when using kde 4.2.4 and split qt 4.5.1, when compiling, some files aren't found. which causes for deceleration errors. Reproducible: Always
Created attachment 195683 [details] build log
Please post your "emerge --info".
Created attachment 196631 [details] emerge --info output
Well, I don't think I'll spend time on fixing this. sim was nice 3 years ago, but now... Upstream still works _hard_ to make bug-less qt3! version of sim. Currently I think it's better not to use sim at all. So unless somebody provides us with patch to fix this issue I'll queue it for masking/removal.
I came across to the bug #253624 and it seems possible to fix issue by using the right slot of qt. Can we do the same to work around the problem?
Created attachment 207850 [details, diff] A patch to depend on slotted kdelibs:3.5 Please try this patch, it fixes building with KDE4 installed here.
I don't see how such change in dependencies fixes original problem. Actually it does nothing since in both cases equal atoms will be pulled in. Also Compare deps that were fixed in bug 253624 with what we have in sim: http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-plugins/mythbrowser/mythbrowser-0.21_p17105.ebuild?r1=1.4&r2=1.5 Anyway if somebody could confirm that without this patch he/she has this problem and with it the problem goes away, I'll apply it. Although I'll be very surprised. )
I don't want kdelibs3 in my system so I disabled "kde" USE flag and compiled it using qt3 smoothly in kde4 environment. I've got "sim" icon in application menu, it runs in systray and everything else works as expected. Please don't remove this package yet, we are still using it :)
Created attachment 207895 [details, diff] sim-0.9.4.3-r3.ebuild-no-kde.patch DaggyStyle, could you check if this patch works in your case? (basically it drops kde support)
-9999 should likely be fixed or dropped entirely too...
(In reply to comment #9) > Created an attachment (id=207895) [details] > sim-0.9.4.3-r3.ebuild-no-kde.patch > > DaggyStyle, could you check if this patch works in your case? (basically it > drops kde support) > if it is masked for removed then the towel has been thrown on this software, I see no reason to check it unless you want me to.
> if it is masked for removed then the towel has been thrown on this software It is masked for removal unless somebody provides a patch to fix it. You are expected to test the patch... Does it still compile slow? /It's sometimes worth to work hard on buggy applications like sim, mc, kde4.../
(In reply to comment #12) > > if it is masked for removed then the towel has been thrown on this software > > It is masked for removal unless somebody provides a patch to fix it. You are > expected to test the patch... Does it still compile slow? > > /It's sometimes worth to work hard on buggy applications like sim, mc, kde4.../ > > > that is why I've asked... will try and report.
I've tried to compile, I've got this: checking for mcopidl... not found configure: error: The important program mcopidl was not found! Please check whether you installed aRts correctly or use --without-arts to compile without aRts support (this will remove functionality). !!! Please attach the following file when seeking support: !!! /var/tmp/portage/net-im/sim-0.9.4.3-r4/work/sim-0.9.4.3/config.log * * ERROR: net-im/sim-0.9.4.3-r4 failed. * Call stack: * ebuild.sh, line 49: Called src_compile * environment, line 2671: Called econf '--with-ssl' '--disable-debug' * ebuild.sh, line 534: Called die * The specific snippet of code: * die "econf failed" * The die message: * econf failed * * If you need support, post the topmost build error, and the call stack if relevant. * A complete build log is located at '/var/tmp/portage/net-im/sim-0.9.4.3-r4/temp/build.log'. * The ebuild environment file is located at '/var/tmp/portage/net-im/sim-0.9.4.3-r4/temp/environment'. * This ebuild is from an overlay named 'Own Portage Overlay': '/usr/local/portage/' * --------------------------- ACCESS VIOLATION SUMMARY --------------------------- LOG FILE "/var/log/sandbox/sandbox-20755.log" VERSION 1.0 FORMAT: F - Function called FORMAT: S - Access Status FORMAT: P - Path as passed to function FORMAT: A - Absolute Path (not canonical) FORMAT: R - Canonical Path FORMAT: C - Command Line F: open_wr S: deny P: /usr/qt/3/etc/settings/.qt_plugins_3.3rc.lock A: /usr/qt/3/etc/settings/.qt_plugins_3.3rc.lock R: /usr/qt/3/etc/settings/.qt_plugins_3.3rc.lock C: /usr/qt/3/bin/uic -L /usr/kde/3.5/lib/kde3/plugins/designer -nounload -impl actest.h actest.ui F: open_wr S: deny P: /usr/qt/3/etc/settings/.qt_plugins_3.3rc.lock A: /usr/qt/3/etc/settings/.qt_plugins_3.3rc.lock R: /usr/qt/3/etc/settings/.qt_plugins_3.3rc.lock C: /usr/qt/3/bin/uic -L /usr/kde/3.5/lib/kde3/plugins/designer -nounload -impl actest.h actest.ui -------------------------------------------------------------------------------- I think it is hal related
Created attachment 207984 [details] sim-0.9.4.3-r3.ebuild-nokde.patch err, true. arts should be explicitly disabled. Could you try this diff, please?
Created attachment 207986 [details] with kde disabled failed again. You need to add "--disable-kde" as well. The correct ebuild attached
Thank you Anton! I've pushed changes to the tree. DaggyStyle could you check if it works for you now?
Created attachment 207992 [details] error log compiles well, fails due to sandbox violation.
(In reply to comment #18) well, I know nothing about sandboxes, but your version (sys-apps/sandbox-2.0) is not in the portage. I only can suggest to roll back to the stable branch sys-apps/sandbox-1.6-r2 and file a separate report if the problem exist. Also, make sure you don't have anything weird in your /etc/make.conf
Hm, interesting. Looking at qt3 eclass takes care of this so I've added inherit qt3. Could you retest with qt3 added to inherit line?
Dear Mr. Volkov, it wasn't a good idea to disable kde useflag on SIM completely. Im still using kde3.5 and SIM works fine and compiles fine with it. So i would suggest to disable this useflag for kde4 only.
(In reply to comment #21) > Dear Mr. Volkov, it wasn't a good idea to disable kde useflag on SIM > completely. Im still using kde3.5 and SIM works fine and compiles fine with > it. So i would > suggest to disable this useflag for kde4 only. > No. KDE3 is unsupported and in process of being removed from Portage. Ebuild with KDE3 support is "a bug".
(In reply to comment #19) > (In reply to comment #18) > well, I know nothing about sandboxes, but your version (sys-apps/sandbox-2.0) > is not in the portage. > I only can suggest to roll back to the stable branch sys-apps/sandbox-1.6-r2 > and file a separate report if the problem exist. > Also, make sure you don't have anything weird in your /etc/make.conf > actually you're wrong, my sandbox is in portage: [I] sys-apps/sandbox Available versions: 1.2.18.1-r2 (~)1.2.18.1-r3 (~)1.2.20_alpha2-r1 1.6-r2 (~)2.1 {multilib} Installed versions: 2.1(06:57:33 AM 09/26/2009)(multilib) Homepage: http://www.gentoo.org/ Description: sandbox'd LD_PRELOAD hack I can't seem to remember why I've went to that version... I'm going stable as we speak.
(In reply to comment #20) > Hm, interesting. Looking at qt3 eclass takes care of this so I've added inherit > qt3. Could you retest with qt3 added to inherit line? > will retest with a stable sandbox asap
ok, it compiles and runs well, ugly as windows 3.11 but still working
(In reply to comment #25) > ok, it compiles and runs well, ugly as windows 3.11 but still working Just note: Can look nice and conserve a lot of space on the desktop (in comparsion with other IMs) on both kde3 and kde4. I understand it is hard to maintain, but premature removal of sim, kde3 and other great and *realiable* applications will result in unusable desktops with experimental kde4 and unusable plasmoids (file watcher e.g.) on it. If there will be something better one day, I will switch, but until... http://ash.ic.cz/images/kde3.png http://ash.ic.cz/images/kde4.png
Just a final note, here is how to upgrade if you've been using it with KDE3: mv ~/.kde3.5/share/apps/sim ~/.sim Peter, you might want to add it to the elog message. Thank you everyone for fixing it.
sim unmasked and this bug (hopefully) is fixed. Thank you guys for cooperation. (In reply to comment #10) > -9999 should likely be fixed or dropped entirely too... Well, I'm not sure here. It has different build system (cmake based) so probably it just works and should be moved to kde-3.5 overlay. I'll keep it as is until somebody reports anything new since I don't have kde around these days.
Anton, btw, I've added note into ebuild. Thank you.
(In reply to comment #28) > (In reply to comment #10) > > -9999 should likely be fixed or dropped entirely too... > Well, I'm not sure here. It has different build system (cmake based) so > probably it just works and should be moved to kde-3.5 overlay. I'll keep it as > is until somebody reports anything new since I don't have kde around these > days. Nobody is going to do that for you, so it's a present bug until you do the move. Every dev has a commit access to the repo. Having it in tree as is, is a *bug*.
