September 2005 - Wine-Devel

Re: patches

by Eric Pouech

Ivan Leo Puoti a écrit : > Eric Pouech wrote: > >> Ivan Leo Puoti a écrit : >> >>> Is anything wrong with these? >>> http://www.winehq.com/pipermail/wine-patches/2005-September/020521.html >>> http://www.winehq.com/pipermail/wine-patches/2005-September/020550.html >> >> >> >> DDK isn't public information. Do we want Wine tree to be tainted by >> non cleanly reversed engineered information ? >> A+ >> > Excuse my language but WTF? The EULA of the DDK is very similar to that > of the SDK, it's freely available for the cost of shipping and handling, > and we can write our own DDK headers just as we can write the SDK ones. > > Ivan. > > from EULA: > 1.1 General License Grant. Microsoft grants to you a personal, non-exclusive, nontransferable, royalty-free license to use the Software, and to make and use five (5) copies of the Software on one or more computers located at your premises solely for the purpose of designing, developing and testing drivers that operate in conjunction with the Software for use with Microsoft Windows 2000 Professional, Microsoft Windows 2000 Server, Microsoft Windows 2000 Advanced Server and Microsoft Windows 2000 Datacenter Server; Microsoft Windows XP, Microsoft Windows XP Service Pack 1; Microsoft Windows Server 2003 Standard Edition, Microsoft Windows Server 2003 Enterprise Edition and Microsoft Windows Server 2003 Datacenter Edition operating system products and any Microsoft operating system product that is a successor to any of the foregoing (each an "OS Product"). isn't that enough ? A+ -- Eric Pouech

Sept. 10, 2005

LTSpice/SwitcherCad III Regression with Rich text patch ID=19048

by paul

I did regression testing and found patch Id=19048 triggers this error wine-pthread: run.c:142: ME_RunOfsFromCharOfs: Assertion `nCharOfs >= nParaOfs' failed. filed http://bugs.winehq.org/show_bug.cgi?id=3239 The application is a free download and the bug has steps to generate the failure. It is still a problem with cvs from 20050909. Paul

Sept. 10, 2005

Re: patches

by Ivan Leo Puoti

Eric Pouech wrote: > Ivan Leo Puoti a écrit : > >> Is anything wrong with these? >> http://www.winehq.com/pipermail/wine-patches/2005-September/020521.html >> http://www.winehq.com/pipermail/wine-patches/2005-September/020550.html > > > DDK isn't public information. Do we want Wine tree to be tainted by > non cleanly reversed engineered information ? > A+ > Excuse my language but WTF? The EULA of the DDK is very similar to that of the SDK, it's freely available for the cost of shipping and handling, and we can write our own DDK headers just as we can write the SDK ones. Ivan.

Sept. 10, 2005

listbox APIs

by Ivan Leo Puoti

I installed the R2 upgrade for win2k3 today, and the listbox tests don't pass anymore, as shown here http://test.winehq.com/data/200509101000/#2003 R2 is the one on the right. Ivan.

Sept. 10, 2005

Possible solution for bug 3028

by Karsten Elfenbein

Hi, after a lot of trail & error I found out that in dlls/kernel/file.c line 1483-1484 the filename/mask is removed. After removing those two codelines the mask was handed over to wine_nt_to_unix_file_name (dlls/ntdll/directory.c) In the wine_nt_to_unix_file_name function it exits at the invalid char check in line 1418-1419 because a "*" is in the file name. If I comment out that loop the updated test case is producing the same error codes as windows xp. Does anyone have a better solution to solve this? (why is the mask removed in dlls/kernel/file.c? why is the "*" an illegal char in dlls/ntdll/directory.c check?) Karsten http://bugs.winehq.org/show_bug.cgi?id=3028

Sept. 10, 2005

Re: Relax Marlett LOGFONT specification

by Richard Cohen

Huw Davies wrote: > On Fri, Sep 09, 2005 at 12:46:10PM +0100, Richard Cohen wrote: >> .. >> Changing SYMBOL_CHARSET -> DEFAULT_CHARSET, FIXED_PITCH -> DEFAULT_PITCH >> .. > > We shouldn't need to do this. Fixing marlett.ttf would be the right > thing to do. > ... I agree that marlett.ttf should be fixed, but I think the patch is correct, because ... 1. AFAICS bug#3284 is caused by a bug in fontforge that was fixed before 20050502, rather than an error in marlett.sfd. The patch works around the broken fontforge. 2. Marlett is not FIXED_PITCH. 3. MSDN says in CreateFont: > Windows 95/98/Me: You can use the DEFAULT_CHARSET value to allow > the name and size of a font to fully describe the logical font. If the > specified font name does not exist, a font from any character set can > be substituted for the specified font, so you should use > DEFAULT_CHARSET sparingly to avoid unexpected results. > > Windows NT/2000/XP: DEFAULT_CHARSET is set to a value based on the > current system locale. For example, when the system locale is English > (United States), it is set as ANSI_CHARSET. > > Fonts with other character sets may exist in the operating system. > If an application uses a font with an unknown character set, it should > not attempt to translate or interpret strings that are rendered with > that font. > > To ensure consistent results when creating a font, do not specify > OEM_CHARSET or DEFAULT_CHARSET. If you specify a typeface name in the > lpszFace parameter, make sure that the fdwCharSet value matches the > character set of the typeface specified in lpszFace. The Wine font mapper seems to copy the Win98 behaviour; specifying SYMBOL_CHARSET sort of implies that the font mapper can fall back to any font with SYMBOL_CHARSET if it can't find Marlett, which really isn't the case. 4. (Appeal to authority!) Despite the warnings against using DEFAULT_CHARSET, a googlefight gives about 53 for Marlett SYMBOL_CHARSET about 122 for Marlett DEFAULT_CHARSET (apparently including MFC -- see http://www.winehq.org/hypermail/wine-patches/2004/03/0010.html ) Richard.

Sept. 10, 2005

patches

by Ivan Leo Puoti

Is anything wrong with these? http://www.winehq.com/pipermail/wine-patches/2005-September/020521.html http://www.winehq.com/pipermail/wine-patches/2005-September/020550.html Ivan

Sept. 10, 2005

Re: winspool/tests: New Testsuite

by Huw D M Davies

On Fri, Sep 09, 2005 at 12:50:26AM +0200, Detlef Riekenberg wrote: > Am Montag, den 29.08.2005, 00:53 +0200 schrieb Detlef Riekenberg: > > > This Patch in not in the CVS yet. > > http://www.winehq.org/pipermail/wine-patches/2005-August/020289.html > > > Changelog: > > > > - New Testsuite for winspool.drv: > > GetDefaultPrinterA/W and > > GetPrinterDriverDirectoryA/W > > - Verbose Output with "WINETEST_DEBUG" > 1 > > > > > Any suggestions / comments on this work ? It's almost a 3000 line patch. You should try to break this down it separate, smaller patches. Huw.

Sept. 10, 2005

Re: winspool: EnumPrinterDrivers

by Huw D M Davies

On Fri, Sep 09, 2005 at 12:53:54AM +0200, Detlef Riekenberg wrote: > Am Sonntag, den 07.08.2005, 19:16 +0200 schrieb Detlef Riekenberg: > > This Patch is not in CVS yet: > > http://www.winehq.org/pipermail/wine-patches/2005-August/019671.html > > > Changelog: > > Let EnumPrinterDrivers return the number of drivers. > > > > > Any comments / suggestions ? Looks good to me. I suggest you resend the patch. Huw.

Sept. 10, 2005

Re: Relax Marlett LOGFONT specification

by Huw Davies

On Fri, Sep 09, 2005 at 12:46:10PM +0100, Richard Cohen wrote: > Index: dlls/user/uitools.c > =================================================================== > RCS file: /home/wine/wine/dlls/user/uitools.c,v > retrieving revision 1.7 > diff -u -p -r1.7 uitools.c > --- dlls/user/uitools.c 7 Jul 2005 18:24:12 -0000 1.7 > +++ dlls/user/uitools.c 9 Sep 2005 11:41:49 -0000 > @@ -926,9 +926,9 @@ static BOOL UITOOLS95_DrawFrameCaption(H > return FALSE; > } > > - hf = CreateFontW(-SmallDiam, 0, 0, 0, FW_NORMAL, FALSE, FALSE, FALSE, > - SYMBOL_CHARSET, OUT_DEFAULT_PRECIS, CLIP_DEFAULT_PRECIS, > - DEFAULT_QUALITY, FIXED_PITCH|FF_DONTCARE, glyphFontName); > + hf = CreateFontW(-SmallDiam, 0, 0, 0, FW_DONTCARE, FALSE, FALSE, FALSE, > + DEFAULT_CHARSET, OUT_DEFAULT_PRECIS, CLIP_DEFAULT_PRECIS, > + DEFAULT_QUALITY, DEFAULT_PITCH|FF_DONTCARE, glyphFontName); > alignsave = SetTextAlign(dc, TA_TOP|TA_LEFT); > bksave = SetBkMode(dc, TRANSPARENT); > clrsave = GetTextColor(dc); We shouldn't need to do this. Fixing marlett.ttf would be the right thing to do. I'll try to look at this next week (but feel free to beat me to it!). Huw.

Sept. 10, 2005

Re: [Now ns7] Dragon ns8 problem

by peter＠piments.com

On Fri, 09 Sep 2005 16:27:47 +0200, Jean Magnan de Bornier <jean(a)bornier.net> wrote: > Now it won't start, but I'm being confident: this thing accepts one more > thing every day, not more; so maybe tomorrow, after a reboot (probably > unneeded, but there seems to be a real addiction about reboot for these > strange animals!), it will run ok, if not I might post the logs... > > thanks, A reboot is hardly ever needed on Linux is you know what to kill/restart. If you dont , it can often be the quickest solution on a desktop system. I've got my WINE/NS closing down pretty cleanly from the menu now. I let it spew out a few messages to the console and then I have to cntl_c it. A couple of seconds later there is no debris on "ps ax" . Re-running wine natspeak afterwards is as good as new. I think that is pretty impressive , hats of to all who have contributed to this effort over the years. And many thanks. -- Using Opera e-mail on Gentoo Linux

Sept. 9, 2005

Re: [Now ns7] Dragon ns8 problem

by wino＠piments.com

On Fri, 09 Sep 2005 16:27:47 +0200, Jean Magnan de Bornier <jean(a)bornier.net> wrote: > > Oh joy! My user's files are there, waiting for me! > > Now it won't start, but I'm being confident: this thing accepts one more > thing every day, not more; so maybe tomorrow, after a reboot (probably > unneeded, but there seems to be a real addiction about reboot for these > strange animals!), it will run ok, if not I might post the logs... > > thanks, Okay, I'm glad to hear getting somewhere. Just refresh my memory about all this business with sound levels and voice training, I have just installed a new user profile. This is also prompted by hardware change since I'd installed a new soundcard. The new Hercules news 5.1 is quite a full featured card but it seems input sensitivity was too low and I couldn't get decent training quality. So after some messing around with the ALSa configurations, I managed to get those soundcard is correctly working with the mixer. I then plugged the microphone into the end sonic soundcard, fired up Dragon NaturallySpeaking has started to create any user profile and run the audio setup process. Here are a few tricks I forgot to mention: Although NaturallySpeaking seems to pick up the sound advice and sets the sound capture level to zero as it enters the audio setup it seems unable to bring it back up again. If this happens, let it get to the stage and then use the Linux mixer to set the capture to suitable level. You should then be able to continue with a quality check. On the Hercules card I was not able to score more than 14, with the Sound Blaster end sonic I was able to score 21. I also encounter some problems which seemed to be more due to Wine than anything else. For example, at one stage I thought it had hung because clicking on buttons had no effect, eventually I remembered having this problem before and getting around it weighs in theor keyboard shortcuts. In any case, I successfully completed a training and this text is my first attempt at using it. As you can see the quality of the dictation recognition is pretty good. None of this text has been edited with the keyboard afterwards, it was all generated by Dragon NaturallySpeaking, and the any editing done was the occasional command issued through the microphone. I will now risk all by trying to update wine to see if you remove some of the glitches. I'll keep you posted, good luck.

Sept. 9, 2005

Re: Need help debugging a memory corruption bug in shfldr_unixfs.c

by Dimi Paun

From: "Phil Krylov" <phil(a)newstar.rinet.ru> > OK, should we then make a patch which adds an IsWindow() check after every > notification in every control? This one fixes a really annoying bug, while > the other places are not proved to crash. Well, if it's a entire class of bugs, we should fix them all. I guess we need a bit of experimentation on windows to see if we need to protect all notifications or only a few ones. I guess we can protect all of them, in which case we have to look to see how we can do it centrally, in the notification function itself. -- Dimi.

Sept. 9, 2005

Re: Need help debugging a memory corruption bug in shfldr_unixfs.c

by Phil Krylov

On Fri, 9 Sep 2005 13:59:05 -0400 "Dimi Paun" <dimi(a)lattica.com> wrote: > From: "Phil Krylov" <phil(a)newstar.rinet.ru> > > Here is a patch which adds checking if the window has been destroyed at > > that point. I don't know if it is acceptable but it fixes the problem. > > But this could potentially happen on any notification, > what makes this particular one special (sorry, I was away > on vacation and I missed this thread). OK, should we then make a patch which adds an IsWindow() check after every notification in every control? This one fixes a really annoying bug, while the other places are not proved to crash. -- Ph.

Sept. 9, 2005

Re: Empire Earth continued

by Lionel Ulmer

On Thu, Sep 08, 2005 at 10:37:56PM +0200, Stefan Dösinger wrote: > I think you're right. After Releasing the DDraw object two times, as EE does, > it's far from beeing freed: It has a refcount of 8!! In particular, creating > the surface and performing QueryInterface on it increase the refcount by 1, > creating a D3D device increases the refcount by 7. What you could test is artifically increase the refcount of 'normal' surfaces and check if these are deleted or not. Lionel -- Lionel Ulmer - http://www.bbrox.org/

Sept. 9, 2005

Re: Need help debugging a memory corruption bug in shfldr_unixfs.c

by Dimi Paun

From: "Phil Krylov" <phil(a)newstar.rinet.ru> > Here is a patch which adds checking if the window has been destroyed at > that point. I don't know if it is acceptable but it fixes the problem. But this could potentially happen on any notification, what makes this particular one special (sorry, I was away on vacation and I missed this thread). -- Dimi.

Sept. 9, 2005

Re: Need help debugging a memory corruption bug in shfldr_unixfs.c

by Duane Clark

Phil Krylov wrote: > Hi Michael, > > On Thu, 8 Sep 2005 23:10:18 +0200 > Michael Jung <mjung(a)iss.tu-darmstadt.de> wrote: > >> Wouldn't it be enough to call notify_click after notify_itemactivate? I've >> attached a modification of your patch, which does just this. Seems to work >> fine for me. > > Probably, but what if some apps depend on order in which these > notifications are sent? (assuming we already send them in the right order) > Comparing the Listview control spy on Wine and Win2k, the notifications appear to be in the correct order in Wine: NM_RELEASEDCAPTURE NM_DBLCLK LVN_ITEMACTIVE

Sept. 9, 2005

Re: Suggestions for improvement of the emulator

by Evil

Mike Hearn wrote: >>From a dictionary: > >"em·u·late: Computer Science. To imitate the function of (another >system), as by modifications to hardware or software that allow the >imitating system to accept the same data, execute the same programs, and >achieve the same results as the imitated system." > >Seems like a good description of Wine to me. It imitates the Win32 API. > > By this definition: Windows XP = Windows 9x Emulator Windows 9x = Dos Emulator etc... In fact, this definition makes a confusing amount of software and hardware an "emulator" solely by virtue of conforming to standards or through backwards compatibility. There's not even an exception that says the "same program" has to be binary code. So, every interpretive programming language, but first implementations, is an "emulator". I refuse to use such a loose definition, probably written by an English major who didn't fully comprehend what he was writing, simply because he was paid to write it. I don't think any of us should, unless you also agree that purchased software always does things more correctly than free software. :) So we need to make sure the media gets it right: "em·u·late: Computer Science, Software: Imitation of the function of (another system), by dynamic recompilation or interpretive translation as to allow the imitating system to accept the same data, execute the same binary programs, and achieve the same results as the imitated system."

Sept. 9, 2005

Re: Suggestions for improvement of the emulator

by Steven Edwards

On 9/9/05, Dmitry Timoshkov <dmitry(a)baikal.ru> wrote: > The better word is "clone" if you don't like a wordy "independent implementation". Heh, no that would be ReactOS. I always like the term "compatibility layer" when speaking of Wine. Thanks Steven

Sept. 9, 2005

Re: Safedisc 1 works on wine

by Ivan Leo Puoti

Oh, and here's an actualy in game shot, if you're a fan of the game you'll probably notice I'm out of practice. http://www003.portalis.it/115/wine/safediscworks2.png Ivan.

Sept. 9, 2005

Re: user: make A->W WM_GETTEXTLENGTH use WM_GETTEXT [try 2]

by Frank Richter

On 09.09.2005 17:09, Alexandre Julliard wrote: > Do a CallWindowProc on a function that is not the current winproc of > the specified window and check if the message gets to the window. Ah, I see. -f.r.

Sept. 9, 2005

Re: user: make A->W WM_GETTEXTLENGTH use WM_GETTEXT [try 2]

by Alexandre Julliard

Frank Richter <frank.richter(a)gmail.com> writes: > On 09.09.2005 16:27, Alexandre Julliard wrote: >> Well, I'm not sure doing a SendMessage is the right thing if you are >> called from inside CallWindowProc, you may have to do a CallWindowProc > > UnmapMsgAtoW() (or so) is also called when interprocess messages are > translated back... in this case, (afaik... correct me when I'm wrong) a > CallWindowProc() does not suffice... Yes, that's what I mean. When unmapping the result of a SendMessage you should do a SendMessage, when unmapping the result of a CallWindowProc you should do a CallWindowProc (I guess, this needs to be checked under Windows). > Hm... not certain how I could determine with what facility(ie > CallWndProc() vs SendMessage() vs something else?) a message (in this > case, WM_GETTEXT) was sent to a window. Do a CallWindowProc on a function that is not the current winproc of the specified window and check if the message gets to the window. -- Alexandre Julliard julliard(a)winehq.org

Sept. 9, 2005

Re: user: make A->W WM_GETTEXTLENGTH use WM_GETTEXT [try 2]

by Frank Richter

On 09.09.2005 16:27, Alexandre Julliard wrote: > Well, I'm not sure doing a SendMessage is the right thing if you are > called from inside CallWindowProc, you may have to do a CallWindowProc UnmapMsgAtoW() (or so) is also called when interprocess messages are translated back... in this case, (afaik... correct me when I'm wrong) a CallWindowProc() does not suffice... > instead. I think a test case will be needed to find out what Windows > does. Hm... not certain how I could determine with what facility(ie CallWndProc() vs SendMessage() vs something else?) a message (in this case, WM_GETTEXT) was sent to a window. -f.r.

Sept. 9, 2005

Re: [Now ns7] Dragon ns8 problem

by Jean Magnan de Bornier

Le 08 septembre à 08:53:19 wino(a)piments.com écrit notamment: | On Wed, 07 Sep 2005 14:21:26 +0200, Jean Magnan de Bornier | <jean(a)bornier.net> wrote: > | > Sound level appears now ok for dns; | Great news. > | > I have to read two phrases, and then a longer text; in this last | > exercise dns gets lost after two lines of text. All the cpu is taken by | > wine-preloader processes and this does not appears to ever stop. There | > were (last time) 23 wine-preloader processes running! > | I have never had that sort of problem. | I cant recall the official minimum mem req for dns7 but as a guide I would | expect it to crawl with 256M and fly with 512M > | Speach rec is memory hungry process and will speed up considerably if | given more RAM, but I have not seen it get stuck as you describe. > | > Would it be something like not enough space allocated to wine?? > | Unless you are near zero on that partition I dont think that is likely. > | It may be worth limitting the number of unknows if you want to get it | working quicker. > | I would suggest you try a known, working combination of wine version , | sound drivers and get it to working and then try to update wine and try | different drivers if you wish. This advice about wine versions has been precious: it was certainly naive to install the program with one version of wine and configure it with another one. I tried the may version to reinstall dns7, but that couldn't be done; then I reverted to the july version, and I both installed and finally configured dns! Oh joy! My user's files are there, waiting for me! Now it won't start, but I'm being confident: this thing accepts one more thing every day, not more; so maybe tomorrow, after a reboot (probably unneeded, but there seems to be a real addiction about reboot for these strange animals!), it will run ok, if not I might post the logs... thanks, -- Jean Magnan de Bornier | Cours Victor Hugo e-mots: jean at bornier.net | 13980 Alleins France T 08 70 39 34 03 | P 06 09 17 35 87

Sept. 9, 2005

Re: user: make A->W WM_GETTEXTLENGTH use WM_GETTEXT [try 2]

by Alexandre Julliard

Frank Richter <frank.richter(a)gmail.com> writes: > On 05.09.2005 18:13, Frank Richter wrote: >> Frank Richter <frank.richter(a)gmail.com> >> When doing A->W WM_GETTEXTLENGTH, use WM_GETTEXT behind the scenes to >> obtain an exact length. This seems to match the behaviour of recent >> Windowses and as well fixes a couple of regressions (caused by theming >> suddenly making the standard controls Unicode). > > Any comment on this one? I mean, getting Delphi apps to work again would > be nice after all... Well, I'm not sure doing a SendMessage is the right thing if you are called from inside CallWindowProc, you may have to do a CallWindowProc instead. I think a test case will be needed to find out what Windows does. -- Alexandre Julliard julliard(a)winehq.org

Sept. 9, 2005