Private slots and symbols file
Hi, I hope this is the right place to ask... I'm maintaining liblastfm, a Qt based library, and I'm using pkgkde-symbolshelper to keep track of the symbols. A recent upstream commit removed a private slot from one of the classes, and it's causing dpkg-gensymbols to complain about the missing symbol. Since it's a private slot, and external programs shouldn't be accessing it anyway, is it safe to mark its symbol as optional? John Stamp -- To UNSUBSCRIBE, email to debian-kde-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130108180152.GA14954@panurge
Bug#691294: phonon-backend-vlc: Segfaults on program close; fixed in upstream a8cda824
Package: phonon-backend-vlc Version: 0.6.0-1 Severity: normal Hi, I've been testing the Last.fm scrobbling client located here: https://github.com/lastfm/lastfm-desktop But if I use phonon's vlc backend, the client consistently segfaults on exit. The relevant snippet from valgrind has this: ==9018== Invalid read of size 8 ==9018==at 0x26653959: QListPhonon::VLC::SinkNode*::removeAll(Phonon::VLC::SinkNode* const) (qlist.h:725) ==9018==by 0x26653332: Phonon::VLC::MediaObject::removeSink(Phonon::VLC::SinkNode*) (mediaobject.cpp:737) ==9018==by 0x26654C00: Phonon::VLC::SinkNode::disconnectFromMediaObject(Phonon::VLC::MediaObject*) (sinknode.cpp:66) ==9018==by 0x266391DB: Phonon::VLC::AudioOutput::~AudioOutput() (audiooutput.cpp:44) ==9018==by 0x26639218: Phonon::VLC::AudioOutput::~AudioOutput() (audiooutput.cpp:46) ==9018==by 0x7DE1433: Phonon::MediaNodePrivate::deleteBackendObject() (medianode.cpp:81) ==9018==by 0x7DCDCEF: Phonon::FactoryPrivate::~FactoryPrivate() (factory.cpp:236) ==9018==by 0x7DCDFD8: Phonon::FactoryPrivate::~FactoryPrivate() (factory.cpp:246) ==9018==by 0x988BEA7: qt_call_post_routines() (in /usr/lib/x86_64-linux-gnu/libQtCore.so.4.8.2) ==9018==by 0x88FC57D: QApplication::~QApplication() (in /usr/lib/x86_64-linux-gnu/libQtGui.so.4.8.2) ==9018==by 0x429F28: main (main.cpp:97) ==9018== Address 0x2526d5c8 is 168 bytes inside a block of size 208 free'd ==9018==at 0x4C279DC: operator delete(void*) (vg_replace_malloc.c:457) ==9018==by 0x7DE1433: Phonon::MediaNodePrivate::deleteBackendObject() (medianode.cpp:81) ==9018==by 0x7DCDCEF: Phonon::FactoryPrivate::~FactoryPrivate() (factory.cpp:236) ==9018==by 0x7DCDFD8: Phonon::FactoryPrivate::~FactoryPrivate() (factory.cpp:246) ==9018==by 0x988BEA7: qt_call_post_routines() (in /usr/lib/x86_64-linux-gnu/libQtCore.so.4.8.2) ==9018==by 0x88FC57D: QApplication::~QApplication() (in /usr/lib/x86_64-linux-gnu/libQtGui.so.4.8.2) ==9018==by 0x429F28: main (main.cpp:97) The following upstream commit fixes it by guarding a pointer. No more segfaults :) https://projects.kde.org/projects/kdesupport/phonon/phonon-vlc/repository/revisions/a8cda824ae8a81059c75e97d8a82d571ba84e459/diff.diff Regards, John Stamp -- To UNSUBSCRIBE, email to debian-qt-kde-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20121023232227.GA8541@panurge
Bug#506814: phonon 4.2.0-2 can cause Qt widget problems
Package: phonon Version: 4:4.2.0-2 Severity: normal It looks like the fix for #498573 created another bug. Compile Qt's musicplayer: http://doc.trolltech.com/4.4/phonon-musicplayer.html Install KDE4, but run musicplayer in a non-KDE environment. The player will be able to find the Phonon plugins, but it will also use KDE's oxygen style, KDE's file dialog, and the menu and toolbar text will be invisible. You can still click on the areas though. I downgraded to 4.2.0-1 and was able to duplicate this by adding /usr/lib/kde4/plugins to QT_PLUGIN_PATH. So it looks like Qt needs to be more selective about which plugins it has access to when it's running in a non-KDE environment. I tried Vincent's suggestion about a symlink just for phonon_backend. That gets me audio and Qt's normal widget behavior. John Stamp -- System Information: Debian Release: lenny/sid APT prefers testing APT policy: (400, 'testing'), (200, 'unstable'), (100, 'experimental') Architecture: i386 (i686) Kernel: Linux 2.6.26-10 (SMP w/2 CPU cores; PREEMPT) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash Versions of packages phonon depends on: ii libphonon44:4.2.0-2 Phonon multimedia framework for Qt ii phonon-backend-gstreamer [pho 4:4.2.0-2 Phonon GStreamer 0.10.x backend ii phonon-backend-xine [phonon-b 4:4.1.3-1 Phonon Xine 1.1.x backend phonon recommends no packages. Versions of packages phonon suggests: pn phonon-backend-mplayernone (no description available) pn phonon-backend-vlcnone (no description available) ii phonon-backend-xine 4:4.1.3-1 Phonon Xine 1.1.x backend -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#506814: phonon 4.2.0-2 can cause Qt widget problems
On Monday 24 November 2008 04:15:56 pm Modestas Vainius wrote: Hello, antradienis 25 Lapkritis 2008, John Stamp rašė: backend ii phonon-backend-xine [phonon-b 4:4.1.3-1 Phonon Xine 1.1.x backend phonon-backend-xine comes from KDE. Please remove it and try again. Try using GStreamer backend instread which the only official Phonon 4.2.0 backend we ship. I removed phonon-backend-xine. Only the GStreamer backend is installed. I see the same problem: KDE dialogs, invisible text, etc. To be more specific about how I tested: I created a clean test account to ensure there weren't any lingering config files, started an xfce4 session, and ran Qt's musicplayer example in an xterm. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#498573: libqt4-dev: Please, enable Phonon support
On Sunday 26 October 2008 10:01:38 am Sune Vuorela wrote: On Friday 24 October 2008 16:01:47 Felix Homann wrote: Hi, thank you Sune for your reply. I'm very sorry I didn't answer earlier. I've tried again building the musicplayer example from the Qt4 examples. It builds but it doesn't work. I'm getting error messages of this kind: Phonon::createPath: Cannot connect Phonon::MediaObject(no objectName) to Phonon::AudioOutput(no objectName). WARNING: bool Phonon::FactoryPrivate::createBackend() phonon backend plugin could not be loaded Unfortunately, the example isn't build in qt4-demos either. What's wrong? which phonon backends do you have installed ? /Sune About the phonon backend error, see the discussion here: http://bugs.archlinux.org/task/11070 Sune, I don't know if you'd consider this a problem with the Qt or phonon packages, but it seems that by default Qt applications look for the phonon plugins under /usr/lib/qt4/plugins/. However, Debian installs them under /usr/lib/kde4/plugins/. startkde takes care of that by setting the QT_PLUGIN_PATH environment var. Something in there also causes ~/.config/Trolltech.conf to update the 4.4\libraryPath entry. Unfortunately, there's two problems with this: 1. If a person hasn't run a KDE4 session at least once, Qt doesn't know where to find phonon's plugins. 2. For some reason the 4.4\libraryPath entry is getting flooded with repetitions. A new pair appears each time a KDE4 session starts. John Stamp -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#372258: kwalletmanager does not appear in kde system tray
It looks like bugs 372564 and 372258 are both related to this change: http://websvn.kde.org/branches/KDE/3.5/kdeutils/kwallet/kwalletmanager.cpp?rev=545035view=rev John Stamp -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#372564: kwalletmanager very often crashes at KDE Logout
It looks like a problem with a patch in Debian's kdeutils branch update (r548145). Under certain circumstances, kwalletmanager exits and deletes _dcopRef before it's actually been created. We just need to create _dcopRef earlier. I've attached a slight modification to the patch that seems to fix it for me. John --- kwallet/kwalletmanager.cpp +++ kwallet/kwalletmanager.cpp @@ -55,22 +55,31 @@ _shuttingDown = false; KConfig cfg(kwalletrc); // not sure why this setting isn't in kwalletmanagerrc... KConfigGroup walletConfigGroup(cfg, Wallet); + _dcopRef = new DCOPRef(kded, kwalletd); if (walletConfigGroup.readBoolEntry(Launch Manager, true)) { _tray = new KSystemTray(this, kwalletmanager tray); _tray-setPixmap(loadSystemTrayIcon(wallet_closed)); QToolTip::add(_tray, i18n(KDE Wallet: No wallets open.)); - connect(_tray,SIGNAL(quitSelected()),SLOT(shuttingDown())); + connect(_tray, SIGNAL(quitSelected()), SLOT(shuttingDown())); QStringList wl = KWallet::Wallet::walletList(); + bool isOpen = false; for (QStringList::Iterator it = wl.begin(); it != wl.end(); ++it) { if (KWallet::Wallet::isOpen(*it)) { _tray-setPixmap(loadSystemTrayIcon(wallet_open)); QToolTip::remove(_tray); QToolTip::add(_tray, i18n(KDE Wallet: A wallet is open.)); +isOpen = true; break; } } + if (!isOpen kapp-isRestored()) { + delete _tray; + _tray = 0L; + kapp-exit(); + return; + } } else { - _tray = 0; + _tray = 0L; } _iconView = new KWalletIconView(this, kwalletmanager icon view); @@ -81,7 +90,6 @@ setCentralWidget(_iconView); _iconView-setMinimumSize(320, 200); - _dcopRef = new DCOPRef(kded, kwalletd); _dcopRef-dcopClient()-setNotifications(true); connect(_dcopRef-dcopClient(), SIGNAL(applicationRemoved(const QCString)),
Bug#372564: kwalletmanager very often crashes at KDE Logout
Duh. Well that was a dumb way for me to submit a new patch. How about this instead? Drop 14_kwalletmanager.diff in debian/patches and recompile. John diff -urN kdeutils-3.5.3.orig/kwallet/kwalletmanager.cpp kdeutils-3.5.3/kwallet/kwalletmanager.cpp --- kdeutils-3.5.3.orig/kwallet/kwalletmanager.cpp 2006-06-13 14:00:19.0 -0700 +++ kdeutils-3.5.3/kwallet/kwalletmanager.cpp 2006-06-13 14:01:05.0 -0700 @@ -55,6 +55,7 @@ _shuttingDown = false; KConfig cfg(kwalletrc); // not sure why this setting isn't in kwalletmanagerrc... KConfigGroup walletConfigGroup(cfg, Wallet); + _dcopRef = new DCOPRef(kded, kwalletd); if (walletConfigGroup.readBoolEntry(Launch Manager, true)) { _tray = new KSystemTray(this, kwalletmanager tray); _tray-setPixmap(loadSystemTrayIcon(wallet_closed)); @@ -89,7 +90,6 @@ setCentralWidget(_iconView); _iconView-setMinimumSize(320, 200); - _dcopRef = new DCOPRef(kded, kwalletd); _dcopRef-dcopClient()-setNotifications(true); connect(_dcopRef-dcopClient(), SIGNAL(applicationRemoved(const QCString)),
Bug#329484: kdesktop: Please add KDE Improvements patch for rounded edges on desktop icons
Package: kdesktop Version: 4:3.4.2-3 Severity: minor Tags: patch kdelibs already uses the KDE Improvements patch that creates rounded edges in Icon View. A similar patch is also available for desktop icons. I've attached the relevant patch. John Stamp -- System Information: Debian Release: testing/unstable APT prefers stable APT policy: (500, 'stable'), (400, 'testing'), (200, 'unstable'), (199, 'experimental') Architecture: i386 (i686) Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.12-6 Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) (ignored: LC_ALL set to en_US.UTF-8) Versions of packages kdesktop depends on: ii kdebase-bin 4:3.4.2-3 core binaries for the KDE base mod ii kdelibs4c24:3.4.2-4 core libraries for all KDE applica ii libart-2.0-2 2.3.17-1 Library of functions for 2D graphi ii libaudio2 1.7-3 The Network Audio System (NAS). (s ii libc6 2.3.5-6GNU C Library: Shared libraries an ii libfontconfig12.3.2-1generic font configuration library ii libfreetype6 2.1.7-2.4 FreeType 2 font engine, shared lib ii libgcc1 1:4.0.1-2 GCC support library ii libglu1-xorg [libglu1]6.8.2.dfsg.1-7 Mesa OpenGL utility library [X.Org ii libice6 6.8.2.dfsg.1-7 Inter-Client Exchange library ii libidn11 0.5.18-1 GNU libidn library, implementation ii libjpeg62 6b-10 The Independent JPEG Group's JPEG ii libkonq4 4:3.4.2-3 core libraries for Konqueror ii libpng12-01.2.8rel-1 PNG library - runtime ii libqt3-mt 3:3.3.5-1 Qt GUI Library (Threaded runtime v ii libsm66.8.2.dfsg.1-7 X Window System Session Management ii libstdc++64.0.1-2The GNU Standard C++ Library v3 ii libx11-6 6.8.2.dfsg.1-7 X Window System protocol client li ii libxcursor1 1.1.3-1X cursor management library ii libxext6 6.8.2.dfsg.1-7 X Window System miscellaneous exte ii libxft2 2.1.7-1FreeType-based font drawing librar ii libxi66.8.2.dfsg.1-7 X Window System Input extension li ii libxinerama1 6.8.2.dfsg.1-7 X Window System multi-head display ii libxrandr26.8.2.dfsg.1-7 X Window System Resize, Rotate and ii libxrender1 1:0.9.0-2 X Rendering Extension client libra ii libxss1 6.8.2.dfsg.1-7 X Screen Saver client-side library ii libxt66.8.2.dfsg.1-7 X Toolkit Intrinsics ii libxxf86misc1 6.8.2.dfsg.1-7 X miscellaneous extensions library ii xlibmesa-gl [libgl1] 6.8.2.dfsg.1-7 Mesa 3D graphics library [X.Org] ii xlibs 6.8.2.dfsg.1-7 X Window System client libraries m ii zlib1g1:1.2.3-4 compression library - runtime Versions of packages kdesktop recommends: ii eject 2.0.13deb-13 ejects CDs and operates CD-Changer -- no debconf information kdebase_desktop_roundedge-icons.diff.gz Description: Binary data
Possible bug with konqueror listviews?
When konqueror 3.3.0 shows files in any of the following formats: Tree View Info List View Detailed List View Text View I frequently have problems: opening files and directories: no action selecting and moving groups of files: the file group can be selected, but when I try to move them, the group is deselected right clicking on a file: doing so deselects the file This doesn't happen all of the time, though. I can sometimes get these things to work again by switching to Icon View and then back. But then I can only travel through a few directories before these problems crop up again. Icon and Multicolumn Views don't seem affected. Has anyone else seen this? John Stamp
Bug#227802: kmix: tray icon displays absurd volume levels (e.g. 24119703%)
Package: kmix Version: 4:3.1.5-1 Severity: normal Tags: patch Kmix 3.1.4 and 3.1.5 show ridiculous volume levels until the volume is manually adjusted. See KDE bug #64274. This was fixed upstream on 11/24/03 but won't be available until KDE 3.2 It's a very simple fix, so I attached a patch. Any chance of including it soon? Thanks. John Stamp -- System Information: Debian Release: testing/unstable Architecture: i386 Kernel: Linux lipsius 2.6.1 #1 Tue Jan 13 10:35:17 PST 2004 i686 Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 Versions of packages kmix depends on: ii kdelibs4 4:3.1.5-1 KDE core libraries ii libart-2.0-2 2.3.16-1 Library of functions for 2D graphi ii libasound2 0.9.8-2 Advanced Linux Sound Architecture ii libaudio2 1.6b-1The Network Audio System (NAS). (s ii libc6 2.3.2.ds1-10 GNU C Library: Shared libraries an ii libfontconfig1 2.2.1-13 generic font configuration library ii libfreetype6 2.1.7-1.1 FreeType 2 font engine, shared lib ii libgcc11:3.3.2-4 GCC support library ii libice64.3.0-0pre1v5 Inter-Client Exchange library ii libpng12-0 1.2.5.0-4 PNG library - runtime ii libqt3c102-mt 3:3.2.3-1 Qt GUI Library (Threaded runtime v ii libsm6 4.3.0-0pre1v5 X Window System Session Management ii libstdc++5 1:3.3.2-4 The GNU Standard C++ Library v3 ii libx11-6 4.3.0-0pre1v5 X Window System protocol client li ii libxcursor11.0.2-4 X Cursor management library ii libxext6 4.3.0-0pre1v5 X Window System miscellaneous exte ii libxft22.1.2-5 FreeType-based font drawing librar ii libxine1 1-rc3a-2 the xine video/media player librar ii libxmu64.3.0-0pre1v5 X Window System miscellaneous util ii libxrender10.8.3-5 X Rendering Extension client libra ii libxt6 4.3.0-0pre1v5 X Toolkit Intrinsics ii xlibmesa-gl [libgl1] 4.3.0-0pre1v5 Mesa 3D graphics library [XFree86] ii xlibs 4.3.0-0pre1v5 X Window System client libraries m ii zlib1g 1:1.2.1-3 compression library - runtime -- no debconf information kmix-volume.diff.gz Description: Binary data
Re: k3b kernel 2.6 (cdrdao)
On Tuesday 13 January 2004 02:20 am, Antonio Rodriguez wrote: On Tue, Jan 13, 2004 at 12:09:46AM +0100, Jean-Michel Kelbert wrote: OK. I'll report a bug anyway. Thanks for checking. I'll solve it in next version. Please set bug to normal. Too much users are using bad levels. -- Jean-Michel Kelbert I have a question: would this fix affect only k3b, or cdrdao in general? I have ATAPI, a very difficult HP cdwritter that has been reported as conflicting with scsi in the kernel (I had to remove scsi from kernel, kept throwing it into panick), therefore I have been using cdrecord with ATAPI. But I also like using cdrdao from command line, besides k3b, etc. If not, could someone explain what the fix would be for cdrdao? I got late on the thread. Thanks to all. The fix only affects k3b. It's bug #227336 For cdrdao to work with kernel 2.6 atapi you will need cdrdao 1.1.7-5. It's in both testing and unstable. To run it from the command line, I use options: --device /dev/hdc --driver generic-mmc Hope this helps. John Stamp
Re: k3b kernel 2.6
On Sunday 11 January 2004 06:06 pm, Germain CHAZOT wrote: Le Mardi 30 Décembre 2003 23:56, Todd Charron a écrit : You may be wright. I've noticed that there are two icons to duplicate a CD; these are translations from spanish, but one's called copy cd and the other is called duplicate cd, and using the second one the rw appears as reader so I can copy cds ;) I can copy CDs through Clone CD as well, however, that always makes an image and doesn't appear to burn on the fly. That's the same for me, copy cd doesn't show any avaible reader and clone cd shows one but always makes an image and doesn't offer to copy on-the-fly. I tried to copy an audio cd but it doesn't work... maybe because of the copy with an image? I didn't understand why it made an image (and only one) for an audio cd. Maybe someone could also explain why copy cd doesn't show any reader? Thanks I had the same problem, too. It looks like k3b assumes that cdrdao 1.1.7-5 cannot handle the new kernel ATAPI interface. You need to change a line to let k3b recognize cdrdao's new abilities. Anyway, this did the trick for me: --- k3b-0.10.3.orig/src/core/k3bdefaultexternalprograms.cpp +++ k3b-0.10.3/src/core/k3bdefaultexternalprograms.cpp @@ -521,7 +521,7 @@ // SuSE 9.0 ships with a patched cdrdao 1.1.7 which contains an updated libschily - if( bin-version K3bVersion( 1, 1, 7 ) || + if( bin-version K3bVersion( 1, 1, 6 ) || ( bin-version == K3bVersion( 1, 1, 7 ) bin-version.suffix() == -suse ) ) { bin-addFeature( plain-atapi ); bin-addFeature( hacked-atapi );
Re: k3b kernel 2.6
Germain CHAZOT wrote: shouldn't this patch be proposed for k3b, if isn't already set in mainstream? But I don't know and didn't find if the version of cdrdao in Cid supports atapi with kernel 2.6? If so, the code above should also check the kernel version rather than base only on cdrdao version. Correct me if I am wrong but it is the combination of cdrdao 1.1.7 and kernel 2.6 that makes trouble. Because cdrdao handles atapi devices only with kernel 2.6 (without scsi emulation I mean). Maybe I misunderstand all this Cdrdao in Sid does support atapi on kernel 2.6. And yes, the problem appears to be related to cdrdao using the new atapi interface. K3b assumes that cdrdao can't handle it. I should file a bug report, but I haven't tested this much. I just wanted something that would quickly work for me. Can you test if the patch works for you too? If so, I'll file a report and include the little kludge.
Re: k3b kernel 2.6
Todd Charron wrote: Cdrdao in Sid does support atapi on kernel 2.6. As I've just found out recently (through reporting it incorrectly as a bug myself) cdrdao is indeed patched to support ATAPI. I can confirm this as I've tested it myself today. cdrdao copy --source-device /dev/hdc --device /dev/hdd --speed 24 (etc options) The problem from what I can see is that K3B assumes that the support is not there based on the version number and thus disables usage of the reader in copying. I'll test that patch out myself later on when I get a moment to see if I can confirm it, but if you could report a bug on it including the patch it would be much appreciated. Btw what effect will this line have on the Debian sources? bin-version.suffix() == -suse ) ) { Should this be changed to something to reflect debian's cdrdao package? Todd OK. I'll report a bug anyway. Thanks for checking. I don't think the suse line has much effect. Based on the comment in k3b, my guess is that Suse appended -suse to their version of cdrdao because it included the backported atapi patch. Debian has included the patch since cdrdao 1.1.7-4, but it doesn't seem to play around with a version suffix. That's why I resorted to changing the numeric check. John Stamp