat Enterprise Linux 3 and RedHat Desktop 3 2.4.21-20 (heavily > modified, as you said!!!) is still broken The fixed version is 2.6.8, earlier versions were problematic. I don't think a 2.4.21 variant can be called "still broken" since it predates the fix. Vincent

Dec. 3, 2004

Re: wine-preloader defunct

by Bill Medland

On December 2, 2004 10:43 pm, Mike McCormack wrote: > Kenneth Porter wrote: > >> The bug present in Linux 2.6.[0-7] and is fixed in Linux 2.6.8 and > >> later. > > > > Running 2.6.8-1.521smp on Fedora Core 2. > > When I say Linux 2.6.8 fixes the bug, I mean the standard kernel shipped > by Linus. I've got no idea what Redhat does, and they tend to apply > alot of patches to their kernel, so it wouldn't suprise me if it's still > broken. > > Mike Yep. RedHat Enterprise Linux 3 and RedHat Desktop 3 2.4.21-20 (heavily modified, as you said!!!) is still broken -- Bill Medland mailto:billmedland(a)mercuryspeed.com http://webhome.idirect.com/~kbmed

Dec. 3, 2004

Re: US export regulations and rsaenh.dll

by Marcus Meissner

On Fri, Dec 03, 2004 at 01:40:42PM +0100, Michael Jung wrote: > On Friday 03 December 2004 13:06, Marcus Meissner wrote: > > Actually, in the closest meaning of the export regulations, > > I am not even allowed to build RPMs for SourceForge anymore > > here at SUSE/Novell :( > > > > Fun without end, I had hoped we keep openssl dynamic > > linking ;) > > Sorry, I didn't mean to cause problems for anybody. Actually, I would not have > a problem with going back to an OpenSSL based implementation. It's just that > there were some compilation problems with older RH versions (which I think > would not be hard to resolve) and if I remember correctly, the ReactOS > project has some license problems with OpenSSL. > > We could also make the selection between OpenSSL/LibTomCrypt a compile time > option (Although, as I understand it, we don't want to many of those, right?) No worries. I expect no problems ;) Ciao, Marcus

Dec. 3, 2004

Re: <<PATCH>> Added shell32.dll functions to browse for unix directory; altered winecfg to use it to select the root directory of a virtual drive

by Mike Hearn

Alexandre Julliard wrote: > Well, there's an IShellFolder interface, and it should probably be > doing something with that. In any case it should definitely be more > integrated into the shell way of doing things, there shouldn't be a > need for a whole new set of APIs. This needs more thought IMO... I have to disagree. I did try and write a shell extension once - even with the Delphi wizards helping with the boilerplate it was still some kind of Buddhist patience-enhancing mental excercise. In the end I trashed Explorer and had to reinstall Windows. There's a *lot* of work involved in writing one. Anyway part of the reason for having the new APIs and restructuring the code is so we can have a non-standard UI. If it's implemented as a shell namespace extension the code will get a lot more complex, and the UI will still show the mapped drives alongside the new namespace which would be confusing to say the least. thanks -mike

Dec. 3, 2004

Re: Providing success stories

by Dan Kegel

Stefan Munz wrote: > In order to push the use of Open Source Software we plan a project with the > "Wirtschaftsförderung Region Stuttgart GmbH" ... to > find software producers who pay us for a wine compatibility test of their > software. ... What's the URL, by the way? Is it http://opensource.region-stuttgart.de/ ? > This project is starting at the moment. I thought a good idea to give > something back to the community is to provide some success stories i.e. for > use on winehq.com. Is this interesting? Yes. I don't see anything on winehq.com like that, but http://www.codeweavers.com/site/products/case_studies seems to be along the lines you're thinking of, maybe? Try just writing up a first draft of a success story on a web page somewhere, and post a link to wine-devel so we can see what you're thinking of. Then I'm sure getting it onto winehq.org will be easy. - Dan -- Trying to get a job as a c++ developer? See http://kegel.com/academy/getting-hired.html

Dec. 3, 2004

Re: US export regulations and rsaenh.dll

by Michael Jung

On Friday 03 December 2004 13:06, Marcus Meissner wrote: > Actually, in the closest meaning of the export regulations, > I am not even allowed to build RPMs for SourceForge anymore > here at SUSE/Novell :( > > Fun without end, I had hoped we keep openssl dynamic > linking ;) Sorry, I didn't mean to cause problems for anybody. Actually, I would not have a problem with going back to an OpenSSL based implementation. It's just that there were some compilation problems with older RH versions (which I think would not be hard to resolve) and if I remember correctly, the ReactOS project has some license problems with OpenSSL. We could also make the selection between OpenSSL/LibTomCrypt a compile time option (Although, as I understand it, we don't want to many of those, right?) Bye, Michael

Dec. 3, 2004

Re: rsaenh/implglue.c: Added missing break

by Marcus Meissner

On Fri, Dec 03, 2004 at 11:49:22AM +0100, Michael Jung wrote: > Greetings, > Michael > > Changelog: > Added a missing break in encrypt_block_impl > > Index: dlls/rsaenh/implglue.c JFYI, we (Novell) are now working on reapproving WINE for US export due to your cryptographic library import. I will keep you posted, but I expect approval. Actually, in the closest meaning of the export regulations, I am not even allowed to build RPMs for SourceForge anymore here at SUSE/Novell :( Fun without end, I had hoped we keep openssl dynamic linking ;) Ciao, Marcus

Dec. 3, 2004

Providing success stories

by Stefan Munz

Hi, I already mailed with some developers a few days ago. Basicly wether and in which way the wine logo can be used to communicate wine compatibility for software producers (this is already solved). For those who didn't read it, i copied a passage of the orig. mail: --- In order to push the use of Open Source Software we plan a project with the "Wirtschaftsförderung Region Stuttgart GmbH" (I don't know whats the equivalent in english but its a public organization that promotes OSS and assists companies that resident in Stuttgart). Within this project we want to find software producers who pay us for a wine compatibility test of their software. --- This project is starting at the moment. I thought a good idea to give something back to the community is to provide some success stories i.e. for use on winehq.com. Is this interesting? If the answer is yes, who is responsible for something of the kind. Regards, Stefan Munz

Dec. 3, 2004

Re: Wine book

by Lionel Ulmer

> Just kidding. I remember Tom saying last year that native DirectX > actually works under Wine. Maybe it's worth covering? Am I way off > base? Well, native DInput somewhat works (for which I am pretty surprised, I would need to trace it to see what Win32 calls it uses) and QUARTZ works too. The 'high level' parts (like d3drm) should work too, but the rest (DDraw, D3D and I think DSound too) would not work. Lionel -- Lionel Ulmer - http://www.bbrox.org/

Dec. 3, 2004

Re: <<PATCH>> Added shell32.dll functions to browse for unix directory; altered winecfg to use it to select the root directory of a virtual drive

by Shachar Shemesh

Steven Edwards wrote: >--- Mike Hearn <mh(a)codeweavers.com> wrote: > > >>What's wrong with additional APIs? The browse folder API isn't COM based >>anyway. >> >> > >It would be much better to use the API provided to create a namespace for the Unix filesystem. The >Control Panel and Network Neighborhood are good examples of this. Quite a few third party apps >create a virtual namespace for things like CVS so I don't see why we can't work in the existing >API to do this with the Unix fs. > >Thanks >Steven > > > It's been a LONG while since I messed with shell extensions. However, if my memory serves me right, you cannot pass virtual namespaces to "CreateFile" and friends. The only one who knows how to work with virtual namespaces is explorer. That may not be much of a problem for the winecfg utility, but it's something to consider. Esp. being that we don't have a proper explorer (and none is planned for Wine that I know of, and ReactOs doesn't seem to need Posix filenames...) Shachar -- Shachar Shemesh Lingnu Open Source Consulting ltd. http://www.lingnu.com/

Dec. 3, 2004

Re: Kernel 2.6.10... and amd64

by Jose Alberto Reguero

I read in the list that ptrace break wine in this kernels, and find a patch for this, but the patch is only for i386 kernels. There is a patch for amd64 kernels? Thanks. El Jueves, 2 de Diciembre de 2004 18:50, Jose Alberto Reguero escribió: > I can't use wine with kernels 2.6.10-rc2.... > With kernel 2.6.9 it works. > My system is an amd64 linux. > Thanks. > Jose Alberto

Dec. 3, 2004

Re: wine-preloader defunct

by Mike McCormack

Kenneth Porter wrote: >> The bug present in Linux 2.6.[0-7] and is fixed in Linux 2.6.8 and later. > Running 2.6.8-1.521smp on Fedora Core 2. When I say Linux 2.6.8 fixes the bug, I mean the standard kernel shipped by Linus. I've got no idea what Redhat does, and they tend to apply alot of patches to their kernel, so it wouldn't suprise me if it's still broken. Mike

Dec. 3, 2004

Re: wine-preloader defunct

by Kenneth Porter

--On Thursday, December 02, 2004 11:15 PM -0500 James Hawkins <truiken(a)gmail.com> wrote: > Does 'kill -9 processnum' not work? Son of a gun, that did it. I'm so used to avoiding that that it didn't occur to me to try it.

Dec. 3, 2004

Re: wine-preloader defunct

by Kenneth Porter

> --On Friday, December 03, 2004 3:18 PM +0900 Mike McCormack > <mike(a)codeweavers.com> wrote: > The bug present in Linux 2.6.[0-7] and is fixed in Linux 2.6.8 and later. Running 2.6.8-1.521smp on Fedora Core 2.

Dec. 3, 2004

Re: wine-preloader defunct

by Mike McCormack

>>How can kill -9 not work? It should kill the process without question >>unless I'm wrong. > > I think this is a kernel bug that was fixed in more recent kernels. The bug present in Linux 2.6.[0-7] and is fixed in Linux 2.6.8 and later. Mike

Dec. 3, 2004

Re: wine-preloader defunct

by Steven Edwards

Hi, --- James Hawkins <truiken(a)gmail.com> wrote: > > Probably not. > > I regularly face them and the only way I know is to reboot > > How can kill -9 not work? It should kill the process without question > unless I'm wrong. I think this is a kernel bug that was fixed in more recent kernels. Thanks Steven __________________________________ Do you Yahoo!? Yahoo! Mail - You care about security. So do we. http://promotions.yahoo.com/new_mail

Dec. 3, 2004

Re: wine-preloader defunct

by James Hawkins

On Thu, 2 Dec 2004 21:28:01 -0800, Bill Medland <billmedland(a)mercuryspeed.com> wrote: > On December 2, 2004 08:15 pm, James Hawkins wrote: > > > > On Thu, 02 Dec 2004 20:01:48 -0800, Kenneth Porter > > > > <shiva(a)sewingwitch.com> wrote: > > > I've now got several instances of "[wine-preloader] <defunct>" in my ps > > > listing from a game server that's died in some strange way. Anyone have > > > experience in clearing these short of bouncing the whole box? They're > > > holding onto a couple of TCP ports that I'd like to use. The ports are > > > stuck in CLOSE_WAIT, because the processes are stuck in memory so the > > > sockets can't close. > > > > Does 'kill -9 processnum' not work? > > Probably not. > I regularly face them and the only way I know is to reboot > > -- > Bill Medland > mailto:billmedland(a)mercuryspeed.com > http://webhome.idirect.com/~kbmed > How can kill -9 not work? It should kill the process without question unless I'm wrong. -- James Hawkins

Dec. 2, 2004

Preliminary Darwine SDK Released

by Sanjay Connare

Hello All, I have released a preliminary version of the Darwine SDK version 0.1. Many people are interested in getting Darwine up and running so they can compile programs on OS X and this release should satisfy their curiosity. The SDK features the following: -source code for the current Darwine binary 20040820 -source code for WineHelper -Xcode Integration -Xcode Template -Updated Uninstaller The Xcode integration allows for a user to compile windows source code from within the Xcode IDE using the wine tools. Users may want to note that there are some changes to the way Darwine is installed and functions. Due to limitations in Xcode, the wine source tools such as winegcc, winemaker, winebuild are installed in the /usr/bin directory and not the /Library/Darwine/bin directory. This may be changed in a future release but was necessary so Xcode integration would work correctly. As a result there is an updated uninstaller as well. The uninstaller will remove everything except for the Xcode project template folder, which is moved to the trash and needs to be removed manually. To start a new Darwine project, simply launch Xcode, select new project, and select Darwine (Windows) Application. Currently in Xcode the "Build and Run" does not work, it will build the program but not launch. However, the Build button works as expected. When compiling a Darwine application for the first time Xcode may give the error: "ld: table of contents for archive: /Library/Darwine/lib/wine/lib<name>.a is out of date; rerun ranlib(1) (can't load from it)" Open a terminal and type the following command: "sudo ranlib /Library/Darwine/lib/wine/lib<name>.a" <name> being replaced by the name of the library whose table of contents is out of date. The source code should then compile fine. There are many examples of source in the wine source code "programs" folder which is located in the root of the source directory. Simple programs should compile fine. However, bigger programs that depend on more of the windows headers will probably not compile due to header incompatabitlies between the wine and OS X headers. You can download the preliminary release here http://prdownloads.sourceforge.net/darwine/Darwine_SDK0.1.dmg?download or through the Darwine web site http://darwine.opendarwin.org Post any comments, questions or concerns to the Darwine Mailing List "darwine-devel(a)lists.sourceforge.net" . Remember this is a preliminary release and there are many bugs and issues that need to be worked out and features that need to be implemented. More information will be posted on the Darwine website in the near feature. Last but not least enjoy. Sanjay Connare

Dec. 2, 2004

Re: wine-preloader defunct

by Bill Medland

On December 2, 2004 08:15 pm, James Hawkins wrote: > On Thu, 02 Dec 2004 20:01:48 -0800, Kenneth Porter > > <shiva(a)sewingwitch.com> wrote: > > I've now got several instances of "[wine-preloader] <defunct>" in my ps > > listing from a game server that's died in some strange way. Anyone have > > experience in clearing these short of bouncing the whole box? They're > > holding onto a couple of TCP ports that I'd like to use. The ports are > > stuck in CLOSE_WAIT, because the processes are stuck in memory so the > > sockets can't close. > > Does 'kill -9 processnum' not work? Probably not. I regularly face them and the only way I know is to reboot -- Bill Medland mailto:billmedland(a)mercuryspeed.com http://webhome.idirect.com/~kbmed

Dec. 2, 2004