- Wine-devel - list.winehq.org

Dynamically linking with a Unix library - or not
by Shachar Shemesh 19 Jun '02

19 Jun '02

Hi WINE-devel list (cross posted to Linux-IL), After a lively discussion in the Israeli Linux users mailing list, it appears that the best solution will be to give a command line (or config) option to use an external library (fribidi) for the RTL rendering. I would still implement this internally, as the extrenal library is more Unicode and less Windows complient, but that's besides the point. Adding a ./configure option to do so would probably be quite feasable (I don't know how to do it at the moment, but I am sure that will not be the disqualifying factor), but I have been quite hesitant to go that way until now, for the following reasons: A. Fribidi will have to be present during building WINE. That, in itself, is not a majour obstacle, but B. Having Fribidi present during compile will add Fribidi to the WINE dependancies. I am afraid that the consequence of B will be that no packager in his right mind will use that option. As that works in direct contradiction to the reason I entered the WINE project to begin with, I am trying to opt for a better solution. I'm seeing two possible solutions to this problem: A. If Fribidi was present during compile, check for its existance during run time. If not present, don't enable the run time option. or B. Copy (port?) Fribidi into the WINE code. It's LGPL, so the license does allow that. The consequence of B is that I will need to keep an eye on Fribidi, and keep it up to date whenever something changes there. I am very reluctant to do so. I would much rather go for the "A" solution. The problem is that I know how to do this in Windows, but not in Linux. Any help anyone can offer me? Shachar

5 10

RE: Configure problems on MSYS
by Steven Edwards 18 Jun '02

18 Jun '02

Still having the problem.....did a clean checkout. No DIff here is the MD5sum $ md5sum.exe configure 3493dd32fa6c3299459c78a9c4cd4e79 *configure __________________________________________________ Do You Yahoo!? Yahoo! - Official partner of 2002 FIFA World Cup http://fifaworldcup.yahoo.com

1 0

What happens with...
by Medland, Bill 18 Jun '02

18 Jun '02

Can someone save me the trouble of hunting and experimenting on the following please? Our program currently is installed through installshield. It does the usual loading of newer Microsoft DLLs etc as well as of our own objects. Rather than registering them immediately it adds the registration tasks to the RunOnce part of the registry and requires a reboot. When I had a go a couple of days ago I didn't actually "do the reboot". (At the time I didn't know how our installer worked) and registered the dlls etc by hand (when I figured out that they weren't). Can anyone tell me if the "RunOnce" registry section works and if "rebooting" in the Windows sense actually reboots the wine server? Bill

2 1

PATCH: fix for RPM
by Pavel Roskin 18 Jun '02

18 Jun '02

Hello, Kristian! I have downloaded the latest Wine RPM from http://wine.dataparty.no/ It doesn't work at all, at least in the configuration without Windows. The trace shows that Wine cannot load KERNEL32.dll because it looks in /tmp/wine-rpm-root/usr/lib/wine/ instead of /usr/lib/wine/: 08072a70:warn:module:BUILTIN32_dlopen cannot open .so lib for builtin kernel32.dll: /tmp/wine-rpm-root/usr/lib/wine/kernel32.dll.so: cannot open shared object file: No such file or directory 08072a70:warn:module:MODULE_LoadLibraryExA Failed to load module 'C:\WINDOWS\SYSTEM\KERNEL32.dll'; error=2 08072a70:trace:heap:RtlFreeHeap (40370000,00000002,40370e50): returning TRUE 08072a70:err:module:PE_fixup_imports Module (file) KERNEL32.dll (which is needed by C:\windows\IsUninst.exe) not found Wine should not be configured with the /usr prefix. The prefix should only be changed to $RPM_BUILD_ROOT/usr during the install. I'm copying my message to the list, because I'm afraid that the RPMs on http://wine.dataparty.no/ are not really maintained. I remember seeing the same error a few months ago, I just didn't have time to report it. How can this error remain unfixed for months, if it makes Wine useless? This patch has been tested. The resulting RPM works. ================================== --- wine.spec.unstrip +++ wine.spec.unstrip @@ -18,14 +18,14 @@ %setup -q %build -./configure --without-curses --prefix=$RPM_BUILD_ROOT/usr \ +./configure --without-curses --prefix=/usr \ --sysconfdir=/etc/wine make depend make CFLAGS='-O2 -Wall' %install rm -rf $RPM_BUILD_ROOT -make install +make install prefix=$RPM_BUILD_ROOT/usr mkdir -p $RPM_BUILD_ROOT/etc/wine cp documentation/samples/config $RPM_BUILD_ROOT/etc/wine/config cp tools/fnt2bdf $RPM_BUILD_ROOT/usr/bin ================================== -- Regards, Pavel Roskin

3 3

__CxxFrameHandler
by juergen.schmied＠debitel.net 18 Jun '02

18 Jun '02

Hi! After getting problems with many aplications generated with VC i did a search for informations and found our implementation is not complete. The information is at: http://www.thecodeproject.com/cpp/Exceptionhandler.asp there is a example code for a replacement of the VC exception handling what looks like it could be ported to wine. Anybody interested in looking into it? juergen PS: created BUG 789 for it.

1 0

make test and ~/.wine/config
by Sylvain Petreolle 18 Jun '02

18 Jun '02

Hi, A little thought about running test and the $HOME config file: If you want to run make test after changing a parameter in the config file, you must run "make testclean", change the parametrer, and rerun make test. So couldn't we include $HOME/.wine/config in the dependancies ? ___________________________________________________________ Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! Yahoo! Mail : http://fr.mail.yahoo.com

1 0

(Demo CD available) Has anyone tested AutoCad 2000 with any
by Matts 18 Jun '02

18 Jun '02

>> AutoDesk now offers a free 15 day demo CD of AutoCad. I hope some of the >> talanted wine developers take the opportunity to get the CD and try to get >> it to work. >> >> http://usa.autodesk.com/adsk/item/0,,616392-123112,00.html >> >> In my opinion there are two windows application that are more important >> than others and that is AutoCad and Quark express. Linux does not have good >> alternatives to theese, and for instance AutoCad users are engineers which >> are interested in technics and I now several that would like to ha linux at >> work but cant since it cant run AutoCad. >> >> --- Matts <kivik(a)firstlinux.net> wrote: >> >Great, it sounds as if it can be working seriously sooon. >> > >> >Is there any way a non wine developer can help with point 3 in the HOWTO ? >> > = (ie finding why ARX apps hang). Solving that will solve a lot of the >> > other = points and it will be closer to wider usage. >> > >> >3) Open SYSTEM.REG file in ~/.wine and remove ALL lines with >> >LOADCTRLS=3D....... and save it. >> >Don't ask why, I dunno but this prevents loading of an ARX >> >app that hangs Autocad. >> > >> >The FONTS issue is a one time problem, and the regen might not be to hard >> > t= o solve. (dirty hack, somehoe regen at each mouserelease). >> > >> >--- Sylvain Petreolle <spetreolle(a)yahoo.fr> wrote: >> >>see >> >>http://appdb.codeweavers.com/appview.php?appId=3D86&versionId=3D102 >> >> >> >>> Once upon a time I got Autocad R13 to work, only >> >>> toolbars and screenrefresh was not working to 100%, >> >>> then it broke completely, and 10 months ago when I >> >>> tested with AutoCad 2000 it was still broken.=20 >> >>>=20 >> >>> Has anyone done any testing recently ? >> >>>=20 >> >>> AutoCad and Quark express are two very important >> >>> pieces of software for wine, since they both lack >> >>> useable alternatives natively built for Linux. _____________________________________________________________ Want a new web-based email account ? ---> http://www.firstlinux.net _____________________________________________________________ Promote your group and strengthen ties to your members with email(a)yourgroup.org by Everyone.net http://www.everyone.net/?btn=tag

1 0

Re: please report this (quicken error)
by Andriy Palamarchuk 17 Jun '02

17 Jun '02

Shane, thank you very much for this information! --- Shane Shields <locutusenterprises(a)softhome.net> wrote: > hi all > > ive been lurking for a while and reading some very > lively discussions but > thats off topic. the reason ive come out of my hole > is the following > messages. just being a good (?) citizen. let me know > what other information > you need and how i can collect it. Would it be difficult for you to submit a bug report with description of the problem to http://bugs.winehq.com? The reason I'm asking you is that I don't know if anybody can handle this information right now and I don't want your report to be lost. In Bugzilla your bug reports are always processed. Again, thank you for the bug report, Andriy Palamarchuk __________________________________________________ Do You Yahoo!? Yahoo! - Official partner of 2002 FIFA World Cup http://fifaworldcup.yahoo.com

1 0

PowerBuilder 6.0 generated applications crash accessing date fields
by Luigi Capriotti 17 Jun '02

17 Jun '02

I'm trying to run a PB-generated application under Wine, but every panel with date-time fields generate a crash. By means of winedbg I've managed to track the crashes to a limited set of function calls into the runtime engine PBVM60.DLL (namely PBE_SetData, PBE_FmtSetDataString, PBE_FmtSetMask ) which are called by the engine itself whenever creating dialogs. It seems to me a dialog property is not correctly initialised, leading to a segfault accessing an invalid pointer. I've managed to package a minimal test application into http://web.tiscali.it/l_capriotti/WineTest.zip hoping some Wine expert can provide advices-directions, Thanks in advance for any help. Luigi

2 1

Re: Return and absolut path in DRIVE_FindDriveRoot
by Michael Wetherell 17 Jun '02

17 Jun '02

On Friday 14 Jun 2002 6:20 pm, Uwe Bonnes wrote: > some application (xilinx webpack) dind't find it's initialisation file when > started in the debugger ("winedbg bin/nt/webpack.exe" versus "wine > bin/nt/exepack" started in "l:/xilinx"). As trap there is also a directory > "l:/xilinx/xilinx". The starting process (wine or winedbg) set the current > working directory to l: and when winedbg started the user process, > DRIVE_Chdir in DIR_Init set the current directory to "l:/xilinx/xilinx". > This was caused by DRIVE_FindDriveRoot not returning an absolute path. I'm sorry about that, I should have forseen it. Your patch does of course fix it (and that's how it was before) but I think it might be better to make the change to DIR_Init() instead though. I think the trouble with trying to return the leading slash in the result of DRIVE_FindDriveRoot() is that whether you get an absolute or relative path depends on whether we are at a drive's root or not, rather than whether the input unix path is absolute or not. So since DRIVE_FindDriveRoot() can't always do it correctly I think it would probably be better if it didn't do it at all, that way the caller knows they have to deal with it rather than it working some of the time, but other times giving rise subtle problems.

2 2