LyX is a document processor that encourages an approach to writing based on the structure of your documents and not simply their appearance (). LyX combines the power and flexibility of TeX/LaTeX with the ease of use of a graphical interface. This results in world-class support for creation of mathematical content (via a fully integrated equation editor) and structured documents like academic. LyX is a modern approach to writing documents which breaks with the obsolete 'typewriter paradigm' of most other document preparation systems. Nov 23, 2008 - Fix a crash on the Mac when using the pixmap cache with combining characters. The tex2lyx man page reflects the current state (bug 2770).
You can subscribe to this list. Oops, the link to the earlier email was messed up: I bet you found it already, anyway. On Jul 31, 2006, at 3:39 PM, Jens Noeckel wrote: On Jul 31, 2006, at 1:08 PM, Jens Noeckel wrote: On Jul 31, 2006, at 12:42 PM, Martin Costabel wrote: Jens Noeckel wrote: Jeff - I do still see the crash on LyX-Mac 1.4.2, though. It will be interesting to see if it is still there in the fink package, too. I have now a lyx-qtmac package in my exp directory that uses the Fink qt3mac package and builds a LyX.app in /sw/Applications. I have not managed to get this crash there.
Comments on this package would be welcome. Martinlyx-qtmac from your experimental directory works exactly as expected. That means, everything works fine, except for the following two things, which I've reported to lyx-devel:.
After loading a file and starting to work on it, the LyX menu rearranges itself so that menu items don't correspond to the correct LyX functions. I've been told that this probably won't be addressed until the 1.4.3 release. When cutting a math or table inset and doing a save-and-quit immediately afterwards, the program crashes. Fortunately, this never causes any loss of data because the save is done before the crash occurs. The latter is the bug I mentioned on this list before. I'm surprised that you can't reproduce it.
Did you see my follow-up clarification of how to get the bug? Jeff did manage to reproduce this with lyx-qt after I corrected my initial sloppy wording. ThanksJens. On Jul 31, 2006, at 1:08 PM, Jens Noeckel wrote: On Jul 31, 2006, at 12:42 PM, Martin Costabel wrote: Jens Noeckel wrote: Jeff - I do still see the crash on LyX-Mac 1.4.2, though. It will be interesting to see if it is still there in the fink package, too. I have now a lyx-qtmac package in my exp directory that uses the Fink qt3mac package and builds a LyX.app in /sw/Applications. I have not managed to get this crash there.
Comments on this package would be welcome. Martin, lyx-qtmac from your experimental directory works exactly as expected.
That means, everything works fine, except for the following two things, which I've reported to lyx-devel:. After loading a file and starting to work on it, the LyX menu rearranges itself so that menu items don't correspond to the correct LyX functions. Feature 216 please support ui with retina display for mac. I've been told that this probably won't be addressed until the 1.4.3 release. When cutting a math or table inset and doing a save-and-quit immediately afterwards, the program crashes. Fortunately, this never causes any loss of data because the save is done before the crash occurs. The latter is the bug I mentioned on this list before. I'm surprised that you can't reproduce it.
Did you see my follow-up clarification of how to get the bug? Jeff did manage to reproduce this with lyx-qt after I corrected my initial sloppy wording. Thanks, Jens. On Jul 31, 2006, at 12:42 PM, Martin Costabel wrote: Jens Noeckel wrote: Jeff - I do still see the crash on LyX-Mac 1.4.2, though.
It will be interesting to see if it is still there in the fink package, too. I have now a lyx-qtmac package in my exp directory that uses the Fink qt3mac package and builds a LyX.app in /sw/Applications. I have not managed to get this crash there. Comments on this package would be welcome. By the way, while compiling lyx 1.4.2 myself, I discovered (or Google discovered for me) that there is a way to patch QT-mac so that the Ctrl key regains its emacs-like functionality, and the Command key becomes a meta key. I tried it and it works with the Mac versions of qt3 and lyx 1.4.2. I wonder if one can also do something like that with the X11 qt3 that fink's lyx uses.
That way lyx-qt would finally have a meta key, the way it did with xforms. If there's interest in this, I could send more information. I am not sure if I understand what this patch is doing and whether this would be interesting for the qt3mac package. So please send some more information. Martin, that's great news - I'll give it a try as soon as possible. The patch I mentioned works for qt-mac, but I looked at the qt-x11 source and saw that it's organized quite differently so that I didn't see an easy way to go ahead with patching the x11 version. Maybe this will be easier with qt4 (I just don't do GUI programming, so it's perfectly possible that it's all very easy and I just don't see it).
That's why I haven't gotten back with more info. Anyway, since you now have qt-mac, here is the link to a patch that works perfectly for the mac version of qt-1.3.6. It's a post by Jason Woodard on the lyx-users mailing list, and all you need is the small patch listed at the bottom. I've tested it and it's a great improvement for emacs users (if you use something like xemacs.bind, as I do). Without the patch, lyx with qt-mac produces the Ctrl-key behavior when you press Command, and you get a meta key by pressing Option. If you compare that with how fink's xemacs works, it's confusing. Jens Noeckel wrote: Jeff - I do still see the crash on LyX-Mac 1.4.2, though.
It will be interesting to see if it is still there in the fink package, too. I have now a lyx-qtmac package in my exp directory that uses the Fink qt3mac package and builds a LyX.app in /sw/Applications.
I have not managed to get this crash there. Comments on this package would be welcome. By the way, while compiling lyx 1.4.2 myself, I discovered (or Google discovered for me) that there is a way to patch QT-mac so that the Ctrl key regains its emacs-like functionality, and the Command key becomes a meta key. I tried it and it works with the Mac versions of qt3 and lyx 1.4.2.
I wonder if one can also do something like that with the X11 qt3 that fink's lyx uses. That way lyx-qt would finally have a meta key, the way it did with xforms. If there's interest in this, I could send more information. I am not sure if I understand what this patch is doing and whether this would be interesting for the qt3mac package. So please send some more information. Philip Ershler wrote: Thanks for the suggestion. I changed over to the 10.4 tree, did a selfupdate and then tried to build soqt21 again.
But the versions of the soqt21 and qt3 packages you show: Failed: phase compiling: soqt21-1.2.0-3 failed i qt3 3.3.6-27 Cross-Platform GUI application framework are those of the 10.4-transitional tree. Something must have gone wrong with your switch to 10.4.
Did you use the 10.4-update script from the Fink homepage or did you do the switch by hand? In any case, your 'Distribution:' in /sw/etc/fink.conf must now be 10.4, and /sw/fink/dists must be a symbolic link to 10.4. This doesn't seem to be the case on your system. On Jul 28, 2006, at 1:25 AM, Martin Costabel wrote: Philip Ershler wrote: Failed: phase compiling: soqt21-1.2.0-3 failed Any reason why you are still using the 10.4-transitional tree? What you are seeing is probably some effect of it rotting in its grave.
On the Fink home page it says that that tree will be officially unsupported from August 1st, but in reality it has not been completely supported for some time already. In particular, there was a patch related to this problem for the soqt21 package in the 10.4 tree (now at version 1.2.0-1004), but not in the 10.4-transitional tree. - Martin Thanks for the suggestion. I changed over to the 10.4 tree, did a selfupdate and then tried to build soqt21 again.
Things have changed a bit. But now my head is spinning round and round. Below I have included a portion of the install output, the value of QTDIR, a look at which parts of qt3 are installed and the top of the config.log.
Is there any more info I can provide to solve this issue? Thanks once again, Phil fink install soqt21 checking if the Open Inventor SoPolygonOffset node is available. True checking if the Open Inventor SoVRMLMaterial VRML node is available. True checking if the Open Inventor SoVRMLViewpoint VRML node is available. True checking if the Open Inventor SoVRMLBackground VRML node is available.
True checking if the Open Inventor SoVRMLFog VRML node is available. True checking for SoKeyboardEvent::DELETE enum. True checking for SoMouseButtonEvent::BUTTON5 enum. True checking for SoCamera::setStereoMode method. True checking for SoGLRenderAction::SORTEDLAYERSBLEND transparency mode.
True checking whether the X11 miscellaneous utilities library is available. Yes checking for moc. /sw/bin/moc checking version of Qt library. 336 checking for Qt library devkit. UNRESOLVED checking install sanity.
Ok Could not find any way to build against the Qt library. If you are sure Qt is properly installed on your system, see the file config.log to find out what went wrong.
One common way to solve this problem is to set up the environment variable QTDIR to the base of the Qt installation directory. Or if it was already set up, double check to see that it is pointing at the correct directory. Note that it is possible to override the combinations of Qt libraries configure tries to link against by setting up the environment variable CONFIGQTLIBS. This might be necessary if you have installed the Qt library in a manner that is not compatible with any of the known configurations, for instance by changing the name of the library or the Qt library having more dependencies on other libraries that we are aware of. If you for instance know the correct set of libraries to be qt304.lib, qtmain.lib and gdi32.lib (this would be on an MSWindows system), then set CONFIGQTLIBS to the string '-lqt304 -lqtmain -lgdi32' before re-running configure.
If you still can't get the configure script to detect the presence of and how to use the Qt library, please mail the config.log file to and ask for help. Configure: error: aborting ### execution of./configure failed, exit code 1 Removing build lock. /sw/bin/dpkg-lockwait -r fink-buildlock-soqt21-1.2.0-3 (Reading database. 106847 files and directories currently installed.) Removing fink-buildlock-soqt21-1.2.0-3. I have a new install on OS X 10.4 of Fink, and tried to install Etherape. I am getting stuck at a point where the error:. /sw/bin/apt-get-lockwait -q -ignore-breakage -download-only install tetex-texmf=3.0-1 Reading Package Lists.
Building Dependency Tree. The following NEW packages will be installed: tetex-texmf 0 packages upgraded, 1 newly installed, 0 to remove and 0 not upgraded. Need to get 148MB of archives. After unpacking 364MB will be used. Get:1 10.4/release/main tetex-texmf 3.0-1 148MB Err 10.4/release/main tetex-texmf 3.0-1 Error reading from server Remote end closed connection Failed to fetch fink-0.8.1/main/binary-darwin-i386//text/tetex-texmf3.0-1darwin- i386.deb Error reading from server Remote end closed connection E: Some files failed to download ### execution of /sw/bin/apt-get-lockwait failed, exit code 100 Failed: Downloading the binary package 'tetex-texmf3.0-1darwin- i386.deb' failed.
appears. This is only the third program I've installed under Fink, the first packages being Ethereal and ClamAV. Any suggestions what would be causing this error?
I've done the selfupdate twice. Fink Commander states 5100 packages displayed, 162 installed. This is on an Intel Macbook. Fink -V gives Package manager version: 0.24.25 Distribution version: 0.8.1.rsync i386 Any ideas, pointers, etc.? Thanks, -Bart Silverstrim. Markus Kamp wrote: Hello folksI use fink 0.8.1 on my Tiger powered ibook G4 and have recently switched from Apples X11 to fink's Xorg package. So far, everything works as planned except some minor cosmetic issues in rootless mode (Graphics behind KDE-Menus won't redraw properly in some circumstances).
The only 'real' problem I have now is, that some X11 based programs (okay, games:-) ) which were not installed via fink, com as self contained application bundles, so you're able to double-click 'em in an Aqua environment and X11 should start up. The bundles I've tried so far, FreeCiv and Scorched3D seem to rely exclusively on Apple's X11, at least for FreeCIV I get an error- message telling me this. What can I do to convince these bundles to work with XDarwin.app?
Well, these are open source, aren't they? So, simply take the sources, replace 'X11.app' by 'XDarwin.app' at the appropriate place, and recompile. Well, at least this would be the recipe if they were.really. open source in the sense of the GPL.
Unfortunately, many of these app bundles that are distributed for MacOSX are in more or less open violation of the GPL, so you can't do what I said. Sometimes you are lucky and the 'X11.app' is only a text string in a wrapper script that you can change in the distributed product. This is often the case in those app bundles that are basically rip-offs of Fink packages, coated with some AppleScript and shell script wrappers and stuck into an app frame. But take the case of freeciv which is quite typical (I haven't looked at Scorched3D in detail; does it really require X11.app? It looks like some big self-contained binary, no app bundle): They put their stuff under the GPL, but the MacOSX binary they distribute on the dmg violates the GPL in several respects: - They don't give you the sources for the main Mach-o binary 'Freeciv Client' that is started when you open the app and which in turn starts X11.app. So you are screwed with your request.
They say that the binaries are 'Based on 2.0.8 source code', whatever that means. The GPL would require that they not only give you the exact sources for the distributed binaries, in particular for the 'Freeciv Client' binary, but also the exact scripts and config files they used to compile it. They stick several dozen dylibs into the app bundle, most of them probably distributed under the GPL. Again, under the GPL they are obliged to distribute the sources for the exact versions of these libraries and of the scripts and config files they used to compile them. This is one of my favorite rants with which I have already annoyed several people, among them the author of Gimp.app (successfully; he published the scripts to compile the app) and even once the quasi-maintainer of our very own FinkCommander.app (not completely successfully; you can't build the distributed version from the published sources, AFAICT). I'm stuck:-) Any more hints or ideas what to try next?
Go to the freeciv mailing list and complain. Or use Fink's freeciv. If it isn't up to date with the latest version, nudge the maintainer:-) - Martin. Forwarded message - From: 'Steven Boothe' To: fink-users@.
Date: Sat, 29 Jul 2006 21:53:31 -0700 Subject: apt-get update broken??? Hello all: I've been getting this error all day and wonder if anyone else has noticed a similar problem lately? Hello folks, I use fink 0.8.1 on my Tiger powered ibook G4 and have recently switched from Apples X11 to fink's Xorg package.
So far, everything works as planned except some minor cosmetic issues in rootless mode (Graphics behind KDE-Menus won't redraw properly in some circumstances). The only 'real' problem I have now is, that some X11 based programs (okay, games:-) ) which were not installed via fink, com as self contained application bundles, so you're able to double-click 'em in an Aqua environment and X11 should start up. The bundles I've tried so far, FreeCiv and Scorched3D seem to rely exclusively on Apple's X11, at least for FreeCIV I get an error- message telling me this. What can I do to convince these bundles to work with XDarwin.app? What I've tried so far is: From Finder/Aqua: - started XDarwin/KDE manually and then tried to double-click the bundle - placed an Alias titled 'X11' in /Applications/Utilities/ pointing to /Applications/XDarwin.app - Opened the bundle (right-click and choose (de): 'Paketinhalt zeigen') and tried to execute the Unix-executable therein.
From XDarwin/KDE: - Opened the bundle (appears as a Folder) and tried to execute the Unix-executable. Here it isn't recognised as an executable but as a document and KDE asks me which Program I want to open it with:-/ So far I didn't have the heart to try chmod+x inside the bundle although it might be the right way to go I'm stuck:-) Any more hints or ideas what to try next? Thanks, Markus - Markus Kamp Wuppertal, Germany. On Jul 29, 2006, at 5:10 AM, Jeff Whitaker wrote: Jens Noeckel wrote: On Jul 28, 2006, at 10:39 AM, Jeff Whitaker wrote: Jens Noeckel wrote: HiI have lyx-qt 1.4.1-1 installed from unstable with fink 0.8.1.rsync i386 on a MacBook Pro, and get a crash when I do a particular sequence of steps in lyx. I've reported this to the lyx developers but also wanted to report it here, in case anyone else experiences this (i've now checked this on three different machines, PPC and Intel, and also on the Non-X11 Qt LyX-Mac version 1.4.2): Try opening a new file, type a few words with a math inset among them.
Save the file. Then erase the math inset and save again.
Immediately afterwards, exit lyx. That causes a crash.
The error message on the terminal is: Mutex destroy failure: Resource busy lyx: SIGSEGV signal caught Sorry, you have found a bug in LyX. Please read the bug-reporting instructions in Help-Introduction and send us a bug report, if necessary. /sw/bin/lyx: line 15: 4178 Abort trap /sw/bin/lyxx $@ I've also checked this on a PPC machine - there, the program also aborts abnormally but with fewer messages after the 'Mutex destroy failure: Resource busy'. Jens Jens; From the 1.4.2 release notes:. Fix a crash on exit on MacOS X, and the well known 'Mutex destroy failure' error message on Linux (qt frontend, bugs 2549 and 1029). I'm testing a package for 1.4.2 now, will let you know when it is committed.
-Jeff Jeff - I do still see the crash on LyX-Mac 1.4.2, though. It will be interesting to see if it is still there in the fink packagetoo. By the way, while compiling lyx 1.4.2 myself, I discovered (or Google discovered for me) that there is a way to patch QT-mac so that the Ctrl key regains its emacs-like functionality, and the Command key becomes a meta key. I tried it and it works with the Mac versions of qt3 and lyx 1.4.2. I wonder if one can also do something like that with the X11 qt3 that fink's lyx uses. That way lyx-qt would finally have a meta key, the way it did with xforms. If there's interest in this, I could send more information.
ThanksJens Jens: I've updated lyx-qt in 10.4 unstable to 1.4.2, and I can't reproduce your crash with it. Jeff, the new lyx-qt 1.4.2 in unstable is still showing the crash on my MacBook Pro.
I re-read my initial email and noticed that I was inaccurate in describing how to reproduce the crash. When I said 'erase the math inset and save again,' I meant you have to cut the math inset by using the 'cut' function from the Edit menu (i.e., it goes into the 'clipboard') - not just erase by using backspace or delete-forward. When I go through the editing steps described above, here is the crash: lyx: SIGSEGV signal caught Sorry, you have found a bug in LyX. Please read the bug-reporting instructions in Help-Introduction and send us a bug report, if necessary.
Jens Noeckel wrote: On Jul 28, 2006, at 10:39 AM, Jeff Whitaker wrote: Jens Noeckel wrote: HiI have lyx-qt 1.4.1-1 installed from unstable with fink 0.8.1.rsync i386 on a MacBook Pro, and get a crash when I do a particular sequence of steps in lyx. I've reported this to the lyx developers but also wanted to report it here, in case anyone else experiences this (i've now checked this on three different machines, PPC and Intel, and also on the Non-X11 Qt LyX-Mac version 1.4.2): Try opening a new file, type a few words with a math inset among them. Save the file.
Then erase the math inset and save again. Immediately afterwards, exit lyx. That causes a crash. The error message on the terminal is: Mutex destroy failure: Resource busy lyx: SIGSEGV signal caught Sorry, you have found a bug in LyX. Please read the bug-reporting instructions in Help-Introduction and send us a bug report, if necessary.
/sw/bin/lyx: line 15: 4178 Abort trap /sw/bin/lyxx $@ I've also checked this on a PPC machine - there, the program also aborts abnormally but with fewer messages after the 'Mutex destroy failure: Resource busy'. Jens Jens; From the 1.4.2 release notes:. Fix a crash on exit on MacOS X, and the well known 'Mutex destroy failure' error message on Linux (qt frontend, bugs 2549 and 1029). I'm testing a package for 1.4.2 now, will let you know when it is committed. -Jeff Jeff - I do still see the crash on LyX-Mac 1.4.2, though.
It will be interesting to see if it is still there in the fink package, too. By the way, while compiling lyx 1.4.2 myself, I discovered (or Google discovered for me) that there is a way to patch QT-mac so that the Ctrl key regains its emacs-like functionality, and the Command key becomes a meta key. I tried it and it works with the Mac versions of qt3 and lyx 1.4.2.
I wonder if one can also do something like that with the X11 qt3 that fink's lyx uses. That way lyx-qt would finally have a meta key, the way it did with xforms. If there's interest in this, I could send more information. ThanksJens Jens: I've updated lyx-qt in 10.4 unstable to 1.4.2, and I can't reproduce your crash with it. As for the qt patch, Benjamin Reed is the qt3 package maintainer so I'd send the patch to him.Jeff - Jeffrey S.
Whitaker Phone: (303)497-6313 NOAA/OAR/CDC R/PSD1 FAX: (303)497-6449 325 Broadway Boulder, CO, USA. On 7/29/06, Andrew Hartung wrote: On Jul 29, 2006, at 12:23 AM, Chris Zubrzycki wrote: On Jul 28, 2006, at 8:19 PM, Alexander K.
Tex 2 Lyx For Mac
Hansen wrote: You should be able to do a recursive remove via gnome-base. What's failing in that case? Also, sudo apt-get remove gnome-base should work too. That gives me Reading Package Lists.
Done Building Dependency Tree. Done You might want to run `apt-get -f install' to correct these: Sorry, but the following packages have unmet dependencies: gnome-mime-data: Depends: gnome-base but it is not going to be installed gnome-print-shlibs: Depends: gnome-base but it is not going to be installed gtkhtml-shlibs: Depends: bonobo (= 1.0.20-13) but it is not going to be installed guppi16-dev: Depends: guppi16-shlibs (= 0.40.3-1015) but 0.40.3-1016 is to be installed libmikmod: Depends: libmikmod-shlibs (= 3.1.11-2) but 3.1.11-3 is to be installed E: Unmet dependencies.
Try 'apt-get -f install' with no packages (or specify a solution). 'fink remove -recursive' uses 'apt-get remove' to do its thing, so the outputs indeed ought to be the same.
What's going on (I think) is that apt-get and fink don't use identical dependency tracking, so it's possible for the system to get in a state that's OK for fink operations on source tarballs, but not for apt-get; and the latter won't do anything for you untill you resolve its gripes. Alexander K. Hansen Fink Documenter (still). On Jul 28, 2006, at 7:19 PM, Alexander K. Hansen wrote: On 7/28/06, Andrew Hartung wrote: I searched the archives and failed to find a solution that worked. When I try to remove gnme I get: $ fkr gnome-base Information about 5937 packages read in 13 seconds. /sw/bin/dpkg-lockwait -remove gnome-base dpkg: dependency problems prevent removal of gnome-base: gnome-print-shlibs depends on gnome-base.
gnome-mime-data depends on gnome-base. /sw/bin/dpkg: error processing gnome-base (-remove): dependency problems - not removing Errors were encountered while processing: gnome-base ### execution of /sw/bin/dpkg-lockwait failed, exit code 1 ERROR: Can't remove package(s). If the above error message mentions dependency problems, you can try fink remove -recursive gnome-base This will attempt to remove the package(s) specified as well as ALL packages that depend on it.
Failed: can't remove package gnome-base -recursive doesn't work either. I'm really tired of chasing and removing dependencies manually, is there a way to get rid of all of gnome? You should be able to do a recursive remove via gnome-base. What's failing in that case?
I get: Information about 5939 packages read in 5 seconds. /sw/bin/apt-get-lockwait remove gnome-base Reading Package Lists. Done Building Dependency Tree. Done You might want to run `apt-get -f install' to correct these: Sorry, but the following packages have unmet dependencies: gnome-mime-data: Depends: gnome-base but it is not going to be installed gnome-print-shlibs: Depends: gnome-base but it is not going to be installed gtkhtml-shlibs: Depends: bonobo (= 1.0.20-13) but it is not going to be installed guppi16-dev: Depends: guppi16-shlibs (= 0.40.3-1015) but 0.40.3-1016 is to be installed libmikmod: Depends: libmikmod-shlibs (= 3.1.11-2) but 3.1.11-3 is to be installed E: Unmet dependencies. Try 'apt-get -f install' with no packages (or specify a solution).
### execution of /sw/bin/apt-get-lockwait failed, exit code 100 Failed: can't remove package gnome-base. On Jul 29, 2006, at 12:23 AM, Chris Zubrzycki wrote: On Jul 28, 2006, at 8:19 PM, Alexander K. Hansen wrote: You should be able to do a recursive remove via gnome-base.
What's failing in that case? Also, sudo apt-get remove gnome-base should work too.
That gives me Reading Package Lists. Done Building Dependency Tree.
Done You might want to run `apt-get -f install' to correct these: Sorry, but the following packages have unmet dependencies: gnome-mime-data: Depends: gnome-base but it is not going to be installed gnome-print-shlibs: Depends: gnome-base but it is not going to be installed gtkhtml-shlibs: Depends: bonobo (= 1.0.20-13) but it is not going to be installed guppi16-dev: Depends: guppi16-shlibs (= 0.40.3-1015) but 0.40.3-1016 is to be installed libmikmod: Depends: libmikmod-shlibs (= 3.1.11-2) but 3.1.11-3 is to be installed E: Unmet dependencies. Try 'apt-get -f install' with no packages (or specify a solution). BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Jul 28, 2006, at 8:19 PM, Alexander K.
Hansen wrote: You should be able to do a recursive remove via gnome-base. What's failing in that case? Also, sudo apt-get remove gnome-base should work too.
-chris zubrzycki - PGP public key: ID: 0xA2ABC070 Fingerprint: 26B0 BA6B A409 FA83 42B3 1688 FBF9 8232 A2AB C070 I think you've been smoking Cocoa instead of coding with it.BEGIN PGP SIGNATURE- Version: GnuPG v1.4.4 (Darwin) Comment: Please sign reply- iEYEARECAAYFAkTK8MsACgkQ+/mCMqKrwHAzNwCgq6cE4YzZKNfO64dm+bh270xL DI4AnRYSCkTToGmhCz8Yflr4gkcPKEHY =Xbdd -END PGP SIGNATURE. On 7/28/06, Andrew Hartung wrote: I searched the archives and failed to find a solution that worked. When I try to remove gnme I get: $ fkr gnome-base Information about 5937 packages read in 13 seconds. /sw/bin/dpkg-lockwait -remove gnome-base dpkg: dependency problems prevent removal of gnome-base: gnome-print-shlibs depends on gnome-base. gnome-mime-data depends on gnome-base.
/sw/bin/dpkg: error processing gnome-base (-remove): dependency problems - not removing Errors were encountered while processing: gnome-base ### execution of /sw/bin/dpkg-lockwait failed, exit code 1 ERROR: Can't remove package(s). If the above error message mentions dependency problems, you can try fink remove -recursive gnome-base This will attempt to remove the package(s) specified as well as ALL packages that depend on it. Failed: can't remove package gnome-base -recursive doesn't work either.
I'm really tired of chasing and removing dependencies manually, is there a way to get rid of all of gnome? You should be able to do a recursive remove via gnome-base. What's failing in that case? - Alexander K. Hansen Fink Documenter (still). I searched the archives and failed to find a solution that worked.
When I try to remove gnme I get: $ fkr gnome-base Information about 5937 packages read in 13 seconds. /sw/bin/dpkg-lockwait -remove gnome-base dpkg: dependency problems prevent removal of gnome-base: gnome-print-shlibs depends on gnome-base.
Gnome-mime-data depends on gnome-base. /sw/bin/dpkg: error processing gnome-base (-remove): dependency problems - not removing Errors were encountered while processing: gnome-base ### execution of /sw/bin/dpkg-lockwait failed, exit code 1 ERROR: Can't remove package(s).
If the above error message mentions dependency problems, you can try fink remove -recursive gnome-base This will attempt to remove the package(s) specified as well as ALL packages that depend on it. Failed: can't remove package gnome-base -recursive doesn't work either.
I'm really tired of chasing and removing dependencies manually, is there a way to get rid of all of gnome? On Jul 28, 2006, at 10:39 AM, Jeff Whitaker wrote: Jens Noeckel wrote: HiI have lyx-qt 1.4.1-1 installed from unstable with fink 0.8.1.rsync i386 on a MacBook Pro, and get a crash when I do a particular sequence of steps in lyx. I've reported this to the lyx developers but also wanted to report it here, in case anyone else experiences this (i've now checked this on three different machines, PPC and Intel, and also on the Non-X11 Qt LyX-Mac version 1.4.2): Try opening a new file, type a few words with a math inset among them. Save the file. Then erase the math inset and save again.
Immediately afterwards, exit lyx. That causes a crash. The error message on the terminal is: Mutex destroy failure: Resource busy lyx: SIGSEGV signal caught Sorry, you have found a bug in LyX. Please read the bug-reporting instructions in Help-Introduction and send us a bug report, if necessary. /sw/bin/lyx: line 15: 4178 Abort trap /sw/bin/lyxx $@ I've also checked this on a PPC machine - there, the program also aborts abnormally but with fewer messages after the 'Mutex destroy failure: Resource busy'. Jens Jens; From the 1.4.2 release notes:. Fix a crash on exit on MacOS X, and the well known 'Mutex destroy failure' error message on Linux (qt frontend, bugs 2549 and 1029).
I'm testing a package for 1.4.2 now, will let you know when it is committed. -Jeff Jeff - I do still see the crash on LyX-Mac 1.4.2, though. It will be interesting to see if it is still there in the fink package, too. By the way, while compiling lyx 1.4.2 myself, I discovered (or Google discovered for me) that there is a way to patch QT-mac so that the Ctrl key regains its emacs-like functionality, and the Command key becomes a meta key. I tried it and it works with the Mac versions of qt3 and lyx 1.4.2. I wonder if one can also do something like that with the X11 qt3 that fink's lyx uses. That way lyx-qt would finally have a meta key, the way it did with xforms.
If there's interest in this, I could send more information. Thanks, Jens. Jens Noeckel wrote: HiI have lyx-qt 1.4.1-1 installed from unstable with fink 0.8.1.rsync i386 on a MacBook Pro, and get a crash when I do a particular sequence of steps in lyx. I've reported this to the lyx developers but also wanted to report it here, in case anyone else experiences this (i've now checked this on three different machines, PPC and Intel, and also on the Non-X11 Qt LyX-Mac version 1.4.2): Try opening a new file, type a few words with a math inset among them.
Save the file. Then erase the math inset and save again.
Immediately afterwards, exit lyx. That causes a crash. The error message on the terminal is: Mutex destroy failure: Resource busy lyx: SIGSEGV signal caught Sorry, you have found a bug in LyX. Please read the bug-reporting instructions in Help-Introduction and send us a bug report, if necessary. /sw/bin/lyx: line 15: 4178 Abort trap /sw/bin/lyxx $@ I've also checked this on a PPC machine - there, the program also aborts abnormally but with fewer messages after the 'Mutex destroy failure: Resource busy'.
Jens Jens; From the 1.4.2 release notes:. Fix a crash on exit on MacOS X, and the well known 'Mutex destroy failure' error message on Linux (qt frontend, bugs 2549 and 1029). I'm testing a package for 1.4.2 now, will let you know when it is committed.Jeff - Jeffrey S. Whitaker Phone: (303)497-6313 Meteorologist FAX: (303)497-6449 NOAA/OAR/PSD R/PSD1 Email: Jeffrey.S.Whitaker@. 325 Broadway Office: Skaggs Research Cntr 1D-124 Boulder, CO, USA Web. Hi, I have lyx-qt 1.4.1-1 installed from unstable with fink 0.8.1.rsync i386 on a MacBook Pro, and get a crash when I do a particular sequence of steps in lyx.
I've reported this to the lyx developers but also wanted to report it here, in case anyone else experiences this (i've now checked this on three different machines, PPC and Intel, and also on the Non-X11 Qt LyX-Mac version 1.4.2): Try opening a new file, type a few words with a math inset among them. Save the file. Then erase the math inset and save again. Immediately afterwards, exit lyx.
That causes a crash. The error message on the terminal is: Mutex destroy failure: Resource busy lyx: SIGSEGV signal caught Sorry, you have found a bug in LyX. Please read the bug-reporting instructions in Help-Introduction and send us a bug report, if necessary. /sw/bin/lyx: line 15: 4178 Abort trap /sw/bin/lyxx $@ I've also checked this on a PPC machine - there, the program also aborts abnormally but with fewer messages after the 'Mutex destroy failure: Resource busy'. Philip Ershler wrote: Failed: phase compiling: soqt21-1.2.0-3 failed Any reason why you are still using the 10.4-transitional tree?
What you are seeing is probably some effect of it rotting in its grave. On the Fink home page it says that that tree will be officially unsupported from August 1st, but in reality it has not been completely supported for some time already. In particular, there was a patch related to this problem for the soqt21 package in the 10.4 tree (now at version 1.2.0-1004), but not in the 10.4-transitional tree.
I just did a rebuild so soqt on my 10.4.7 laptop using fink 10.4 and was not able to reproduce and trouble. I do see that I have a newer version of qt3 installed compared to what you have. I know it takes a long time, but it might be worth doing a self update and then building the latest qt to see if there is something up with older Qt versions.
The other thing is to make sure you do NOT have DYLDLIBRARYPATH set. Fink rebuild soqt21. Checking for moc. /sw/bin/moc checking version of Qt library.
336 checking for Qt library devkit.I/sw/include -lqt-mt checking whether the QGL widget is part of main Qt library. Yes checking whether Qt::Keypad is defined.
True checking whether QWidget::showFullScreen is available. True checking whether operator(QGLFormat&,QGLFormat&) is available.kurt I'm sorry, here's a bit more to go on. True checking whether the X11 miscellaneous utilities library is available. Yes checking for moc. /sw/bin/moc checking version of Qt library. 334 checking for Qt library devkit.I/sw/include -lqt-mt checking install sanity.
Bogus The configure script picked out Qt's install program when trying to find the system-wide install program. This probably happened because you have QTDIR/bin too early in the path.
During installation, you would run into serious problems because of this. Fix your PATH variable so the system-wide install program is found instead of the Qt installer and re-run configure. configure: error: aborting ### execution of./configure failed, exit code 1 Removing build lock.
/sw/bin/dpkg-lockwait -r fink-buildlock-soqt21-1.2.0-3 (Reading database. 105497 files and directories currently installed.) Removing fink-buildlock-soqt21-1.2.0-3. Failed: phase compiling: soqt21-1.2.0-3 failed Do You Yahoo!? Tired of spam?
Mail has the best spam protection around.