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

Re: FILE_GetNtStatus Converting errno 38 to STATUS_UNSUCCESSFUL

by Alexandre Julliard

Mike Hearn <mike(a)navi.cx> writes: > It just means there was some information loss in converting error codes > internally inside Wine ... you'll have to look elsewhere to figure out > *why* it failed. There is a table of errno codes on MSDN you can use or > look in winerror.h to find out what 38 is. Actually it's a Unix errno, 38 is ENOSYS. It would be interesting to find out what function causes that, a relay trace is probably the best way. -- Alexandre Julliard julliard(a)winehq.org

June 25, 2004

Re: problem with linkage of so under RHEL3

by Alexandre Julliard

Bill Medland <billmedland(a)mercuryspeed.com> writes: > So for future reference is there an easy way to see what functions a so > provides that don't show up in nm? Try nm -D. > Then I guess the question will be a matter of why the dlopen doesn't see the > existing functions when underneath Wine. Just a wild guess, but you could try adding RTLD_GLOBAL to the dlopen call where Wine opens libodbc.so. -- Alexandre Julliard julliard(a)winehq.org

June 25, 2004

Re: FW: FW: WM_NEXTDLGCTL changes the default button ID and does not rest ore default control identifier

by Alexandre Julliard

Krishna Murthy <Krishna.Murthy(a)guptaworldwide.com> writes: > Alexandre, > Is these changes looks good? Do I have to submit this patch to > 'wine-patches(a)winehq.org' ? Well, I'm not really happy with it, that whole bSetDefID thing is fairly ugly IMO. I'd really appreciate if you could write a regression test that illustrates the behavior of both DM_SETDEFID and WM_NEXTDLGCTL so that the code can be restructured in a more readable way. -- Alexandre Julliard julliard(a)winehq.org

June 25, 2004

Re: FILE_GetNtStatus Converting errno 38 to STATUS_UNSUCCESSFUL

by Mike Hearn

On Fri, 25 Jun 2004 18:16:14 +0200, Jürgen Leibner wrote: > # fixme:ntdll:FILE_GetNtStatus Converting errno 38 to STATUS_UNSUCCESSFUL It just means there was some information loss in converting error codes internally inside Wine ... you'll have to look elsewhere to figure out *why* it failed. There is a table of errno codes on MSDN you can use or look in winerror.h to find out what 38 is. thanks -mike

June 25, 2004

Re: problem with linkage of so under RHEL3

by Bill Medland

On June 24, 2004 05:46 pm, Bill Medland wrote: > Guys > > Command line application opens a dll using LoadLibrary. > That Dll links to odbc > odbc uses unixODBC to link to the new Oracle 10g Beta ODBC client. > That client so access rather a lot of Oracle libraries > > Relocation error occurs > > /usr/local/bin/wine-pthread: relocation error: > /opt/oracle/product/10.1.0/client_1/lib/libsqora.so: undefined symbol: > SQLGetPrivateProfileString Can anyone help me on this please. I am basically constrained by my lack of deep knowledge about Linux and would appreciate guidance. I am quite prepared to do my best to help fix the problem with Wine but cannot do it alone. First, can someone explain why the Linux equivalent does work. unittest links to mylib.so using dlopen (..., RTLD_NOW) mylib.so is linked to libodbc.so At the request of mylib via a SQLDriverConnect libodbc.so opens a link to $ORACLE_HOME/lib/libsqora.so. I don't know whether it is RTLD_NOW or RTLD_LAZY because it's down in libtools somewhere Now here is where I get confused. ldd -r libsqora.so shows 7 unresolved functions, including SQLGetPrivateProfileString It also does not show a dependency on libodbc itself. SQLGetPrivateProfileString does exist in libodbc. nm libodbc.so does not show it. Presumably the link works because since it was the libodbc that linked to libsqora.so SQLGetPrivateProfileString was already defined when the dlopen occurred. So for future reference is there an easy way to see what functions a so provides that don't show up in nm? Then I guess the question will be a matter of why the dlopen doesn't see the existing functions when underneath Wine. -- Bill Medland mailto:billmedland(a)mercuryspeed.com http://webhome.idirect.com/~kbmed

June 25, 2004

Scrollbars, an application bug or a wine regression ? (#2314)

by Gabriele Giorgetti

Hello, I've noticed a problem with scrollbars in the latest Wine's release (20040615). With wine version 20040505 everything was just fine. I've attacched a zip file with two screenshots (pls. see wine BUG #2314 @ bugs.winehq.org ) the first one is the correct behavior of the app using wine 20040505 the second one is wine 20040615 and the app in its wrong behavior. Trying to describe the problem: The application show up with a bottom scrollbar and a right scrollbar even if they are not needed. When executing the very same application, with the very same settings and size with wine 20040505 the scrollbars are not there. I hope you get it Thanks. NB: BUG #2314 at bugs.winehq.org

June 25, 2004

FILE_GetNtStatus Converting errno 38 to STATUS_UNSUCCESSFUL

by Jürgen Leibner

Hi there, i'm trying to install a Software for Schools using wine-20040505-SuSELinux90.i586.rpm. After launching wine as user with: # wine setup.exe i get this message: .... # fixme:ntdll:FILE_GetNtStatus Converting errno 38 to STATUS_UNSUCCESSFUL Is anybody out there, who is able to explain what happens and what to do to get succsess? Thanks, Jürgen

June 25, 2004

Re: appdb dump

by Jeremy Newman

This is getting to be a common request. I think what I am going to do is make a script that tars up the appdb every night and puts it up for download. The only table it will not dump will be the userdb, for privacy reasons. I'll email you all again when I have something whipped up. On Thu, 2004-06-24 at 18:12, Joel Konkle-Parker wrote: > I'm trying to set up a copy of appdb on localhost to play with (and > hopefully use as a base for patches for you guys). Do you think I could > get a dump of the apps tables so that I can have a good set to test with? -- _ _ WebGeek/SysAdmin CodeWeavers -= http://www.codeweavers.com | \| |_____ __ ___ __ __ _ _ _ -= http://www.dracowulf.com | .` / -_) | | / ' \/ _` | ' \ -= mailto:jnewman(a)codeweavers.com |_|\_\___|\_/\_/|_|_|_\__,_|_||_| -= ICQ: 1711553 Yahoo: laxdragon

June 25, 2004

does wineshelllink work for use with GNOME?

by Santosh Siddheshwar

Hi, Does anybody know of the issues when making shortcuts on the main menu on GNOME (redhat 9)? It seems to me that the .desktop files are created under the right folder ($HOME/.gnome/apps) but the shortcuts dont appear on the menu. As far as I could figure the way to go about adding to the main or system menu is to make entries in the Applications.directory file in /etc/X11/desktop-menus and then have your app .desktop files in /usr/share/applications. Why dont the .desktop files under $HOME/.gnome/apps show up on the menu? Any other ideas? Similarly on redhat 9 for KDE the shortcuts seem to be required to be under $HOME/.kde/applnk-redhat but wineshelllink puts them in applnk folder. Is this correct? Thanks in advance! Santosh Siddheshwar ********************************************************************* Disclaimer: The information in this e-mail and any attachments is confidential / privileged. It is intended solely for the addressee or addressees. If you are not the addressee indicated in this message, you may not copy or deliver this message to anyone. In such case, you should destroy this message and kindly notify the sender by reply email. Please advise immediately if you or your employer does not consent to Internet email for messages of this kind. *********************************************************************

June 25, 2004

problem with linkage of so under RHEL3

by Bill Medland

Guys (I'll be looking into it tomorrow but I thought I'd see if anyone already knows) Command line application opens a dll using LoadLibrary. That Dll links to odbc odbc uses unixODBC to link to the new Oracle 10g Beta ODBC client. That client so access rather a lot of Oracle libraries Relocation error occurs /usr/local/bin/wine-pthread: relocation error: /opt/oracle/product/10.1.0/client_1/lib/libsqora.so: undefined symbol: SQLGetPrivateProfileString The application can also be built under Linux, as can "That Dll" and link directly to unixODBC. Under that configuration it works fine. In all the messing about with the threading etc can anyone think immediately of where we might have messed up the linking? (Wine is rather old - cvs compile of about 20040408) -- Bill Medland mailto:billmedland(a)mercuryspeed.com http://webhome.idirect.com/~kbmed

June 25, 2004