Bug#658236: cups: unbuildable on hurd-i386: missing libusb-1.0
Package: cups Version: 1.5.0-16 Severity: important Tags: patch User: debian-h...@lists.debian.org Usertags: hurd Hi, since cups 1.5.0-14, the libusb library is the new one, 1.0, which is not available on Hurd; this makes the cups source unbuildable. Attached there is a patch to exclude the libusb-1.0-0-dev b-d on hurd, and disable the libusb detection in configure. Thanks, -- Pino --- a/debian/control +++ b/debian/control @@ -20,7 +20,7 @@ libijs-dev, libjpeg-dev, libldap2-dev, libkrb5-dev, libpam0g-dev, libpaper-dev, libpng12-dev, libpoppler-dev (>= 0.12.2-2), - libslp-dev, libtiff4-dev, libtool, libusb-1.0-0-dev, + libslp-dev, libtiff4-dev, libtool, libusb-1.0-0-dev [!hurd-any], patch, pkg-config, po4a (>= 0.31~), po-debconf, poppler-utils, sharutils, zlib1g-dev Homepage: http://www.cups.org --- a/debian/rules +++ b/debian/rules @@ -17,7 +17,12 @@ include /usr/share/cdbs/1/rules/debhelper.mk include /usr/share/cdbs/1/class/autotools.mk -DEB_CONFIGURE_EXTRA_FLAGS := --with-optim=$(DEB_OPTFLAGS) --libdir=/usr/lib/$(DEB_HOST_MULTIARCH) --mandir=/usr/share/man --with-docdir=/usr/share/cups/doc-root --localedir=/usr/share/cups/locale --enable-slp --enable-libpaper --enable-ssl --enable-gnutls --disable-openssl --enable-threads --enable-static --enable-debug --enable-libusb --enable-dbus --with-dbusdir=/etc/dbus-1 --enable-gssapi --enable-avahi --with-pdftops=/usr/bin/gs --disable-launchd --with-cups-group=lp --with-system-groups=lpadmin --with-printcap=/var/run/cups/printcap --with-log-file-perm=0640 CFLAGS="$(CFLAGS)" --with-local_protocols='CUPS dnssd' --with-remote_protocols='CUPS dnssd' +DEB_CONFIGURE_EXTRA_FLAGS := --with-optim=$(DEB_OPTFLAGS) --libdir=/usr/lib/$(DEB_HOST_MULTIARCH) --mandir=/usr/share/man --with-docdir=/usr/share/cups/doc-root --localedir=/usr/share/cups/locale --enable-slp --enable-libpaper --enable-ssl --enable-gnutls --disable-openssl --enable-threads --enable-static --enable-debug --enable-dbus --with-dbusdir=/etc/dbus-1 --enable-gssapi --enable-avahi --with-pdftops=/usr/bin/gs --disable-launchd --with-cups-group=lp --with-system-groups=lpadmin --with-printcap=/var/run/cups/printcap --with-log-file-perm=0640 CFLAGS="$(CFLAGS)" --with-local_protocols='CUPS dnssd' --with-remote_protocols='CUPS dnssd' +ifeq ($(DEB_HOST_ARCH_OS),hurd) +DEB_CONFIGURE_EXTRA_FLAGS += --disable-libusb +else +DEB_CONFIGURE_EXTRA_FLAGS += --enable-libusb +endif DEB_MAKE_INSTALL_TARGET := install BUILDROOT=$(DEB_DESTDIR) DEB_INSTALL_CHANGELOGS_ALL := CHANGES.txt
Bug#658237: cups: extra libpoppler-dev build dependency
Package: cups Version: 1.5.0-16 Severity: wishlist Tags: patch Hi, with cups 1.5.0-16 the external filters, and mong them the ones using the libpoppler API, have been moved to the cups-filters source. This makes the libpoppler-dev build dependency redundant now, and it can be removed; patch attached for it. Thanks, -- Pino --- a/debian/control +++ b/debian/control @@ -19,7 +19,7 @@ Build-Depends: libdbus-1-dev, libfontconfig1-dev, libfreetype6-dev, libgnutls-dev, libijs-dev, libjpeg-dev, libldap2-dev, libkrb5-dev, - libpam0g-dev, libpaper-dev, libpng12-dev, libpoppler-dev (>= 0.12.2-2), + libpam0g-dev, libpaper-dev, libpng12-dev, libslp-dev, libtiff4-dev, libtool, libusb-1.0-0-dev, patch, pkg-config, po4a (>= 0.31~), po-debconf, poppler-utils, sharutils, zlib1g-dev
Bug#648993: ecl: FTBFS on hurd-i386
Hi,
Alle mercoledì 16 novembre 2011, Svante Signell ha scritto:
> The attached patch fixes the FTBFS problems of ecl on GNU/Hurd. When
> native threads are enabled the "./ecl_min compile" command hangs.
Preface: libgc, used by ecl, makes use of the SIGUSR1 & SIGUSR2 signals
(handling them on its own) to respectively suspend and & threads when
the main thread needs to "stop the world" and garbage collect or so.
In ecl, src/c/unixint.d, there is the following function:
| static void
| install_process_interrupt_handler()
| {
| #ifdef SIGRTMIN
| # define DEFAULT_THREAD_INTERRUPT_SIGNAL SIGRTMIN + 2
| #else
| # define DEFAULT_THREAD_INTERRUPT_SIGNAL SIGUSR1
| #endif
| #if defined(ECL_THREADS) && !defined(ECL_MS_WINDOWS_HOST)
| if (ecl_get_option(ECL_OPT_TRAP_INTERRUPT_SIGNAL)) {
| int signal = ecl_get_option(ECL_OPT_THREAD_INTERRUPT_SIGNAL);
| if (signal == 0) {
| signal = DEFAULT_THREAD_INTERRUPT_SIGNAL;
| ecl_set_option(ECL_OPT_THREAD_INTERRUPT_SIGNAL,
|signal);
| }
| mysignal(signal, non_evil_signal_handler);
| #ifdef HAVE_SIGROCMASK
| sigdelset(cl_core.default_sigmask, signal);
| pthread_sigmask(SIG_SETMASK, cl_core.default_sigmask, NULL);
| #endif
| }
| #endif
| }
given on Hurd we don't have realtime signals yet, SIGRTMIN (and
SIGRTMAX) is not defined, which makes SIGUSR1 the signal being used for
the inter-thread communication, overriding the SIGUSR1 handler set by
libgc.
The solution (a bit of hack, actually) I have is use a different signal
for DEFAULT_THREAD_INTERRUPT_SIGNAL (I chose SIGINFO, generally not used
and not POSIX); this allowed me to compile ecl correctly.
I attached the patch for it; note I don't think it should be forwarded
upstream, as it is not clean and kind of working around the lack of
realtime signals in Hurd.
--
Pino Toscano
--- a/src/c/unixint.d
+++ b/src/c/unixint.d
@@ -1073,6 +1073,8 @@
{
#ifdef SIGRTMIN
# define DEFAULT_THREAD_INTERRUPT_SIGNAL SIGRTMIN + 2
+#elif defined(__GNU__)
+# define DEFAULT_THREAD_INTERRUPT_SIGNAL SIGINFO
#else
# define DEFAULT_THREAD_INTERRUPT_SIGNAL SIGUSR1
#endif
signature.asc
Description: This is a digitally signed message part.
Bug#648993: ecl: FTBFS on hurd-i386
Alle domenica 5 febbraio 2012, Samuel Thibault ha scritto: > Pino Toscano, le Sun 05 Feb 2012 12:56:39 +0100, a écrit : > > note I don't think it should be forwarded upstream, as it is not > > clean and kind of working around the lack of realtime signals in > > Hurd. > > Well, given that ecl explicitly uses libgc, which explicitly uses > SIGUSR1, it sounds reasonable to at least notify them of the issue. libgc doesn't uses SIGUSR1 everywhere, just on FreeBSD (not kFreeBSD) and on Hurd. Everywhere else it uses SIGRTMIN + 5 & SIGRTMIN + 6. What ecl could do is check whether there's a signal handler already set for DEFAULT_THREAD_INTERRUPT_SIGNAL, and try again with other signal(s). Although, this still wouldn't help if ecl would first set the signal handler, and then libgc would do the same. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#659048: avogadro: FTBFS on hurd-i386: missing .desktop file and its icon
Package: avogadro
Version: 1.0.3-3.1
Severity: important
Tags: patch
User: debian-h...@lists.debian.org
Usertags: hurd
Hi,
currently[1], avogadro does not build on hurd-i386.
The problem is that the .desktop file and its icon are not installed
on Hurd, so dh_install fails to install them.
Attached there is a patch which I also forwarded upstream (see [2]).
[1]
https://buildd.debian.org/status/fetch.php?pkg=avogadro&arch=hurd-i386&ver=1.0.3-3.1&stamp=1328631588
[2]
https://sourceforge.net/tracker/?func=detail&aid=3485433&group_id=165310&atid=835079
Thanks,
--
Pino
--- a/avogadro/src/CMakeLists.txt
+++ b/avogadro/src/CMakeLists.txt
@@ -169,10 +169,10 @@
endif(APPLE AND ENABLE_AVO_PACKAGE)
# Install a .desktop file and icon for Linux users
-if (CMAKE_SYSTEM_NAME MATCHES Linux OR CMAKE_SYSTEM_NAME MATCHES kFreeBSD)
+if (CMAKE_SYSTEM_NAME MATCHES Linux OR CMAKE_SYSTEM_NAME MATCHES kFreeBSD OR CMAKE_SYSTEM_NAME STREQUAL GNU)
install(FILES "avogadro.desktop" DESTINATION ${XDG_APPS_INSTALL_DIR})
install(FILES "icons/avogadro-icon.png" DESTINATION ${XDG_ICON_INSTALL_DIR})
-endif(CMAKE_SYSTEM_NAME MATCHES Linux OR CMAKE_SYSTEM_NAME MATCHES kFreeBSD)
+endif(CMAKE_SYSTEM_NAME MATCHES Linux OR CMAKE_SYSTEM_NAME MATCHES kFreeBSD OR CMAKE_SYSTEM_NAME STREQUAL GNU)
#add_subdirectory(projectplugins)
Bug#659150: libgstreamer-perl: unbuildable on non-linux archs
Package: libgstreamer-perl
Version: 0.16-1
Severity: important
Tags: patch
Hi,
currently libgstreamer-perl cannot be built on non-Linux architectures,
because it build-depends on gstreamer0.10-alsa, which exists on Linux
only (since ALSA in specific to Linux).
It seems it is used in some tests of the test suite, although they
seem to skip checks if there is no alsa bin available; hence, the
proposed change is to make gstreamer0.10-alsa, both as build dependency
and recommend, restricted to linux architectures.
Thanks,
--
Pino
--- a/debian/control
+++ b/debian/control
@@ -2,7 +2,7 @@
Section: perl
Priority: optional
Build-Depends: debhelper (>= 8), perl,
- gstreamer0.10-alsa,
+ gstreamer0.10-alsa [linux-any],
gstreamer0.10-plugins-base,
libextutils-depends-perl,
libextutils-pkgconfig-perl,
@@ -19,7 +19,7 @@
Package: libgstreamer-perl
Architecture: any
Depends: ${perl:Depends}, ${shlibs:Depends}, ${misc:Depends}, libglib-perl
-Recommends: gstreamer0.10-plugins-base, gstreamer0.10-alsa
+Recommends: gstreamer0.10-plugins-base, gstreamer0.10-alsa [linux-any]
Description: Perl interface to the GStreamer media processing framework
libgstreamer-perl allows a Perl developer to use the GStreamer libraries
in a perlish way, freeing the programmer from any memory management and
Bug#659225: libgstreamer-interfaces-perl: unbuildable on non-linux archs
Package: libgstreamer-interfaces-perl Version: 0.06-1 Severity: important Tags: patch Hi, currently libgstreamer-interfaces-perl cannot be built on non-Linux architectures, because it build-depends on gstreamer0.10-alsa and alsa-utils, which exist on Linux only (since ALSA in specific to Linux). It seems they are used in a test of the test suite, although it seems to skip checks if there is no alsamixer available; hence, the proposed change is to make the gstreamer0.10-alsa and alsa-utils build dependencies restricted to linux architectures. (Note that wildcard archs are officially supported in Policy >= 3.9.1, while the packaging is still at 3.8.4.) Thanks, -- Pino --- a/debian/control +++ b/debian/control @@ -2,9 +2,10 @@ Section: perl Priority: optional Build-Depends: debhelper (>= 7), libgstreamer-plugins-base0.10-dev (>= 0.10.12), - gstreamer0.10-alsa, alsa-utils, gstreamer0.10-plugins-base, libgstreamer0.10-dev, + gstreamer0.10-plugins-base, libgstreamer0.10-dev, libextutils-depends-perl (>= 0.205), libextutils-pkgconfig-perl (>= 1.07), - libglib-perl (>= 1.180), libgstreamer-perl, perl + libglib-perl (>= 1.180), libgstreamer-perl, perl, + gstreamer0.10-alsa [linux-any], alsa-utils [linux-any] Maintainer: Debian Perl Group Uploaders: Antonio Radici , gregor herrmann
Bug#680946: okular: fails to start, symbol lookup error
tag 680946 + pending thanks Alle lunedì 9 luglio 2012, Damien Couroussé ha scritto: > Package: okular > Version: 4:4.4.5-2 Why didn't you upgrade okular too? Also, > ii libokularcore1 4:4.6.5-3libraries for the Okular document why this old version (not stable, and not testing either)? Ok, the dependencies should be enough (and will be fixed so frankenstein like yours won't happen anymore), but why not upgrade all the okular packages together, and not keep them up-to-date with testing? -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#680929: [okular] unable to open PDF file
Hi, Alle lunedì 9 luglio 2012, Marco Righi ha scritto: > Okular can not open the attached file. It opens fine here (fully up-to-date testing), 3 mostly white pages with some lines of text each. What is the problem you see? Do you get any error dialog? Can you please paste all the output you get when lanching from a terminal okular with your document (`okular thefile.pdf`)? > This is a provate file, please > use only in order to debug the application. Then you should not have attached it to a public bug tracking system. Thanks, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#680929: [okular] unable to open PDF file
Hi, (Note: please do not top-post, thanks.) Alle lunedì 9 luglio 2012, Marco Righi ha scritto: > Command 514 of 20 $okular cessrighi_12070616541842210.pdf > okular(16602)/kdecore (KConfigSkeleton) > KCoreConfigSkeleton::writeConfig: okular(16602)/kdecore > (KConfigSkeleton) KCoreConfigSkeleton::writeConfig: > okular(16602)/kdecore (KConfigSkeleton) > KCoreConfigSkeleton::writeConfig: okular(16602)/kdecore > (KConfigSkeleton) KCoreConfigSkeleton::writeConfig: > okular(16602)/kdecore (KConfigSkeleton) > KCoreConfigSkeleton::writeConfig: > > okular open a windows with a big red square that has inside a big > white X next to the error message > > "Impossibile aprire > /media/data/users/home/marco/Scrivania/tmp/marco_nav/cessrighi_120706 > 16541842210.pdf" Please run kdebugdialog, enable the areas 4700, 4701 and the "okular" one without a number (you can use the line edit to filter them), and save. Then run again okular, it should give a bit more information. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#680929: [okular] unable to open PDF file
Hi, Alle lunedì 9 luglio 2012, Marco Righi ha scritto: > KCoreConfigSkeleton::writeConfig: okular(18989)/kdecore (trader) > KMimeTypeTrader::query: query for mimeType "application/pdf" , > "okular/Generator" : returning 1 offers okular(18989) > PDFGeneratorPopplerDebugFunction: [Poppler] "Error: PDF file is > damaged - attempting to reconstruct xref table..." > okular(18989) PDFGeneratorPopplerDebugFunction: [Poppler] "Error: > Couldn't find trailer dictionary" > okular(18989) PDFGeneratorPopplerDebugFunction: [Poppler] "Error: > Couldn't read xref table" Indeed those could mean that what you are loading is not a PDF file or it is corrupted. But, I don't get those when loading the document you attached in this bug report: are you sure you sent the very same document you are opening here? The one you attached is: $ md5sum cessrighi_12070616541842210.pdf 7a8fa6ff39a0c708edda9bc09fa517dc cessrighi_12070616541842210.pdf Also, can you please paste the output of $ dpkg -l '*poppler*' | grep ^ii ? Thanks, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#681081: poppler: Build-Depend on libtiff5-dev
severity 681081 wishlist tag 681081 + wontfix thanks Hi, Alle martedì 10 luglio 2012, Michael Terry ha scritto: > libtiff5 is now available (source tiff), and it would be nice if > poppler used that instead of libtiff4 (source tiff3). What's the rationale for deviating from the default tiff currently used in Debian? Also, most probably for wheezy+1 libtiff-dev is going to be provided by tiff v4 instead of v3, so poppler will follow suit automatically (and poppler 0.20.x and greater is targeted for wheezy+1, anyway). Hence, tagging it as wontfix. Thanks for the patch anyway, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#680929: [okular] unable to open PDF file
tag 680929 + unreproducible thanks Alle martedì 10 luglio 2012, Marco Righi ha scritto: > > $ md5sum cessrighi_12070616541842210.pdf > > 7a8fa6ff39a0c708edda9bc09fa517dc cessrighi_12070616541842210.pdf > > Command 93 of 3 $md5sum cessrighi_12070616541842210.pdf > 7a8fa6ff39a0c708edda9bc09fa517dc cessrighi_12070616541842210.pdf > > After the reboot okular open the file... Thanks, marking as unreproducible then. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#664681: transition: KDE's 4.8 release of platform, applications and workspace
Alle giovedì 12 luglio 2012, Adam D. Barratt ha scritto: > On 04.07.2012 21:05, Adam D. Barratt wrote: > > On Wed, 2012-06-27 at 10:35 +0200, Pino Toscano wrote: > >> here it is a recap of the status of the various bits currently > >> ongoing > >> so far. > >> 5) split of kdeutils > >> > >> the monolithic kdeutils source has been replaced in KDE 4.8 > >> > >> with the following sources: > >> ark filelight kcalc kcharselect kdf kfloppy kgpg > >> kremotecontrol ktimer kwallet superkaramba printer-applet > >> sweeper > >> > >> the only binaries lost is the monolithic kdeutils-dbg (as > >> > >> expected); > >> > >> this would need to hint the old source away from testing > >> > >> -> this is not ready to migrate yet > > > > kfloppy, kgpg, kremotecontrol, printer-applet and superkaramba > > appear to > > be the sources still to migrate. They should all be ready on > > Friday night, left to their own devices. > > Indeed, those all seem to have migrated now. Is there anything > remaining for 4.8, or could we declare this done now? All the split bits have finally migrated, so I can now file a RM for src:kdeutils. I think we can consider this bug done, yes (there's kde-l10n still at 4.7, but that's waiting for a fixed tasksel to migrate). Thanks for all your patience, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#681386: RM: kdeutils -- ROM; replaced by split sources
Package: ftp.debian.org Severity: normal Hi, the monolithic kdeutils module has been split by upstream since KDE SC 4.8 in few smaller sources; the set of the KDE SC sources replacing kdeutils is: ark, filelight, kcalc, kcharselect, kdf, kfloppy, kgpg, kremotecontrol, ktimer, kwallet, superkaramba, printer-applet, sweeper The only loss package is kdeutils-dbg, but that is expected of course. Thanks, -- Pino -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#679883: calibre: Port to libpoppler25 API
tags 679883 + fixed-upstream thanks Hi, Alle lunedì 2 luglio 2012, Andy Whitcroft ha scritto: > Version: 0.8.51+dfsg-1 > [...] > We are using this patch in Ubuntu to handle the changes in > libpoppler's API version 19 to version 25/26. Please consider for > Debian. It seems newer versions of calibre (I tested the latest one, 0.8.60, released on July 13th) do not use poppler API at all but just invoke the some of the poppler-utils executable. This means packaging a newer upstream release would allow to drop the libpoppler-qt4-dev & libpoppler-private-dev build dependencies and no more be bothered by libpoppler API/ABI bumps. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#679891: pdf2djvu: Port to libpoppler25 API
Hi, Alle lunedì 2 luglio 2012, Andy Whitcroft ha scritto: > Version: 0.7.12-2 > [...] > We are using this patch in Ubuntu to handle the changes in > libpoppler's API version 19 to version 25/26. Please consider for > Debian. Please note that pdf2djvu 0.7.13 compiles fine with poppler >= 0.20, with its test suite (which needs python-nose) passing too. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#679892: pdftoipe: Port to libpoppler25 API
forwarded 679892 https://sourceforge.net/apps/mantisbt/ipe7/view.php?id=112 thanks Hi, Alle lunedì 2 luglio 2012, Andy Whitcroft ha scritto: > We are using this patch in Ubuntu to handle the changes in > libpoppler's API version 19 to version 25/26. Please consider for > Debian. This patch has two issues: - it is not compatible with the poppler version (0.18.x) currently in Debian - the format for the parameters in the error string is wrong (in >= 0.20, it is not printf-like but follows the GooString::format() syntax) I just reported upstream the issue (it was not reported), proposing a patch which would switch the call to error() to a simplier fprintf(). This avoids issues if error() changes again in the future, and it is compatible with any poppler version. Thanks, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#681807: okular: sometimes fails to start, but no error message
severity 681807 important thanks Alle lunedì 16 luglio 2012, Felix Zweig ha scritto: > Sometimes, starting Okular fails without any error message. "fails" is bit too generic; what does it happen? You get back to the console prompt, or it hangs? Since when it is happening? Does the problem happens only when okular is started without a document? > How can I get a more verbose output the further diagnose the problem? Run kdebugdialog and enable all the okular debug areas (even those without a number, use the filtering line edit on the top). -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#662041: FTBFS on several architectures
clone 662041 -1 retitle -1 FTBFS on hurd-i386 severity -1 important user debian-h...@lists.debian.org usertags -1 hurd thanks Hi, (splitting off the Hurd issues) Alle sabato 3 marzo 2012, Kurt Roeckx ha scritto: > On Sat, Mar 03, 2012 at 09:12:13PM +0100, Michael Biebl wrote: > > Source: elfutils > > Version: 0.153-1 > > Severity: serious > > > > The current version of elfutils FTBFS on several architectures: > > hurd-i386 > > Yes, nothing new. It should never have built on hurd, but it > seems some people think it's a good idea to run buildds that > don't run the testsuite. Sorry if it gives you a bad impression, although OTOH there's a number of packages (of which few important) that requires elfutils libraries. > I filed a bug for the problem in hurd's upstream bug tracker ages > ago. Do you mean http://savannah.gnu.org/bugs/?32770 ? Also please note that elfutils is full of linuxisms, such as: - it assumes that /proc/$pid/maps exist (libdwfl/linux-proc-maps.c) - it looks for Linux kernels and modules (libdwfl/linux-kernel-modules.c) there's no bad thing in doing them... as long as they are done only on Linux. Should elfutils be assumed as "Linux" stuff (so other OSes must emulate all the things it assumes to exist), or could it be fixed to check for Linux stuff only on Linux? -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#686147: Bug#580706:
unmerge 686147 thanks Hi, Alle sabato 1 settembre 2012, David Smith ha scritto: > It seems Akonadi, by default is trying to use MySQL even though by > default, the distribution installs akonadi-backend-sqlite, and by > default, mysql isn't installed. No, the default is akonadi-backend-mysql, which is also the first to be installed if no "backends" are installed already: $ apt-cache show akonadi-server Package: akonadi-server Depends: [...], akonadi-backend-mysql (= 1.7.2-1) | akonadi-backend- sqlite (= 1.7.2-1) | akonadi-backend-postgresql (= 1.7.2-1) Can you please paste the results of $ dpkg -l '*akonadi*' | grep ^ii $ dpkg -l '*mysql*' | grep ^ii ? > Is there any possibility at all that the packaging for Akonadi can, > by default, use the sqllite backend? Upstream does not consider it the optimal choice yet, so no. > As that is what the package > manager installs by default for Akonadi when pulling in all of KDE. As said above, it is not. > When you start pulling > in mysql, things start getting complicated as even mysql won't work > unless you configure usernames and passwords and such. akonadi-backend-mysql depends only on the mysql server , without the init scripts not other system-wide setup (including global passwords). Also note that this bug (#580706) is turning into a black hold of different issues, and merging #686147 with this was not a good move and has been reverted. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#686677: freemedforms-libs: hardcoded qt4 library dependendies
Package: freemedforms-libs Version: 0.7.4-1 Severity: important Hi, freemedforms-libs hardcodes the library dependencies for all the qt4 libraries it uses; in particular - libqt4-core, libqt4-gui: these two are transitional packages, and their counterparts would get a dependency on >= 4:4.7.0~beta1 already - libqt4-network, libqt4-sql, libqt4-svg, libqt4-xml: these are added automatically, at version >= 4:4.5.3 - libqt4-script: this is not used by freemedforms-libs (but by other binaries of src:freemedforms-project) - libqt4-sql-sqlite, libqt4-sql-mysql: these two are valid dependencies, but does freemedforms-project really needs both of two at the same time? would not it be better to have then as alternatives, i.e. libqt4-sql-sqlite | libqt4-sql-mysql ? Thanks, -- Pino -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#682101: Bug#662041: FTBFS on several architectures
Hi, (sending to the right bug now) Alle sabato 3 marzo 2012, Kurt Roeckx ha scritto: > On Sat, Mar 03, 2012 at 09:12:13PM +0100, Michael Biebl wrote: > > Source: elfutils > > Version: 0.153-1 > > Severity: serious > > > > The current version of elfutils FTBFS on several architectures: > > hurd-i386 > > Yes, nothing new. It should never have built on hurd, but it > seems some people think it's a good idea to run buildds that > don't run the testsuite. Sorry if it gives you a bad impression, although OTOH there's a number of packages (of which few important) that requires elfutils libraries. > I filed a bug for the problem in hurd's upstream bug tracker ages > ago. Do you mean http://savannah.gnu.org/bugs/?32770 ? Also please note that elfutils is full of linuxisms, such as: - it assumes that /proc/$pid/maps exist (libdwfl/linux-proc-maps.c) - it looks for Linux kernels and modules (libdwfl/linux-kernel-modules.c) there's no bad thing in doing them... as long as they are done only on Linux. Should elfutils be assumed as "Linux" stuff (so other OSes must emulate all the things it assumes to exist), or could it be fixed to check for Linux stuff only on Linux? -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#683354: apt unusable on hurd
tag 683354 + patch
thanks
Hi,
Alle martedì 31 luglio 2012, Gabriele Giacone ha scritto:
> Upgrade to 0.9.7.3 then run apt-get update:
> [...]
> Reading package lists... Done
> E: The package cache file is corrupted
>
> You won't be able to install/remove any package.
> Downgrading apt and libapt-pkg4.12 to 0.9.7.2 fixes that.
This happens in 0.9.7.3 because this version was fixed to properly use
the fallback code (i.e. not using mremap()) in any occasion on non-Linux
OSes. This means that in MMap::Map() mmap() is not used, but SyncToFd +
buffer is; when synchronizing this (in the two MMap::Sync()), the write
of the buffer to SyncToFd or the msync() call is done inside
preprocessor blocks "#ifdef _POSIX_SYNCHRONIZED_IO"... and that symbol
is not defined by glibc on Hurd, as msync() is not implemented (and
returns ENOSYS). OTOH, you don't need the POSIX synchronized I/O to just
write to the file, so the _POSIX_SYNCHRONIZED_IO blocks can be reduced
to the actual msync() calls.
Attached there is a patch implementing what said above; it makes apt
working again on Hurd.
Thanks,
--
Pino Toscano
--- a/apt-pkg/contrib/mmap.cc
+++ b/apt-pkg/contrib/mmap.cc
@@ -160,7 +160,6 @@
if ((Flags & UnMapped) == UnMapped)
return true;
-#ifdef _POSIX_SYNCHRONIZED_IO
if ((Flags & ReadOnly) != ReadOnly)
{
if (SyncToFd != NULL)
@@ -170,11 +169,12 @@
}
else
{
+#ifdef _POSIX_SYNCHRONIZED_IO
if (msync((char *)Base, iSize, MS_SYNC) < 0)
return _error->Errno("msync", _("Unable to synchronize mmap"));
+#endif
}
}
-#endif
return true;
}
/*}}}*/
@@ -186,7 +186,6 @@
if ((Flags & UnMapped) == UnMapped)
return true;
-#ifdef _POSIX_SYNCHRONIZED_IO
unsigned long long PSize = sysconf(_SC_PAGESIZE);
if ((Flags & ReadOnly) != ReadOnly)
{
@@ -198,11 +197,12 @@
}
else
{
+#ifdef _POSIX_SYNCHRONIZED_IO
if (msync((char *)Base+(unsigned long long)(Start/PSize)*PSize,Stop - Start,MS_SYNC) < 0)
return _error->Errno("msync", _("Unable to synchronize mmap"));
+#endif
}
}
-#endif
return true;
}
/*}}}*/
signature.asc
Description: This is a digitally signed message part.
Bug#683856: kajongg: Unable to start local game
tag 683856 + pending thanks Hi, the problem has been discovered few weeks ago and fixed already in the packaging VCS with backports of a couple of commits from upstream. Thanks for your report, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#683354: Bug#684038: /usr/bin/apt-get: apt-get remove says " package cache file is corrupted"
Alle lunedì 6 agosto 2012, David Kalnischkies ha scritto: > On Mon, Aug 6, 2012 at 6:45 PM, Nigel Horne wrote: > > Could you give me the URL, please, where I download the file > > manually? > > Huh. hurd-i386 is unfortunately not a release architecture (= no > testing), so the files are gone from the archive. For any architecture which is on the main archive or in ports, you can use http://snapshot.debian.org/ to get old sources and binaries. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#684414: unblock (pre-approval): libspectre/0.2.7-1
Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: freeze-exception
Hi,
yesterday a new bugfix version of libspectre, 0.2.7, has been released.
This version quite small in the actual changes it brings, while fixing
a couple of issues.
The full source diff between 0.2.6 and 0.2.7 is:
ChangeLog| 333 +--
Makefile.in | 104 -
NEWS | 10
aclocal.m4 | 2501 --
config.guess | 263 +-
config.sub | 204 +-
configure| 3339 ++-
configure.ac |4
depcomp | 238 ++
install-sh | 544 +++--
libspectre/Makefile.am | 12
libspectre/Makefile.in | 201 +-
libspectre/ps.c | 12
libspectre/spectre-version.h |6
ltmain.sh| 4051 ---
missing | 145 -
test/Makefile.am | 19
test/Makefile.in | 123 -
18 files changed, 7840 insertions(+), 4269 deletions(-)
but most of the changes are autoconf/automake/libtool-related stuff;
stripping those and stripping the ChangeLog (output of `git log`) and
NEWS files, what is left is:
configure.ac |4 ++--
libspectre/Makefile.am | 12 ++--
libspectre/ps.c | 12 +++-
libspectre/spectre-version.h |6 +++---
test/Makefile.am | 19 +--
5 files changed, 31 insertions(+), 22 deletions(-)
which is also attached.
The changes are the following:
- configure.ac, libspectre/spectre-version.h:
bump the version to 0.2.7
- libspectre/Makefile.am, test/Makefile.am:
fix the automake build system to handle own CFLAGS properly and not
ignore user-specified ones (like the Debian ones)
- libspectre/ps.c
as written in NEWS,
"fixes two crashes than can happen when %%Pages: or
%%Page: comments are missing in the document or when %%Pages: comment
is present but it's invalid."
On the Debian side, other than the usual "new upstream release",
I would need to just add "export V=1" in debian/rules to have the build
system output the compilation commands used.
I think it should be mostly harmless overall for wheezy, hence
I'm asking for a pre-approval (and then unblock) before uploading it
to unstable.
Thanks,
--
Pino
--- libspectre-0.2.6/configure.ac
+++ libspectre-0.2.7/configure.ac
@@ -2,7 +2,7 @@
m4_define([spectre_major_version], [0])
m4_define([spectre_minor_version], [2])
-m4_define([spectre_micro_version], [6])
+m4_define([spectre_micro_version], [7])
m4_define([spectre_version],
[spectre_major_version.spectre_minor_version.spectre_micro_version])
@@ -25,7 +25,7 @@
## increment any time the source changes; set to
## 0 if you increment CURRENT
-LT_REVISION=6
+LT_REVISION=7
## increment if any interfaces have been added; set to 0
## if any interfaces have been changed or removed. removal has
--- libspectre-0.2.6/libspectre/Makefile.am
+++ libspectre-0.2.7/libspectre/Makefile.am
@@ -1,9 +1,3 @@
-INCLUDES = \
- -I$(top_builddir) \
- -I$(srcdir) \
- -I$(top_srcdir) \
- $(SPECTRE_CFLAGS)
-
EXTRA_DIST=spectre-version.h.in
lib_LTLIBRARIES = libspectre.la
@@ -47,5 +41,11 @@
spectre-utils.c \
$(gv_sources)
+libspectre_la_CFLAGS = \
+ -I$(top_builddir) \
+ -I$(srcdir) \
+ -I$(top_srcdir) \
+ $(SPECTRE_CFLAGS)
+
libspectre_la_LIBADD = $(LIB_GS)
libspectre_la_LDFLAGS = -version-info @VERSION_INFO@ -no-undefined
--- libspectre-0.2.6/libspectre/ps.c
+++ libspectre-0.2.7/libspectre/ps.c
@@ -597,6 +597,8 @@
if (maxpages > 0) {
doc->pages = (struct page *) PS_calloc(maxpages,
sizeof(struct page));
+if (!doc->pages)
+maxpages = 0;
CHECK_MALLOCED(doc->pages);
}
}
@@ -1002,14 +1004,14 @@
section_len += line_len;
}
+if (maxpages == 0) {
+ maxpages = 1;
+ doc->pages = (struct page *) PS_calloc(maxpages, sizeof(struct page));
+ CHECK_MALLOCED(doc->pages);
+}
newpage:
while (DSCcomment(line) && iscomment(line+2, "Page:")) {
- if (maxpages == 0) {
- maxpages = 1;
- doc->pages = (struct page *) PS_calloc(maxpages, sizeof(struct page));
-CHECK_MALLOCED(doc->pages);
- }
label = ps_gettext(line+length("%%Page:"), &next_char);
if (sscanf(next_char, "%u", &thispage) != 1) thispage = 0;
if (nextpage == 1) {
--- libspectre-0.2.6/libspectre/spectre-version.h
+++ libspectre-0.2.7/libspectre/spectre-version.h
@@ -27,11 +27,11 @@
#define SPECTRE_MAJOR_VERSION 0
#define SPECTRE_MINOR_VERSION 2
-#define SPECTRE_MICRO_VERSION 6
+#define SPECTRE_MICRO_VERSION 7
-#define SPECTRE_VERSION_STRING "0.2.6"
+#define SPECTRE_VERSION_STRING "0.2.7"
-#defi
Bug#704283: Hurd: fix calculation of elapsed time
Package: time Version: 1.7-24 Severity: normal Tags: patch User: debian-h...@lists.debian.org Usertags: hurd Hi, On Hurd, due to the fact that configure detects a wait3 that does not fill in rusage, the non-wait3 code path in resuse.c is used. One of the first things in !HAVE_WAIT3 code is getting or defining a HZ define; neither sys/times.h nor sys/param.h provide HZ, CLOCKS_PER_SEC, or CLK_TCK, leading to HZ being defined as 60, which is not correct. On the other hand, including time.h is enough to provide CLOCKS_PER_SEC, which would then be used for HZ. Attached patch to make configure look for time.h, and use it if available. (The "wait3 does not fill in rusage" is another issue though, although fixing a fallback code path in time should not harm.) Thanks, -- Pino Subject: Search for time.h, and use it in !HAVE_WAIT3 case. This helps getting a correct define for HZ when wait3 is detected to not be properly filling in rusage. Author: Pino Toscano --- a/configure.in +++ b/configure.in @@ -11,7 +11,7 @@ AC_PROG_INSTALL dnl Checks for header files. AC_HEADER_STDC AC_HEADER_SYS_WAIT -AC_CHECK_HEADERS(unistd.h string.h sys/rusage.h sys/resource.h) +AC_CHECK_HEADERS(unistd.h string.h sys/rusage.h sys/resource.h time.h) dnl Checks for typedefs, structures, and compiler characteristics. AC_C_CONST --- a/resuse.c +++ b/resuse.c @@ -25,6 +25,9 @@ #if !HAVE_WAIT3 # include +# if HAVE_TIME_H +# include +# endif # ifndef HZ # include # endif
Bug#704283: Hurd: fix calculation of elapsed time
Alle lunedì 1 aprile 2013, Bob Proulx ha scritto: > I have been looking at making a new package upload which due to the > freeze would be made into experimental. I will address this in the > next package upload. Then after release of Wheezy there will be a > new upload to Sid. Sounds good. I know that Debian is frozen so this should wait before getting into unstable, so I still filed this patch to avoid forgetting to send it later. > There wouldn't happen to be a Debian Hurd machine available for > testing by a non-DD such as myself? Perhaps a server of yours? It > would help facilitate these types of things if I had access to one > hands on. Or perhaps a VM image that I could run on my system? We have two porterboxes, exodar and strauss, but (as Hurd is not a release architecture) the usual DSA access for non-DDs does not apply for them). On the other hand, we can grant access on request, so just email ad...@exodar.debian.net and we'll give you access to exodar. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#704598: libc0.1-dev: sys/mount.h requires C99
Package: libc0.1-dev
Version: 2.13-38
Severity: normal
Hi,
currently, using sys/mount.h provided on kFreeBSD by glibc requires
compiling it with C99.
Basically, compiling a very tiny test case like:
vvv
#include
int main(){return 0;}
^^
$ gcc -D_BSD_SOURCE -std=c90 -o mount mount.c
In file included from mount.c:1:0:
/usr/include/i386-kfreebsd-gnu/sys/mount.h:389:15: error: expected '=', ',',
';', 'asm' or '__attribute__' before 'int'
while it succeeds when using -std=c99 (or equivalents).
--
Pino
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#704598: libc0.1-dev: sys/mount.h requires C99
Alle mercoledì 3 aprile 2013, Steven Chamberlain ha scritto: > > $ gcc -D_BSD_SOURCE -std=c90 -o mount mount.c > > Do any packages actually do this? Compile with -std=c90 or -ansi or > -std and use this header? This breaks the sys/mount.h detection done by kdelibs, which forces c90 to stay compatible with older/broken C compilers. In any case, whatever is the reason, a libc header should compile with any C version requested by the application/library. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#704998: hurd: Broken upgrade due to disappeared shared libraries
Hi, Alle lunedì 8 aprile 2013, Guillem Jover ha scritto: > A dist-upgrade broke the system as the hurd package got upgraded > first, which made the shared libraries disappear, and when > hurd-libs0.3 was being unpacked tar failed to execute. > > Any “Essential: yes” package needs to Pre-Depend at least on their > shared library dependencies to be able to be reliably unpacked (an > any other dependencies required for it to run at unpack time), the > same for pseudo-essential packages (like hurd-libs0.3 itself). It > seems this problem has always been present but only got visible with > the libs split. There's something I miss: hurd already pre-depends (other than the usual shlib dependency) on hurd-libs0.3, which is where the shared libraries were moved, so hurd-libs0.3 should have been at least unpacked before hurd. Also, considering the files in hurd-libs0.3 are in a multiarch libdir, unlike the old location in hurd, so there is nothing which should have prevented unpacking/configuring hurd-libs0.3 with hurd < 20130226-2. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#704998: hurd: Broken upgrade due to disappeared shared libraries
tag 704998 + pending thanks Alle venerdì 12 aprile 2013, Guillem Jover ha scritto: > On Fri, 2013-04-12 at 12:02:54 +0200, Pino Toscano wrote: > > Alle lunedì 8 aprile 2013, Guillem Jover ha scritto: > > > A dist-upgrade broke the system as the hurd package got upgraded > > > first, which made the shared libraries disappear, and when > > > hurd-libs0.3 was being unpacked tar failed to execute. > > > > > > Any “Essential: yes” package needs to Pre-Depend at least on > > > their shared library dependencies to be able to be reliably > > > unpacked (an any other dependencies required for it to run at > > > unpack time), the same for pseudo-essential packages (like > > > hurd-libs0.3 itself). It seems this problem has always been > > > present but only got visible with the libs split. > > > > There's something I miss: hurd already pre-depends (other than the > > usual shlib dependency) on hurd-libs0.3, which is where the shared > > libraries were moved, so hurd-libs0.3 should have been at least > > unpacked before hurd. > > Heh, you made me doubt of myself for a moment: > > $ dpkg-deb -I hurd_20130226-2_hurd-i386.deb |grep Depends > Depends: hurd-libs0.3 (= 20130226-2), sysv-rc, netdde (>= > 0.0.20120518~), libc0.3 (>= 2.13-38~), libncursesw5 (>= > 5.6+20070908), libtinfo5, libx11-6, xkb-data WTF, I was pretty sure to have added it, that's why I even assumed it was there... Thanks for the notice, I just fixed in git. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#704598: libc0.1-dev: sys/mount.h requires C99
Alle giovedì 4 aprile 2013, Dmitrijs Ledkovs ha scritto: > On 3 April 2013 21:39, Pino Toscano wrote: > > Alle mercoledì 3 aprile 2013, Steven Chamberlain ha scritto: > >> > $ gcc -D_BSD_SOURCE -std=c90 -o mount mount.c > >> > >> Do any packages actually do this? Compile with -std=c90 or -ansi > >> or -std and use this header? > > > > This breaks the sys/mount.h detection done by kdelibs, which forces > > c90 to stay compatible with older/broken C compilers. > > > > In any case, whatever is the reason, a libc header should compile > > with any C version requested by the application/library. > > While I understand why kdelibs is doing that, I see no point in > Debian to stay compatible with 23 year old compilers. And I see no point in breaking compatibility with FreeBSD for really no reason, especially when the GNU/kFreeBSD port makes uses of FreeBSD parts for low-level stuff. (And this issue has nothing to do with "23 year old compilers", but it's about the C standard used.) We have and still are uploading KDE 4.10.x (currently .2) pieces to experimental, and so far this issue has caused build issues in kde4libs, libkcompactdisc, kde-workspace. KDE compiles fine on FreeBSD and it did with previous versions of sys/mount.h on GNU/kFreeBSD, so I expect it to continue to do so. GNU/kFreeBSD porters, can you please take a look at this? Thanks, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#705406: kdepim-groupware: Please provide kdepim-groupware & kdepim-wizard functionality in kdepim >= 4.10.2
Hi, Alle domenica 14 aprile 2013, Johannes Graumann ha scritto: > Please make the functionality of kdepim-groupware and kdepim-wizard > available for kdepim >= 4.10.2. These two were dropped upstream, so I'm not sure what we should bring back. > This is particularly interesting as > starting with that kdepim version the kolab3 protocol used by the > recently released kolab3 is supported. If your actual interest is the kolab stuff, those are available as akonadi resources in kdepim-runtime, which we cannot enable yet due to missing kolab libraries in Debian. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#705567: calligrawords: Cannot create encrypted documents
Hi, Alle martedì 16 aprile 2013, Richard Smith ha scritto: > 1) Error message produced by Words: "Could not save > /home/richard/Documents/Blah.odt Reason: Could not create the file > for saving" > 2) Check terminal output: words(3984)/koffice (lib > store) KoEncryptionChecker::isEncryptionSupported: QCA is enabled > but sha1, pbkdf2(sha1) or blowfish-cfb are not supported. Encryption > is disabled. Do you have libqca2-plugin-ossl installed? If not, does installing it allow the saving of encrypted files? -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694551: unblock (pre-approval): soprano/2.7.6+dfsg.1-2
Alle martedì 27 novembre 2012, Adam D. Barratt ha scritto: > On Tue, 2012-11-27 at 17:39 +0100, Pino Toscano wrote: > > I would like to upload soprano 2.7.6+dfsg.1-2 with the following > > two changes: > > - add breaks/replaces in libsoprano-doc for the move of soprano.tag > > > > in an older version > > > > - fix epoch in the version of the libqt4-dev build dependency > > They both look >= important to me, so don't really need > pre-approval. :-) I just wanted to be sure :) > Please go ahead; thanks. Thanks, uploaded. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694553: unblock (pre-approval): kwallet/4.8.4-3
Alle martedì 27 novembre 2012, Adam D. Barratt ha scritto: > On Tue, 2012-11-27 at 18:16 +0100, Pino Toscano wrote: > > I would like to upload kwallet 4.8.4-3. > > The only change is the backport of an upstream patch to fix the > > slow startup of the kwalletmanager application. Its startup gets > > slower and slower wrt the number of passwords stored in the > > wallets. > > Please go ahead; thanks. Thanks, uploaded. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694600: unblock (pre-approval): kgamma/4:4.8.4-2
Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: unblock
Hi,
I would like to upload kgamma 4.8.4-2.
The only change is the backport of an upstream patch to fix the crash
of this configuration module when being run as different user than the
one currently logged in X.
Attached the current diff out of the packaging repo of the changes
above.
Thanks,
--
Pino
diff --git a/debian/changelog b/debian/changelog
index c8f3e08..8af9522 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,13 @@
+kgamma (4:4.8.4-2) UNRELEASED; urgency=low
+
+ [ Pino Toscano ]
+ * Backport upstream commit 1918f145261a52a3b2a917c6784c2cebecbb4064 to fix
+crash when run as different user (e.g. root); patch
+upstream_Fix-crash-when-XVidExtWrap-is-not-available.patch.
+(Closes: #688767)
+
+ -- Debian Qt/KDE Maintainers Wed, 28 Nov 2012 10:42:59 +0100
+
kgamma (4:4.8.4-1) unstable; urgency=low
* Team upload.
diff --git a/debian/patches/series b/debian/patches/series
new file mode 100644
index 000..b4726e5
--- /dev/null
+++ b/debian/patches/series
@@ -0,0 +1 @@
+upstream_Fix-crash-when-XVidExtWrap-is-not-available.patch
diff --git a/debian/patches/upstream_Fix-crash-when-XVidExtWrap-is-not-available.patch b/debian/patches/upstream_Fix-crash-when-XVidExtWrap-is-not-available.patch
new file mode 100644
index 000..abe8479
--- /dev/null
+++ b/debian/patches/upstream_Fix-crash-when-XVidExtWrap-is-not-available.patch
@@ -0,0 +1,50 @@
+From 1918f145261a52a3b2a917c6784c2cebecbb4064 Mon Sep 17 00:00:00 2001
+From: Christoph Feck
+Date: Sun, 11 Nov 2012 02:19:01 +0100
+Subject: [PATCH] Fix crash when XVidExtWrap is not available
+
+This can happen when run as root
+BUG: 292335
+FIXED-IN: 4.10
+---
+ kcmkgamma/kgamma.cpp | 10 +-
+ 1 file changed, 5 insertions(+), 5 deletions(-)
+
+diff --git a/kcmkgamma/kgamma.cpp b/kcmkgamma/kgamma.cpp
+index ffe29ea..890ba99 100644
+--- a/kcmkgamma/kgamma.cpp
b/kcmkgamma/kgamma.cpp
+@@ -70,7 +70,7 @@ KGamma::KGamma(QWidget* parent_P, const QVariantList &)
+ :KCModule(KGammaConfigFactory::componentData(), parent_P), rootProcess(0)
+ {
+ bool ok;
+- GammaCorrection = true;
++ GammaCorrection = false;
+ xv = new XVidExtWrap(&ok, NULL);
+ if (ok) { /* KDE 4: Uneccessary test, when all KCM wrappers do conditional loading */
+ xv->getGamma(XVidExtWrap::Red, &ok);
+@@ -94,6 +94,7 @@ KGamma::KGamma(QWidget* parent_P, const QVariantList &)
+ xv->setScreen(currentScreen);
+
+ rootProcess = new KProcess;
++ GammaCorrection = true;
+ setupUI();
+ saved = false;
+
+@@ -107,10 +108,9 @@ KGamma::KGamma(QWidget* parent_P, const QVariantList &)
+ }
+ load();
+ }
+-else { //something is wrong, show only error message
+- GammaCorrection = false;
+- setupUI();
+-}
++ }
++ if (!GammaCorrection) { //something is wrong, show only error message
++setupUI();
+ }
+ }
+
+--
+1.7.10.4
+
Bug#694617: unblock (pre-approval): konsole/4:4.8.4-2
Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: unblock
Hi,
I would like to upload konsole 4:4.8.4-2 with the following two
changes:
- konsole-dbg recommends kdelibs5-dbg
- backport an upstream patch to simplify/fix the search of executables;
upstream recommended us to provide it, he fixed that while working
on another bug
Attached the current diff out of the packaging repo of the changes
above.
Thanks,
--
Pino
diff --git a/debian/changelog b/debian/changelog
index 5afb9a2..bf56122 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,15 @@
+konsole (4:4.8.4-2) UNRELEASED; urgency=low
+
+ [ Lisandro Damián Nicanor Pérez Meyer ]
+ * konsole-dbg now recommends kdelibs5-dbg.
+
+ [ Pino Toscano ]
+ * Backport upstream commit e48cee6cca742a5bdb3daf6fa0b7c76073217faa to avoid
+looking up executables when an absolute path is specified; patch
+upstream_No-need-to-lookup-executable-PATH-when-absolute-path.patch.
+
+ -- Debian Qt/KDE Maintainers Tue, 23 Oct 2012 18:29:05 -0300
+
konsole (4:4.8.4-1) unstable; urgency=low
* New upstream release.
diff --git a/debian/control b/debian/control
index 755aaf2..2daa69c 100644
--- a/debian/control
+++ b/debian/control
@@ -34,6 +34,7 @@ Priority: extra
Depends: ${misc:Depends}, konsole (= ${binary:Version})
Breaks: kdebase-dbg (<< 4:4.6.80)
Replaces: kdebase-dbg (<< 4:4.6.80)
+Recommends: kdelibs5-dbg
Description: debugging symbols for the KDE X terminal emulator
This package contains debugging files used to investigate problems with
binaries included in the KDE X terminal emulator.
diff --git a/debian/patches/series b/debian/patches/series
index 93acb4d..5047970 100644
--- a/debian/patches/series
+++ b/debian/patches/series
@@ -1,2 +1,3 @@
one_profile_per_window.diff
debian-T-addition.diff
+upstream_No-need-to-lookup-executable-PATH-when-absolute-path.patch
diff --git a/debian/patches/upstream_No-need-to-lookup-executable-PATH-when-absolute-path.patch b/debian/patches/upstream_No-need-to-lookup-executable-PATH-when-absolute-path.patch
new file mode 100644
index 000..4d2d53c
--- /dev/null
+++ b/debian/patches/upstream_No-need-to-lookup-executable-PATH-when-absolute-path.patch
@@ -0,0 +1,30 @@
+From e48cee6cca742a5bdb3daf6fa0b7c76073217faa Mon Sep 17 00:00:00 2001
+From: Jekyll Wu
+Date: Thu, 1 Nov 2012 11:57:42 +0800
+Subject: [PATCH] No need to lookup executable PATH when absolute path is
+ given (cherry picked from commit
+ d12aead588ec09359633eb9de5f9ac2b2a7e1e06)
+
+---
+ src/Session.cpp |5 +
+ 1 file changed, 5 insertions(+)
+
+diff --git a/src/Session.cpp b/src/Session.cpp
+index 732858d..cf325b0 100644
+--- a/src/Session.cpp
b/src/Session.cpp
+@@ -369,6 +369,11 @@ QString Session::checkProgram(const QString& program)
+ if (exec.isEmpty())
+ return QString();
+
++QFileInfo info(exec);
++if (info.isAbsolute() && info.exists() && info.isExecutable()) {
++return exec;
++}
++
+ exec = KRun::binaryName(exec, false);
+ exec = KShell::tildeExpand(exec);
+ QString pexec = KStandardDirs::findExe(exec);
+--
+1.7.10.4
+
Bug#694600: unblock (pre-approval): kgamma/4:4.8.4-2
Alle mercoledì 28 novembre 2012, Niels Thykier ha scritto: > On 2012-11-28 10:54, Pino Toscano wrote: > > Package: release.debian.org > > Severity: normal > > User: release.debian@packages.debian.org > > Usertags: unblock > > > > Hi, > > > > I would like to upload kgamma 4.8.4-2. > > The only change is the backport of an upstream patch to fix the > > crash of this configuration module when being run as different > > user than the one currently logged in X. > > > > Attached the current diff out of the packaging repo of the changes > > above. > > > > Thanks, > > Hi, > > Please go ahead. Let us know when it has been rebuilt on all > (relevant) architectures and has some time in unstable. Thanks, it was uploaded some hours ago and it build almost everywhere. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694578: unblock (pre-approval): okular/4.8.4-3
Hi, Alle mercoledì 28 novembre 2012, Niels Thykier ha scritto: > On 2012-11-28 01:04, Lisandro Damián Nicanor Pérez Meyer wrote: > > Package: release.debian.org > > Severity: normal > > User: release.debian@packages.debian.org > > Usertags: unblock > > > > Please (pre-approve) unblock package okular > > einclude/attach the debdiff against the package in testing) > > > > This packages solves: > > > > - #680946, severity important, avoiding different versions of > > libokularcore and okular at the same time. The first 3 entries in > > the changelog are part of this fix. > > > > Three commits backported from upstream to solve: > > > > - the extraction of the page bounding box, used by the trimming > > function; - restore word detection in DVI documents; > > - avoid a crash with broken fonts. > > > > Diffstat: > > changelog > > | 22 + control > > |3 > > patches/avoid_crash_with_broken_fonts.diff > >| 31 ++ patches/fix_word_detection_with_dvi_docs.diff > >| 153 ++ patches/series > >|3 > > patches/upstream_Call-Generator-signalPixmapRequestDone-_after_-s > > avin.patch | 40 ++ 6 files changed, 251 insertions(+), 1 > > deletion(-) > > > > Kinds regards, Lisandro. > > > > unblock okular/4.8.4-3 > > > > [...] > > Hi, > > Please go ahead. Let us know when it has been rebuilt on all > (relevant) architectures and has some time in unstable. Thanks, it has been uploaded. Unfortunately, I recompiled it using a wrong pbuilder chroot (which has poppler/experimental) -- can you please binNMU okular/amd64? Sorry for the trouble, and thanks again, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#688113: kde-runtime: Some headers files missing e.g. simpleresource.h
Hi, Alle mercoledì 19 settembre 2012, ares ha scritto: > Some headers files are missing. Bug is identical to ubungu bug 928009 > - see > https://bugs.launchpad.net/ubuntu/+source/kde-runtime/+bug/928009 > where are all information described. > > They have already fixed their packages so maybe you could get > inspiration (they've added kde-runtime-dev package). The fix is wrong, simply because kde-runtime is supposed to be a "runtime only" module, not providing any development libraries to be used by 3rd parties. This has been a known problem upstream. In KDE 4.9, kde-runtime will not contain development libraries anymore (the only ones, nepomuk-related, are moved to the new nepmuk-core module). -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#671729: Lokalize uses its own broken version of msgcat
Hi David, Alle giovedì 29 novembre 2012, David Prévot ha scritto: > Le 29/11/2012 06:16, Pino Toscano a écrit : > > Alle domenica 6 maggio 2012, David Prévot ha scritto: > >> I've reported a similar bug upstream [0] some time ago, but > >> despite its current status, it hasn't really been fixed or > >> resolved. > >> > >>0: https://bugs.kde.org/show_bug.cgi?id=256035 > > > > It should have been fixed in some RC of KDE 4.8, and indeed I > > cannot reproduce your issue with the lokalize currently in testing > > (4.8.4). > > Right, the PO file is not as broken anymore, but the no-wrap entries > are still wrapped, and the one-line around 80 characters are not > handled the same way as msgcat does, so it still “produce[s] an > unnecessary and unreadable huge diff on VCS” (aka: upstream bug [0] > is still unfixed). It would seem to me this is a different problem though (i.e. "not honours no-wrap entries" vs "produces broken .po files"). I see no mention of support for "no-wrap" in entries, so could you please open a new bug/wish upstream for that (since I see you have already an account on KDE's bugzilla)? Thanks, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694743: unblock (pre-approval): kgpg/4:4.8.4-4
Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: unblock
Hi,
I would like to upload kgpg 4:4.8.4-4.
The only change is the addition of the "gnupg | gnupg2" dependency;
so far it went unnoticed because apt depends on gnupg (so kgpg would
work in basically almost any Debian installation), but having the
explicit dependency would be a good idea anyway.
Attached the current diff out of the packaging repo of the changes
above.
Thanks,
--
Pino
diff --git a/debian/changelog b/debian/changelog
index a8ab8f2..bd02e1e 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,10 @@
+kgpg (4:4.8.4-4) UNRELEASED; urgency=low
+
+ [ Lisandro Damián Nicanor Pérez Meyer ]
+ * Depend on gnupg or gnupg2 (Closes: #686632).
+
+ -- Debian Qt/KDE Maintainers Tue, 04 Sep 2012 20:40:47 -0300
+
kgpg (4:4.8.4-3) unstable; urgency=low
* Team upload.
diff --git a/debian/control b/debian/control
index 14adc66..74c34f7 100644
--- a/debian/control
+++ b/debian/control
@@ -19,7 +19,7 @@ Vcs-Git: git://git.debian.org/pkg-kde/kde-sc/kgpg.git
Package: kgpg
Section: utils
Architecture: any
-Depends: ${shlibs:Depends}, ${misc:Depends}
+Depends: gnupg | gnupg2, ${shlibs:Depends}, ${misc:Depends}
Description: graphical front end for GNU Privacy Guard
Kgpg manages cryptographic keys for the GNU Privacy Guard, and can encrypt,
decrypt, sign, and verify files. It features a simple editor for applying
Bug#694743: unblock (pre-approval): kgpg/4:4.8.4-4
Hi, Alle giovedì 29 novembre 2012, Adam D. Barratt ha scritto: > On Thu, 2012-11-29 at 20:18 +0100, Pino Toscano wrote: > > I would like to upload kgpg 4:4.8.4-4. > > The only change is the addition of the "gnupg | gnupg2" dependency; > > so far it went unnoticed because apt depends on gnupg (so kgpg > > would work in basically almost any Debian installation), but > > having the explicit dependency would be a good idea anyway. > > Please go ahead; thanks. Thanks; uploaded and built almost everywhere. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694670: kscd does not play any CD
Hi,
Alle venerdì 30 novembre 2012, Michael Stapelberg ha scritto:
> Rainer Dorsch wrote:
> > whatever CD I try to play with kscd player, it does not play
> > anything. Since vlc can play all track, audex can rip them, I think
> > it is a problem with kscd. kscd can still read out the track list.
> > Since I am not aware that I have any special setup, there is a
> > fair chance that the bug affects everybody and makes the package
> > completely unusable. If that is not the case, please downgrade the
> > bug.
>
> I can reproduce this: when just clicking on the play button in kscd,
> it immediately segfaults.
Well, this seems a different issue ("does nothing" vs "click and
crashes") than the reported one...
> After setting the environment variable KDE_DEBUG=1 and ulimit -c
> unlimited, I get a core dump:
>
> gdb $ thread apply all bt
> [...]
> Hopefully that helps to further debug this problem.
Unfortunately it not not useful, since kscd is a "unique instance"
application which forks right after the start, so the trace of the
parent process is not useful.
However, you can get a better one (and try to debug kscd) just by
running kscd as
$ kscd --nofork
also, installing kdemultimedia-dbg will help providing a more useful
stack trace.
Also, you could please paste the output of
$ dpkg -l | grep phonon
?
Thanks,
--
Pino Toscano
signature.asc
Description: This is a digitally signed message part.
Bug#694670: kscd does not play any CD
Hi Michael and Rainer, Alle venerdì 30 novembre 2012, Michael Stapelberg ha scritto: > > However, you can get a better one (and try to debug kscd) just by > > running kscd as > > $ kscd --nofork > > also, installing kdemultimedia-dbg will help providing a more > > useful stack trace. > > Thanks for those hints. Here is the new backtrace: > [...] Thanks, much better. Looking upstream, I found https://bugs.kde.org/show_bug.cgi?id=265597 which now leads me to the question: do you both have udisks installed? If not, does installing it and rerunning kscd changes the situation? Thanks, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694789: poppler-utils: PDF print over evince breaks the complete text
Hi, Alle venerdì 30 novembre 2012, Samuel Wolf ha scritto: > print _some_ pdf files over evince breaks the complete text, see > screenshot. Please provide a (sample) document showing the issue. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694670: kscd does not play any CD
clone 694670 -1 retitle -1 crash with no optical drive forwarded 694670 https://bugs.kde.org/show_bug.cgi?id=293766 thanks Hi, Alle venerdì 30 novembre 2012, Michael Stapelberg ha scritto: > I think we are in fact dealing with two separate issues here: That's what I was saying, yes :-P > 1. The segfault I reported happens when there is no optical drive >recognized on the system. My ThinkPad X200 does not have an > optical drive, but there is one in the docking station. > Unfortunately, it only works correctly when the notebook is booted > with the docking station attached. > >That is, when submitting the backtrace, I had not booted the > notebook in the docking station, so that the optical drive was not > detected. Simplifying this, I suppose the same crash will happen > when you disconnect the OD in your computer on purpose, or rmmod the > driver. > >This issue can be put into a separate bugreport. Should I file a >separate issue or do you want to open it? I will just clone this bug, so backlog etc is preserved. Let's continue that issue on the new bug. > 2. After rebooting my notebook in the docking station and inserting > an audio CD into the optical drive, kscd does not crash anymore when > I press the play button. > >However, neither does it actually play the CD. In the terminal in >which I started it, I get: > > QDBusConnection: session D-Bus connection created before > QCoreApplication. Application may misbehave. Setting new source > New source: QUrl( "" ) > Transitioning to state "playing" > State change > Moving from "null" 0 to "ready" 1 > 1 > Transitioning to state "null" > >And that’s it. The application itself shows the title of the first >track scrolling by but remains at 00:00 and does not play > anything, even after many minutes. I attached a screenshot. > >Using mplayer cdda://0 -cdrom-device /dev/sr0 works fine. > >I noticed that /dev/cdrom was not present on my system, so I used >ln -s /dev/sr0 /dev/cdrom and that made kscd work. Its output in > the terminal is now: > > QDBusConnection: session D-Bus connection created before > QCoreApplication. Application may misbehave. Setting new source > New source: QUrl( "" ) > Transitioning to state "playing" > State change > Moving from "null" 0 to "ready" 1 > Duration message > Duration message > State change > Stream changed to cdda:// > Moving from "ready" 1 to "paused" 4 > State change > Moving from "paused" 4 to "playing" 2 > Transitioning to state "paused" > State change > Moving from "playing" 2 to "paused" 4 Interesting; possibly it won't change much, but what if (without the symlink you just created) you install phonon-backend-vlc, make it the default in `kcmshell4 phonon` and try again? Also, you might want to enable the two debug areas for kscd in `kdebugdialog`, and rerun kscd. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694670: kscd does not play any CD
Hi, Alle venerdì 30 novembre 2012, Michael Stapelberg ha scritto: > Pino Toscano writes: > > Interesting; possibly it won't change much, but what if (without > > the symlink you just created) you install phonon-backend-vlc, make > > it the default in `kcmshell4 phonon` and try again? > > I did that, and it leads to the same result (music not playing). Hm ok, thanks for the test anyway. > I have noticed that I have a lot of CD-ROM drives in > /etc/udev/rules.d/70-persistent-cd.rules due to various hardware > being connected to my computer over time (such as UMTS modems which > bring their driver by emulating a CD-ROM drive). The first one is > /dev/cdrom, the second one cdrom1 and so on. Currently, the symlink > which gets created is /dev/cdrom5. > > I am not sure whether kscd or some other component of the kde > multimedia stack is supposed to look for /dev/cdrom* instead of > /dev/cdrom or what is the best way of figuring out where to look. > > After googling a bit, I stumbled upon this bug report at RedHat, > which concludes that 70-persistent-cd.rules should just be deleted > by the sysadmin: > https://bugzilla.redhat.com/show_bug.cgi?id=570561 I don't have experience with udev and its rules, so I cannot comment about this, sorry. > Tackling this issue from a different angle: What happens when I have > two optical drives? How will kscd chose which one it uses? Will it > always use /dev/cdrom? According to the code, it basically always chooses the first, and the possibility to switch drive has never been implemented... In any case, kscd basically queries the Solid library which in the end asks udisks about the supported drives; you can find out what is detected with $ solid-hardware query "IS OpticalDrive" and then you can get the details of the found devices with $ solid-hardware details /org/... -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#672825: qt4-x11: Various small problems building qt4-x11 on hppa
Alle venerdì 30 novembre 2012, John David Anglin ha scritto: > On 11/29/2012 6:30 AM, Pino Toscano wrote: > > The upload of qt4-x11 4:4.8.2+dfsg-3 now build-conflicts with > > libqtwebkit-dev, so a qt4 build cannot be accidentally affected by > > an already built qtwebkit. Due to the way qt is build and qtwebkit > > is handled within qt4, this is the "best" solution at the moment > > (see also #689265). Thus, I'm closing this bug, since all the > > issues look to me solved. Do not hesitate to reopen if they are > > not. Thank your for your report, > > Confirmed, removing libqtwebkit-dev fixes build. Thanks for the confirmation! -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#690847: poppler.
retitle 690847 patch for RTL issues thanks Hi Alex, Alle venerdì 30 novembre 2012, alex bodnaru ha scritto: > i've fixed and improved some issues related to rtl support. other on > the way. > right, although upstream like the patches, they will be applied to > their unstable tip. Which bugs does this patch fix? Can you please point me to upstream bugs? > untill that tip will be stable, i'll submit them on debian too. It is way too late to introduce new dependencies now for Wheezy; after that, there will be plently of time for updating Poppler, so having them applied to Poppler 0.22 or 0.24 should have a good chance to have to be part of Jessie (the stable release after Wheezy). -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#690847: poppler.
Hi, Alle venerdì 30 novembre 2012, alex bodnaru ha scritto: > of course patches would debute on unstable. > the moment the patch on upstream will maturize, i'll submit it here > too. Sorry, but unstable is not a dump for random stuff. Unstable currently has Poppler 0.18.x, which won't see any more development. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694830: unblock: kde-workspace/4:4.8.4-5
Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: unblock
Hi,
we would like to upload kde-workspace 4:4.8.4-5, which so far would
provide a number of updates/fixes:
- a minor changelog fix to the previous version
- fix the X session registration, which was off due to a mismatch
between an old debian option and the actual configuration key
(#692730)
- the backport of an upstream fix for the taskbar widget, which would
not unnecessarly notify windows as "demanding attention" (#685334)
- the backport of an upstream patch to improve a patch backported
already in 4:4.8.4-3
- the backport of 3 upstream patches to fix the reading of two
configuration files (kwinrc and powermanagementprofilesrc) as
cascading configuration (i.e. so values in configs in global paths
like /u/s/kde4/config and the Debian-specific /e/kde4, for the local
admin, can be used)
- the backport of an upstream fix for the saving of panel launchers of
preferred apps (#686131)
- drop hal from kdm.init (#694021)
- kde-workspace-bin:
- ship a /etc/kde4 directory, so postinst can work (#694253)
- explicitly depend on qdbus, used by startkde
- encoding fixes for the debconf translations of es (#692209) and pl
(#691953)
Generally, all the proposed changes should be safe; in case, the upload
could be aged to more days to give it more days.
Attached the current debdiff.
Thanks,
--
Pino
diff -Nru kde-workspace-4.8.4/debian/changelog kde-workspace-4.8.4/debian/changelog
--- kde-workspace-4.8.4/debian/changelog 2012-10-21 11:07:18.0 -0300
+++ kde-workspace-4.8.4/debian/changelog 2012-11-30 20:04:40.0 -0300
@@ -1,6 +1,44 @@
+kde-workspace (4:4.8.4-5) UNRELEASED; urgency=low
+
+ [ Lisandro Damián Nicanor Pérez Meyer ]
+ * Remove "New upstream release" from the previous changelog entry
+(Closes: #691302). Thanks Filipus Klutiero.
+ * sessreg was merged in kdm (Closes: #692730):
+- Drop the use-sessreg option in kdm.options and its man page, as it is
+ no longer needed.
+- Do not set UseSessReg to false in debian/patches/kdmrc_defaults.diff.
+ The bug it solved has dissapeared during the merge.
+Thanks Wolfgang Schweer and Petter Reinholdtsen for digging into this.
+
+ [ Pino Toscano ]
+ * Backport upstream commit 89161c40e064958aa192b2202670316671d7e377 to fix
+the "demand" status of windows in the taskbar widget; patch
+upstream_Check-if-attention-demanding-status-has-changed-when.patch.
+(Closes: #685334)
+ * Backport upstream commits d0343319fcfc249a38c79171be727d7133984eeb and
+53f6eca921a44a1f5e567b92d45ea84599afea74 to make powermanagementprofilesrc
+read as cascading config, and 9994e178b790b03a464c335e624366f82f0da643 for
+kwinrc.
+ * Backport upstream commit 5b35757fc3613a18e2d784af53d1d23ebe295028 to
+improve the sorting of init .js files (followup of the changes backported
+in 4:4.8.4-3 with upstream_make-sure-scripts-are-executed-sorted.patch);
+patch upstream_Make-sure-the-plasma-desktop-scripts-are-sorted-in-t.patch.
+ * Backport upstream commit 70cd86a5eef103137e64f2e3868a5bd2d58d71d7 to fix
+the deletion of launchers of preferred applications; patch
+upstream_Fix-deletion-of-preferred-application-launchers.patch.
+(Closes: #686131)
+ * Drop hal from kdm.init. (Closes: #694021)
+ * Make kde-workspace-bin ship a /etc/kde4 directory. (Closes: #694253)
+ * Make kde-workspace-bin explicitly depend on qdbus, used by startkde.
+
+ [ Debconf translation updates ]
+ * Fix encoding of Polish (thanks to David Prévot). (Closes: #691953)
+ * Fix encoding of Spanish (thanks to Camaleón). (Closes: #692209)
+
+ -- Debian Qt/KDE Maintainers Wed, 24 Oct 2012 17:06:28 -0300
+
kde-workspace (4:4.8.4-4) unstable; urgency=low
- * New upstream release.
* The package (re)building switches to XZ compression. (Closes: #688761)
[ Pino Toscano ]
diff -Nru kde-workspace-4.8.4/debian/control kde-workspace-4.8.4/debian/control
--- kde-workspace-4.8.4/debian/control 2012-10-21 11:06:24.0 -0300
+++ kde-workspace-4.8.4/debian/control 2012-11-30 20:04:40.0 -0300
@@ -109,7 +109,7 @@
Depends: ${shlibs:Depends}, ${misc:Depends}, iso-codes,
plasma-desktop (= ${binary:Version}) | plasma-netbook (= ${binary:Version}),
kde-workspace-data (= ${source:Version}), x11-utils, x11-xserver-utils,
- kde-workspace-kgreet-plugins (= ${binary:Version})
+ kde-workspace-kgreet-plugins (= ${binary:Version}), qdbus
Recommends: plasma-scriptengines, polkit-kde-1 (>= 0.99) | policykit-1-gnome,
upower [linux-any]
Suggests: x11-xkb-utils
diff -Nru kde-workspace-4.8.4/debian/kde-workspace-bin.dirs kde-workspace-4.8.4/debian/kde-workspace-bin.dirs
--- kde-workspace-4.8.4/debian/kde-workspace-bin.dirs 1969-12-31 21:00:00.0 -0300
+++ kde-workspace-4.8.4/debian/kde-workspace-bin.dirs 2012-11-30 20:04:40.0 -0300
@@ -0,0 +1 @@
+etc/kde4
dif
Bug#694830: unblock: kde-workspace/4:4.8.4-5
retitle 694830 unblock (pre-approval): kde-workspace/4:4.8.4-5 thanks Alle sabato 1 dicembre 2012, Pino Toscano ha scritto: > we would like to upload kde-workspace 4:4.8.4-5, which so far would > provide a number of updates/fixes: Forgot to properly modify the subject, as this is a pre-approval request. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694773: qtwebkit: FTBFS: sh4: Memory exhausted by dh_strip
Hi, Alle venerdì 30 novembre 2012, Nobuhiro Iwamatsu ha scritto: > qtwebkit FTBFS on Renesas sh4. > [...] > This problem is same as debug option problem for mips, arm and other. > I created a patch which reivse this problem. The patch seems fine, although it is post-Wheezy material. Thanks, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694789: poppler-utils: PDF print over evince breaks the complete text
reassign 694789 libcairo2 cairo/1.12.2-2 tag 694789 + fixed-upstream fixed 694789 cairo/1.12.6-1 thanks Hi Samuel, Alle venerdì 30 novembre 2012, Samuel Wolf ha scritto: > Here is a example pdf: > http://www.stoeber.de/TDM/HUKE/441698_4.pdf Thanks for the testcase. It is a cairo bug, most probably similar (if not the same) to #679105 and #692828. Furthermore, this issue has been fixed in (at least, maybe even earlier) cairo 1.12.6, hence marking appropriately. Thanks for your report, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#695012: kde-workspace: Suspend not working after upgrade squeeze->wheezy
Hi, Alle lunedì 3 dicembre 2012, Johannes Ranke ha scritto: > After a fresh installation of Debian wheezy on my ACER TravelMate > 8371 laptop, suspend/resume is not working any more. > > Maybe it has to do with my package selection, as I used the list of > installed packages in squeeze for selecting packages in wheezy via > dpkg --get-selection and dpkg --set-selection. Do you have upower installed? If so, what does $ upower --dump say? -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694830: unblock: kde-workspace/4:4.8.4-5
Alle lunedì 3 dicembre 2012, Julien Cristau ha scritto: > On Sat, Dec 1, 2012 at 00:28:05 +0100, Pino Toscano wrote: > > we would like to upload kde-workspace 4:4.8.4-5, which so far would > > provide a number of updates/fixes: > > - a minor changelog fix to the previous version > > - fix the X session registration, which was off due to a mismatch > > between an old debian option and the actual configuration key > > (#692730) > > - the backport of an upstream fix for the taskbar widget, which would > > not unnecessarly notify windows as "demanding attention" > > (#685334) > > - the backport of an upstream patch to improve a patch backported > > already in 4:4.8.4-3 > > - the backport of 3 upstream patches to fix the reading of two > > configuration files (kwinrc and powermanagementprofilesrc) as > > cascading configuration (i.e. so values in configs in global > > paths like /u/s/kde4/config and the Debian-specific /e/kde4, for > > the local admin, can be used) > > - the backport of an upstream fix for the saving of panel launchers of > > preferred apps (#686131) > > - drop hal from kdm.init (#694021) > > - kde-workspace-bin: > > - ship a /etc/kde4 directory, so postinst can work (#694253) > > - explicitly depend on qdbus, used by startkde > > - encoding fixes for the debconf translations of es (#692209) and pl > > (#691953) > > > > Generally, all the proposed changes should be safe; in case, the > > upload could be aged to more days to give it more days. > > Ack to all except the hal removal in kdm.init, this is still needed > on !linux. Hm, but hal does not ship an init.d script in current wheezy though, so how could that dependency be satisfied even with hal installed? -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694830: unblock: kde-workspace/4:4.8.4-5
Alle lunedì 3 dicembre 2012, Julien Cristau ha scritto: > On Mon, Dec 3, 2012 at 21:59:08 +0100, Pino Toscano wrote: > > Alle lunedì 3 dicembre 2012, Julien Cristau ha scritto: > > > On Sat, Dec 1, 2012 at 00:28:05 +0100, Pino Toscano wrote: > > > > we would like to upload kde-workspace 4:4.8.4-5, which so far > > > > would provide a number of updates/fixes: > > > > - a minor changelog fix to the previous version > > > > - fix the X session registration, which was off due to a mismatch > > > > between an old debian option and the actual configuration key > > > > (#692730) > > > > - the backport of an upstream fix for the taskbar widget, which would > > > > not unnecessarly notify windows as "demanding attention" > > > > (#685334) > > > > - the backport of an upstream patch to improve a patch backported > > > > already in 4:4.8.4-3 > > > > - the backport of 3 upstream patches to fix the reading of two > > > > configuration files (kwinrc and powermanagementprofilesrc) as > > > > cascading configuration (i.e. so values in configs in global > > > > paths like /u/s/kde4/config and the Debian-specific /e/kde4, > > > > for the local admin, can be used) > > > > - the backport of an upstream fix for the saving of panel launchers of > > > > preferred apps (#686131) > > > > - drop hal from kdm.init (#694021) > > > > - kde-workspace-bin: > > > > - ship a /etc/kde4 directory, so postinst can work (#694253) > > > > - explicitly depend on qdbus, used by startkde > > > > - encoding fixes for the debconf translations of es (#692209) and pl > > > > (#691953) > > > > > > > > Generally, all the proposed changes should be safe; in case, > > > > the upload could be aged to more days to give it more days. > > > > > > Ack to all except the hal removal in kdm.init, this is still > > > needed on !linux. > > > > Hm, but hal does not ship an init.d script in current wheezy > > though, so how could that dependency be satisfied even with hal > > installed? > > Hrm. I seem to remember a conversation with Michael about that, > don't remember what the outcome was. OK, I removed that change for now, let's revise that after Wheezy; the rest will be uploaded soon. Thanks for the review, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694830: unblock: kde-workspace/4:4.8.4-5
Alle lunedì 3 dicembre 2012, Julien Cristau ha scritto: > On Sat, Dec 1, 2012 at 00:28:05 +0100, Pino Toscano wrote: > > we would like to upload kde-workspace 4:4.8.4-5, which so far would > > provide a number of updates/fixes: > > - a minor changelog fix to the previous version > > - fix the X session registration, which was off due to a mismatch > > between an old debian option and the actual configuration key > > (#692730) > > - the backport of an upstream fix for the taskbar widget, which would > > not unnecessarly notify windows as "demanding attention" > > (#685334) > > - the backport of an upstream patch to improve a patch backported > > already in 4:4.8.4-3 > > - the backport of 3 upstream patches to fix the reading of two > > configuration files (kwinrc and powermanagementprofilesrc) as > > cascading configuration (i.e. so values in configs in global > > paths like /u/s/kde4/config and the Debian-specific /e/kde4, for > > the local admin, can be used) > > - the backport of an upstream fix for the saving of panel launchers of > > preferred apps (#686131) > > - drop hal from kdm.init (#694021) > > - kde-workspace-bin: > > - ship a /etc/kde4 directory, so postinst can work (#694253) > > - explicitly depend on qdbus, used by startkde > > - encoding fixes for the debconf translations of es (#692209) and pl > > (#691953) > > > > Generally, all the proposed changes should be safe; in case, the > > upload could be aged to more days to give it more days. > > Ack to all except the hal removal in kdm.init, this is still needed > on !linux. With the above change, uploaded and built basically everywhere. Thanks, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#686677: closed by Thorsten Alteholz (Bug#686677: fixed in freemedforms-project 0.8.0-2)
Hi, Alle martedì 4 dicembre 2012, Debian Bug Tracking System ha scritto: >[ Eric Maeker ] >* debian/control: change dependency of freemedforms-libs > (Closes: #686677) libqt4-sql-sqlite and libqt4-sql-mysql are both > needed Note that now control specifies: | libqt4-sql (>= 4:4.7.0), | libqt4-sql-sqlite (>= 4:4.7.0) which is not exactly what described in changelog... (I will not reopen the bug, since this is not an actual issue, but you might want to fix this in your next upload, whenever you are planning to do it.) -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#695160: unblock (pre-approval): akonadi/1.7.2-2
Package: release.debian.org Severity: normal User: release.debian@packages.debian.org Usertags: unblock Hi, I would like to upload akonadi 1.7.2-2. The only change is a fix in the provided README.Debian, to actually match the configuration format. Attached the current diff out of the packaging repo of the changes above. Thanks, -- Pino diff --git a/debian/README.Debian b/debian/README.Debian index fc6d9fc..774b7c5 100644 --- a/debian/README.Debian +++ b/debian/README.Debian @@ -47,7 +47,7 @@ At the moment, the following backends are supported: This backend uses official QSql MySQL driver. The following configuration options are available: - [General] + [%General] Driver=QMYSQL [QMYSQL] @@ -71,7 +71,7 @@ At the moment, the following backends are supported: This backend uses QSql PSQL driver. The following configuration options are available: - [General] + [%General] Driver=QPSQL [QPSQL] @@ -94,7 +94,7 @@ At the moment, the following backends are supported: also shipped in the backend package. The following configuration options are available: - [General] + [%General] Driver=QSQLITE3 [QSQLITE3] diff --git a/debian/changelog b/debian/changelog index a60860c..78c7441 100644 --- a/debian/changelog +++ b/debian/changelog @@ -1,3 +1,11 @@ +akonadi (1.7.2-2) UNRELEASED; urgency=low + + [ Lisandro Damián Nicanor Pérez Meyer ] + * Change [General] to [%General] in README.Debian (Closes: #688236). +Thanks Alexander Wuerstlein for noticing. + + -- Debian Qt/KDE Maintainers Thu, 20 Sep 2012 14:58:24 -0300 + akonadi (1.7.2-1) unstable; urgency=low * New upstream release.
Bug#695160: unblock (pre-approval): akonadi/1.7.2-2
Alle mercoledì 5 dicembre 2012, Julien Cristau ha scritto: > On Tue, Dec 4, 2012 at 20:09:38 +0100, Pino Toscano wrote: > > Package: release.debian.org > > Severity: normal > > User: release.debian@packages.debian.org > > Usertags: unblock > > > > Hi, > > > > I would like to upload akonadi 1.7.2-2. > > The only change is a fix in the provided README.Debian, to actually > > match the configuration format. > > You don't really need pre-approval for trivial changes... OK, uploaded and built everywhere. Thanks, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#694617: unblock (pre-approval): konsole/4:4.8.4-2
Alle venerdì 7 dicembre 2012, Niels Thykier ha scritto: > On 2012-11-28 13:31, Pino Toscano wrote: > > Package: release.debian.org > > Severity: normal > > User: release.debian@packages.debian.org > > Usertags: unblock > > > > Hi, > > > > I would like to upload konsole 4:4.8.4-2 with the following two > > changes: > > - konsole-dbg recommends kdelibs5-dbg > > - backport an upstream patch to simplify/fix the search of > > executables; > > > > upstream recommended us to provide it, he fixed that while > > working on another bug > > > > Attached the current diff out of the packaging repo of the changes > > above. > > > > Thanks, > > -- Pino > > Hi, > > Please go ahead and let us know when it has been rebuilt on the > relevant architectures. Thank you; it has been uploaded a couple of hours ago, and built fine everywhere. Thanks, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#695573: razorqt: allow to compile on !linux archs
Source: razorqt Version: 0.4.1-1~exp1 Severity: normal Tags: patch Hi, currently razorqt cannot be built on non-Linux architectures due to the unconditional libudev-dev build dependency. Udev is Linux-specific, so attached there is a patch to restrict that B-D on Linux archs only. Thanks, -- Pino --- a/debian/control +++ b/debian/control @@ -11,7 +11,7 @@ Build-Depends: debhelper (>= 9~), libpolkit-agent-1-dev, libpolkit-qt-1-dev, libqt4-dev (>= 4.5.1), - libudev-dev, + libudev-dev [linux-any], libxrender-dev, libxcomposite-dev, libxdamage-dev,
Bug#695573: razorqt: allow to compile on !linux archs
Hi, Alle lunedì 10 dicembre 2012, Manuel A. Fernandez Montecelo ha scritto: > 2012/12/10 Pino Toscano : > > currently razorqt cannot be built on non-Linux architectures due to > > the unconditional libudev-dev build dependency. Udev is > > Linux-specific, so attached there is a patch to restrict that B-D > > on Linux archs only. > > Thanks for the report. I do not know if it will even compile fine > without this dependency, but I will try to incorporate it in new > releases. The version reported compiled fine for me on Hurd (so I could suppose on kFreeBSD too). > In the few months that it took for the package to get into Debian, a > new major upstream version was released, so the packaging will need > to be revisited. Ok, thank you. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#695631: ninja-build: FTBFS on !linux archs
Package: ninja-build
Version: 1.0.0-1
Severity: important
Tags: patch
Hi,
ninja-build does not build on non-Linux archs, such as
GNU/kFreeBSD[1][2] and GNU/Hurd[3].
The problem is that the GetProcessorCount() implementation for these
architectures is the sysconf() one, but has not been
included, causing sysconf() and _SC_NPROCESSORS_ONLN to not be
declared. Another solution (which is the one I chose) is to make use
of the "linux" implementation which uses get_nprocs(), which is a GNU
extension and thus available for anything using GNU libc.
If this solution is not deemed wanted, the other solution is to just add
#ifndef _WIN32
#include
#endif
among the other includes in src/util.cc.
[1]
https://buildd.debian.org/status/fetch.php?pkg=ninja-build&arch=kfreebsd-i386&ver=1.0.0-1&stamp=1355174935
[2]
https://buildd.debian.org/status/fetch.php?pkg=ninja-build&arch=kfreebsd-amd64&ver=1.0.0-1&stamp=1355175958
[3]
https://buildd.debian.org/status/fetch.php?pkg=ninja-build&arch=hurd-i386&ver=1.0.0-1&stamp=1355180120
Thanks,
--
Pino
--- a/src/util.cc
+++ b/src/util.cc
@@ -39,7 +39,7 @@
#elif defined(__SVR4) && defined(__sun)
#include
#include
-#elif defined(linux)
+#elif defined(linux) || defined(__GLIBC__)
#include
#endif
@@ -295,7 +295,7 @@ string StripAnsiEscapeCodes(const string
return stripped;
}
-#if defined(linux)
+#if defined(linux) || defined(__GLIBC__)
int GetProcessorCount() {
return get_nprocs();
}
Bug#695573: razorqt: allow to compile on !linux archs
Alle lunedì 10 dicembre 2012, Manuel A. Fernandez Montecelo ha scritto: > 2012/12/10 Pino Toscano : > > The version reported compiled fine for me on Hurd (so I could > > suppose on kFreeBSD too). > > Other than compiling, did you have the opportunity to test it and see > if it runs fine? On Hurd, it seems to start correctly although nothing appears in the panel (might be very well an Hurd-specific issue). I'd think it should work fine even in kFreeBSD. Looking at the release notes of more recent versions, it seems libudev is no more used directly, so if you package >= 0.5 you could just drop that B-D. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#704669: kdelibs-bin: unable to install kdelibs-bin=4:4.10.2-1 due to issues with kdelibs4c2a
Hi, Alle giovedì 18 aprile 2013, robert...@libero.it ha scritto: > you should change the breaks/replaces of kde4.10.2 in order to break > kdelibs4c2a << 4:3.5.10.dfsg.1-6~ so that a version >> -6~ won't > conflict with the current kde. There is no kdelibs4c2a in the Debian archive anymore, so there is no way kdelibs4c2a can be fixed in Debian. Limiting the breaks to a specific version which does not exists in Debian gives zero guarantees that such versions will actually fix the issue. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#704669: kdelibs-bin: unable to install kdelibs-bin=4:4.10.2-1 due to issues with kdelibs4c2a
Hi, Alle giovedì 18 aprile 2013, robert...@libero.it ha scritto: > kdelibs4c2a is actually in debian squeeze: > http://packages.debian.org/squeeze/kdelibs4c2a It does not in wheezy, and KDE 4.10 at the moment is only in experimental. > The only conflict is the presence of ktelnetservice and kmailservice > files that is also present in kdelibs5-plugins. So a custom version > of kdelibs4c2a definitely solves the issue. One file or 100 files doesn't matter: there is a file conflict. > Again, this is the way the bug has been solved in ubuntu: > https://bugs.launchpad.net/ubuntu/+source/kde4libs/+bug/1100622 It would be helpful if you could read what I said previously: > > Limiting the breaks to > > a specific version which does not exists in Debian gives zero > > guarantees that such versions will actually fix the issue. Sorry no, I will not lift the breaks based on a version that may (or may not) exist somewhere else out of Debian, and which we cannot guarantee it will actually fix the issue. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#705708: gmic: FTBFS on hurd-i386
Source: gmic Version: 1.5.1.6+dfsg-4 Severity: important Tags: patch User: debian-h...@lists.debian.org Usertags: hurd Hi, currently[1], gmic fails to compile on GNU/Hurd. The failure happens because in the (home-brew...) Makefile the code path used if OS=GNU is the Windows configuration. The easy fix is doing what has also been done for GNU/kFreeBSD, i.e. turn OS=GNU into Linux. (Side note: patch kfreebsd-support seems obsolete.) [1] https://buildd.debian.org/status/fetch.php?pkg=gmic&arch=hurd-i386&ver=1.5.1.6%2Bdfsg-4&stamp=1341423041 Thanks, -- Pino --- a/src/Makefile +++ b/src/Makefile @@ -56,6 +56,9 @@ CC = g++ ifeq ($(OS),GNU/kFreeBSD) OS = Linux endif +ifeq ($(OS),GNU) +OS = Linux +endif ifeq ($(OS),Darwin) PLUGINDIR = ~/Library/Application\ Support/Gimp/plug-ins USR = /opt/local
Bug#706092: [kdepim-runtime] Akonadi Kolab resources missing since 4.10.x
forcemerge 705406 706092 thanks Hi, Alle mercoledì 24 aprile 2013, Schrober ha scritto: > It seems the Kolab support was accidentally removed during the 4.10.x > upgrade. It was not "accidentally removed", but the support for kolab was migrated into akonadi resources which use libkolab, which is not available in Debian. See the rest of discussion in #705406 (to which this bug is merged) for more details. P.S.: yes, we do read build logs of our packages already. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#706170: xosview: platform detection improvement/fix in debian/rules
Source: xosview Version: 1.14-1 Severity: wishlist Tags: patch Hi, the changes in the experimental version still make xosview compiled as "linux" even in non-Linux OSes, and never uses the right platform string on kFreeBSD. The attached patch factorizes the detection of the platform string to use when invoking make, - making it use the right one on kFreeBSD (which though won't compile yet, and that will need proper fixes) - bailing out when the OS is not recognized, and it is easy (just add the two proper lines in that if chain, e.g. starting from the kfreebsd ones) to add cases for new OSes Thanks, -- Pino --- a/debian/rules +++ b/debian/rules @@ -9,23 +9,20 @@ CXX=g++ DEB_BUILD_ARCH_OS ?=$(shell dpkg-architecture -qDEB_BUILD_ARCH_OS) +ifeq ($(DEB_BUILD_ARCH_OS),linux) + PLATFORM=linux +else ifeq ($(DEB_BUILD_ARCH_OS),kfreebsd) + PLATFORM=bsd +else + $(error Missing implementation for $(DEB_BUILD_ARCH_OS)) +endif build: build-arch build-indep build-arch: build-stamp build-indep: build-stamp build-stamp: dh_testdir - -#experimental kfreebsd-* support -ifeq ($(DEB_BUILD_ARCH_OS),kfreebsd-i386) - $(MAKE) PLATFORM=bsd -else -ifeq ($(DEB_BUILD_ARCH_OS),kfreebsd-amd64) - $(MAKE) PLATFORM=bsd -endif -endif - - $(MAKE) PLATFORM=linux + $(MAKE) PLATFORM=$(PLATFORM) touch $@ clean:
Bug#706170: xosview: platform detection improvement/fix in debian/rules
Alle venerdì 26 aprile 2013, Kartik Mistry ha scritto: > On Fri, Apr 26, 2013 at 2:05 AM, Pino Toscano wrote: > > the changes in the experimental version still make xosview compiled > > as "linux" even in non-Linux OSes, and never uses the right > > platform string on kFreeBSD. > > The attached patch factorizes the detection of the platform string > > to use when invoking make, > > - making it use the right one on kFreeBSD (which though won't > > compile > > > > yet, and that will need proper fixes) > > > > - bailing out when the OS is not recognized, and it is easy (just > > add > > > > the two proper lines in that if chain, e.g. starting from the > > kfreebsd ones) to add cases for new OSes > > Thanks a lot. I was about to commit my fix in git repo (and also > realized that kfreebsd build is not yet ready!), but your patch is > way much better. Committed to git repo. I'll disable kfreebsd-* > build and upload in a day or two. Thanks, but why leaving it commented and removing the kfreebsd build? After all, xosview on kfreebsd has not been compiled before, so having it fail is not a regression wrt previous versions (thus it won't stop xosview migrating to testing, when uploaded to unstable). Just leave the platform selection as it is and readd kfreebsd-any as architecture for xosview, so kfreebsd porters can notice the build failure and possibly provide patches for it. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#706170: xosview: platform detection improvement/fix in debian/rules
Alle venerdì 26 aprile 2013, Kartik Mistry ha scritto: > On Fri, Apr 26, 2013 at 1:13 PM, Pino Toscano wrote: > > Thanks, but why leaving it commented and removing the kfreebsd > > build? After all, xosview on kfreebsd has not been compiled > > before, so having it fail is not a regression wrt previous > > versions (thus it won't stop xosview migrating to testing, when > > uploaded to unstable). > > Just leave the platform selection as it is and readd kfreebsd-any > > as architecture for xosview, so kfreebsd porters can notice the > > build failure and possibly provide patches for it. > > Great. Done. Thanks. Just one niptick more: you can replace «kfreebsd-amd64 kfreebsd-i386» with just «kfreebsd-any» in both the libkvm-dev build dependency and in the Architecture control field. There's nothing potentially specific to amd64 or i386, so (once fixed) it will work fine with any other kfreebsd arch. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#706630: kdevelop: Please update kdevelop to 4.5.0
notfound 706630 4:4.4.1-0ubuntu1 thanks Hi, Alle giovedì 2 maggio 2013, Logan Rosen ha scritto: > kdevelop needs to be updated to 4.5.0 Yes, we know kdevelop 1.5 has been released, and until Wheezy is not released this is an a low priority. Also, please note two things: > Version: 4:4.4.1-0ubuntu1 a) we don't have this version in Debian > -- System Information: > Debian Release: wheezy/sid > APT prefers saucy-updates > APT policy: (500, 'saucy-updates'), (500, 'saucy-security'), (500, > 'saucy'), (100, 'saucy-backports') > Architecture: amd64 (x86_64) > Foreign Architectures: i386 b) please report bugs and wishes to Debian using a Debian system, not an Ubuntu one Thanks, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#706865: libpthread-stubs: FTBFS on hurd-i386: libpthread-stubs library no more compiled
Source: libpthread-stubs Version: 0.3-3 Severity: important Tags: patch User: debian-h...@lists.debian.org Usertags: hurd Hi, given that now Hurd's libpthread is compiled as glibc addon with proper forwards, the libpthread-stubs library is no more compiled and thus the build fails due to missing files referenced by .install.hurd files. Attached there is a patch to remove all the custom hurd-i386 handling in .install files (removing the .install.hurd files) and rules. Thanks, -- Pino --- a/debian/libpthread-stubs0-udeb.install.hurd +++ /dev/null @@ -1 +0,0 @@ -lib/i386-gnu lib --- a/debian/libpthread-stubs0.install.hurd +++ /dev/null @@ -1 +0,0 @@ -lib --- a/debian/rules +++ b/debian/rules @@ -1,21 +1,10 @@ #!/usr/bin/make -f -DEB_HOST_ARCH_OS ?= $(shell dpkg-architecture -qDEB_HOST_ARCH_OS) -DEB_HOST_MULTIARCH ?= $(shell dpkg-architecture -qDEB_HOST_MULTIARCH) - %: dh $@ --with autoreconf --builddirectory=build override_dh_auto_configure: debian/copyright -ifeq ($(DEB_HOST_ARCH_OS), hurd) - # libc.so pulls in libpthread-stubs.so, defeating our configure checks - # by making them think the stubs are already in libc. Point ld at an - # empty libpthread-stubs instead. - echo 'OUTPUT_FORMAT(elf32-i386)' > debian/libpthread-stubs.so - LDFLAGS='-L$(CURDIR)/debian' dh_auto_configure --builddirectory=build -else dh_auto_configure --builddirectory=build -endif override_dh_makeshlibs: dh_makeshlibs -plibpthread-stubs0 --add-udeb=libpthread-stubs0-udeb @@ -23,11 +12,6 @@ override_dh_makeshlibs: override_dh_auto_install: dh_auto_install --builddirectory=build find debian/tmp -name '*.la' -delete -ifeq ($(DEB_HOST_ARCH_OS), hurd) - install -d debian/tmp/lib/$(DEB_HOST_MULTIARCH) - mv debian/tmp/usr/lib/$(DEB_HOST_MULTIARCH)/libpthread-stubs.so.0* debian/tmp/lib/$(DEB_HOST_MULTIARCH) - ln -sf /lib/$(DEB_HOST_MULTIARCH)/libpthread-stubs.so.0 debian/tmp/usr/lib/$(DEB_HOST_MULTIARCH)/libpthread-stubs.so -endif debian/copyright: debian/copyright.debian COPYING cat $+ > $@
Bug#706865: [Xcb] Bug#706865: libpthread-stubs: FTBFS on hurd-i386: libpthread-stubs library no more compiled
Alle domenica 5 maggio 2013, Julien Cristau ha scritto:
> On Sun, May 5, 2013 at 17:24:09 +0200, Pino Toscano wrote:
> > Source: libpthread-stubs
> > Version: 0.3-3
> > Severity: important
> > Tags: patch
> > User: debian-h...@lists.debian.org
> > Usertags: hurd
> >
> > Hi,
> >
> > given that now Hurd's libpthread is compiled as glibc addon with
> > proper forwards, the libpthread-stubs library is no more compiled
> > and thus the build fails due to missing files referenced by
> > .install.hurd files.
> >
> > Attached there is a patch to remove all the custom hurd-i386
> > handling in .install files (removing the .install.hurd files) and
> > rules.
>
> Can we use this opportunity to stop shipping the empty
> libpthread-stubs0, on all archs?
Considering with this patch libpthread-stubs0{,-dev} would be the same
on all the Debian archs, I see nothing wrong with that.
(Semi-related: libpthread-stubs0-udeb could also go, as it is not used
anymore on Hurd.)
--
Pino Toscano
signature.asc
Description: This is a digitally signed message part.
Bug#679892: pdftoipe: Port to libpoppler25 API
Hi, Alle domenica 15 luglio 2012, Pino Toscano ha scritto: > I just reported upstream the issue (it was not reported), proposing a > patch which would switch the call to error() to a simplier fprintf(). > This avoids issues if error() changes again in the future, and it is > compatible with any poppler version. Just a short note: few months ago upstream released pdftoipe-20130124, which should be compilatible with any stable Poppler version so far (like 0.18.x fixed by the patch currently shipped, and the more recent 0.20.x). -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#706971: transition: attica 0.4
Package: release.debian.org Severity: normal User: release.debian@packages.debian.org Usertags: transition Hello, I would like to request a slot for a small libattica 0.4.x transition. The newer attica source is already in experimental, and the transition affects just few (KDE-related) sources. The SONAME change is the following: * libattica-dev (src:attica): libattica0 -> libattica0.4 The sources affected by the transition are: choqok kde4libs kdeplasma-addons kde-runtime smokekde there are also calligra and parley which build-depend on libattica-dev and checking for it at cmake time but not actually using it; should be enough to ignore them. Ben file: title = "attica 0.4"; is_affected = .build-depends ~ /libattica-dev/; is_good = .depends ~ "libattica0.4"; is_bad = .depends ~ "libattica0"; Thanks, -- Pino -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#710699: nmu: xjig_2.4-14
Package: release.debian.org Severity: normal User: release.debian@packages.debian.org Usertags: binnmu nmu xjig_2.4-14 . i386 . -m "Rebuild in a clean environment" It seems the xjig maintainer upload (with an i386 build) was not done in a clean (or up-to-date) environment, and it uses libjpeg62 instead of libjpeg8 (current libjpeg-dev provider). Dave, please make sure to build your packages in clean and up-to-date build environment. Thanks, -- Pino -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#710779: poppler: incorrect rendering: spurious white rectangles
tag 710779 + moreinfo thanks Hi, Alle domenica 2 giugno 2013, Vincent Lefevre ha scritto: > Source: poppler > Version: 0.16.7-2 Reporting a bug against a version that now exist nowhere is not exactly useful... > xpdf and evince now render some PDF files incorrectly. See attached > example; at least the first page is affected. I've also attached two > screenshot parts. > > xpdf-a.png has been obtained with xpdf using: > > /var/cache/apt/archives/libpoppler-glib4_0.12.4-1.2_amd64.deb > /var/cache/apt/archives/libpoppler5_0.12.4-1.2_amd64.deb > /var/cache/apt/archives/poppler-data_0.4.4-1_all.deb > /var/cache/apt/archives/poppler-utils_0.12.4-1.2_amd64.deb > /var/cache/apt/archives/xpdf_3.02-15_amd64.deb > > (status on 2011-07-01). No problems. > > xpdf-b.png has been obtained with xpdf using: > > /var/cache/apt/archives/libpoppler-glib6_0.16.7-2_amd64.deb > /var/cache/apt/archives/libpoppler13_0.16.7-2_amd64.deb > /var/cache/apt/archives/poppler-data_0.4.4-1_all.deb > /var/cache/apt/archives/poppler-utils_0.16.7-2_amd64.deb > /var/cache/apt/archives/xpdf_3.02-17_amd64.deb > > (status on 2011-07-15). Three spurious white rectangles appear. > > The problem is still present with: > * evince 3.4.0-3.1 > * xpdf 3.03-11 / libpoppler19 0.18.4-6 I don't see any spurious white rectangle neither with okular and xpdf (which use the splash renderer) nor with evince (which use the cairo renderer); this using poppler 0.18.4-6 on an up-to-date jessie/testing system. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#711035: override: rsibreak:utils/optional
Package: ftp.debian.org Severity: normal Hi, even if it is a KDE application, it is an non-KDE-specific utility. Thanks, -- Pino -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#711308: override: icecc:devel/optional, libicecc-dev:libdevel/optional
Package: ftp.debian.org Severity: normal icecc is a development tool so it fits more as "devel", more than "kde"; libicecc-dev should also be "libdevel" just like any other -dev package. Also, in PTS and buildd pages I see the icecc source as being in the "kde" section, does that need some manual override change too? Thanks, -- Pino -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#711307: libqt5webkit5-dev: Insufficient dependencies in libqt5webkit5-dev
retitle 711307 libQt5WebKitWidgets.prl filled with private libs thanks Hi, Alle giovedì 6 giugno 2013, Timo Jyrinki ha scritto: > This is an after the fact bug report to (in)validate the recent > dependency changes in the packaging git. > > It seems that the dependencies of libqt5webkit5-dev, or possibly some > of its dependencies, are not sufficient at the moment. The dependencies of libqt5webkit5-dev are perfectly fine, since they cover all the #include's in libqt5webkit5-dev's headers. > I've made a simple test case as follows: > apt-get install bzr-builddeb libqt5webkit5-dev qtdeclarative5-dev > qt5-default apt-get remove libqt5opengl5-dev # only recommended by > qtbase5-dev bzr branch lp:~timo-jyrinki/+junk/testwebkit > cd testwebkit > bzr bd (Since it is really small, having it as standalone tarball attached to this would have been better.) > This fails with: Having the full build log would have been nicer. > /usr/bin/ld: cannot find -lxslt > /usr/bin/ld: cannot find -lgio-2.0 > /usr/bin/ld: cannot find -lgstapp-0.10 > /usr/bin/ld: cannot find -lgstinterfaces-0.10 > /usr/bin/ld: cannot find -lgstpbutils-0.10 > /usr/bin/ld: cannot find -lgstvideo-0.10 > /usr/bin/ld: cannot find -lgstbase-0.10 > /usr/bin/ld: cannot find -lgstreamer-0.10 > /usr/bin/ld: cannot find -lgobject-2.0 > /usr/bin/ld: cannot find -lgmodule-2.0 > /usr/bin/ld: cannot find -lgthread-2.0 > /usr/bin/ld: cannot find -lxml2 > /usr/bin/ld: cannot find -lglib-2.0 > /usr/bin/ld: cannot find -lsqlite3 > /usr/bin/ld: cannot find -lQt5OpenGL Yes, I can see these errors, but... > In addition to the already added dependencies - libqt5opengl5-dev > libgstreamer-plugins-base0.10-dev libxslt1-dev - sqlite3 dependency > seems to be missing. ... no, adding the lot of -dev packages added in Git and the above is not the right solution. Doing a bit of investigation reveals that /usr/lib/*/libQt5WebKitWidgets.prl is filled with these libs (mostly non-qt5, but also some unneeded qt5 ones), so it is an upstream bug that should be reported. In the meanwhile, a (IMHO) better workaround would be pruning these libraries from that .prl file. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#702071: CVE-2013-1788, CVE-2013-1789 and CVE-2013-1790
tag 702071 - moreinfo tag 702071 + confirmed found 702071 poppler/0.18.4-5 thanks Hi, thanks for the tests cases, Salvatore. I've verified the issues, and the situation that I found for current wheezy+sid (= 0.18.4-5) is the following: Alle sabato 2 marzo 2013, Salvatore Bonaccorso ha scritto: > CVE-2013-1788[0]: > invalid memory issues This applies, but not with all the reported documents. > CVE-2013-1789[1]: > crash in broken documents This seems to not apply. > CVE-2013-1790[2]: > uninitialized memory read This applies. I will backport and test the appropriate fixes for this version of poppler, and then upload. Regarding stable, I will do the proper investigation (and eventually backport fixes as needed) once sid is fixed and the fixed version has successfully migrated to wheezy; this way I want to reduce the potential issues. Is that okay for the security team? -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#703423: colord: unbuildable on !linux archs
Source: colord
Version: 0.1.31-1
Severity: serious
Tags: patch
Justification: fails to build from source
Hi,
the new colord 0.1.31-1 cannot be built on non-Linux archs because of
the libsystemd-login-dev B-D (built from systemd, which is
Linux-specific).
Other than that, there are other issues, which my patch fixes in
(I hope) a clean way:
- as said already, limit the libsystemd-login-dev B-D as linux-any
- gusb and systemd-login are Linux-specific, so explicitly enable or
disable them on Linux archs or not
- since the udev directory cannot be detected, on non-Linux archs the
installation path of the udev rules would result as "/rules.d".
To avoid this, I've explicitly set the udevdir to configure (also on
Linux); of course, if this seems notdesiderable, the udevdir can be
passed only on non-Linux archs
- the colorhug stuff depends on gusb, so all the colorhug packages are
now linux-any instead of any
- there seem to be no colord-plugins compiled, so its line in
colord.install would result in a dh_install error; I've "grouped" it
with the colord-sensors line that follows in that .install file,
I hope it is not too generic now
Thanks,
--
Pino
--- a/debian/control
+++ b/debian/control
@@ -21,7 +21,7 @@ Build-Depends:
dh-autoreconf,
gtk-doc-tools,
autopoint,
- libsystemd-login-dev,
+ libsystemd-login-dev [linux-any],
Standards-Version: 3.9.4
XS-Testsuite: autopkgtest
Section: graphics
@@ -101,7 +101,7 @@ Description: GObject introspection data
Package: libcolorhug-dev
Section: libdevel
-Architecture: any
+Architecture: linux-any
Depends:
libcolorhug1 (= ${binary:Version}),
libglib2.0-dev,
@@ -119,7 +119,7 @@ Description: library to access the Color
Package: libcolorhug1
Section: libs
-Architecture: any
+Architecture: linux-any
Depends:
${shlibs:Depends},
${misc:Depends},
@@ -135,7 +135,7 @@ Description: library to access the Color
Package: gir1.2-colorhug-1.0
Section: introspection
-Architecture: any
+Architecture: linux-any
Depends: ${gir:Depends},
${shlibs:Depends},
${misc:Depends}
--- a/debian/rules
+++ b/debian/rules
@@ -21,12 +21,13 @@ confflags = \
--disable-volume-search \
--with-daemon-user=colord \
--with-systemdsystemunitdir=/lib/systemd/system \
+ --with-udevrulesdir=/lib/udev/rules.d \
--enable-vala
ifeq ($(DEB_HOST_ARCH_OS),linux)
- confflags += --enable-gudev --enable-sane
+ confflags += --enable-gudev --enable-sane --enable-gusb --enable-systemd-login
else
- confflags += --disable-gudev --disable-sane
+ confflags += --disable-gudev --disable-sane --disable-gusb --disable-systemd-login
endif
override_dh_auto_configure:
--- a/debian/colord.install
+++ b/debian/colord.install
@@ -2,8 +2,7 @@ debian/tmp/etc/
lib/udev/rules.d/
usr/bin/
usr/lib/colord*
-usr/lib/*/colord-plugins
-usr/lib/*/colord-sensors
+usr/lib/*/colord-*
usr/share/color
usr/share/colord
usr/share/glib-2.0/schemas
Bug#559473: [PATCH] symbols file for fuse
tag 559473 + patch thanks Hi, attached there is a first version of patch for symbols file for libfuse. Since a few years, upstream uses properly versioned symbols, so for most of them (symver) directives can do the job just fine. Regarding the manually added symbols (which don't have any version): - __fuse_* symbols seem to be part of compatibility 22, 21 or 11, if you look at the libfuse 2.5.x headers; thus, I set 2.2 as version for them, which should be more than enough - fuse_main and fuse_new seem very old function (if not just stubs) which are never used, so 2.5 should be more than enough too - libulockmgr seems added in fuse 2.6, hence the version I put Building packages with a libfuse that ships the attached symbols file means some dependencies could be lower than the current shlib dependency (e.g. I tested sshfs, and its libfuse2 dependency changed from >= 2.8.1 to >= 2.7); thus I suggest to apply this only after the Wheezy release, so there's plenty of time to check whether the new dependencies are fine for fuse clients. Of course, if there is any need to manually raise the version for some symbol(s), it can be done easily. -- Pino Toscano --- /dev/null +++ b/debian/libfuse2.symbols @@ -0,0 +1,22 @@ +libfuse.so.2 #PACKAGE# #MINVER# +* Build-Depends-Package: libfuse-dev + (symver)FUSE_2.2 2.2 + (symver)FUSE_2.4 2.4 + (symver)FUSE_2.5 2.5 + (symver)FUSE_2.6 2.6 + (symver)FUSE_2.7 2.7 + (symver)FUSE_2.7.5 2.7.5 + (symver)FUSE_2.8 2.8 + (symver)FUSE_2.9 2.9 + (symver)FUSE_2.9.1 2.9.1 + __fuse_exited@Base 2.2 + __fuse_loop_mt@Base 2.2 + __fuse_process_cmd@Base 2.2 + __fuse_read_cmd@Base 2.2 + __fuse_set_getcontext_func@Base 2.2 + __fuse_setup@Base 2.2 + __fuse_teardown@Base 2.2 + fuse_main@Base 2.5 + fuse_new@Base 2.5 +libulockmgr.so.1 #PACKAGE# #MINVER# + ulockmgr_op@Base 2.6 --- a/debian/rules +++ b/debian/rules @@ -56,7 +56,7 @@ dh_link --remaining-packages override_dh_makeshlibs: - dh_makeshlibs -s -V "libfuse2 (>= 2.8.1)" --add-udeb=libfuse2-udeb + dh_makeshlibs -s --add-udeb=libfuse2-udeb override_dh_strip: dh_strip --dbg-package=fuse-dbg signature.asc Description: This is a digitally signed message part.
Bug#703930: unblock: poppler/0.18.4-6
Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: unblock
Hi,
I just uploaded poppler/0.18.4-6 fixing CVE-2013-1788 and CVE-2013-1790
(#702071). Could you please unblock it?
(Also, I noticed during the upload to have left urgency=low, maybe
is it worth urgency=medium.)
unblock poppler/0.18.4-6
Thanks,
--
Pino
diff -Nru poppler-0.18.4/debian/changelog poppler-0.18.4/debian/changelog
--- poppler-0.18.4/debian/changelog 2013-01-31 15:20:54.0 +0100
+++ poppler-0.18.4/debian/changelog 2013-03-25 21:43:14.0 +0100
@@ -1,3 +1,18 @@
+poppler (0.18.4-6) unstable; urgency=low
+
+ * Backport upstream commits 0388837f01bc467045164f9ddaff787000a8caaa (patch
+upstream_Fix-another-invalid-memory-access-in-1091.pdf.asan.7.patch),
+8b6dc55e530b2f5ede6b9dfb64aafdd1d5836492 (adapted patch
+upstream_Fix-invalid-memory-access-in-1150.pdf.asan.8.69.patch), and
+e14b6e9c13d35c9bd1e0c50906ace8e707816888 (adapted patch
+upstream_Fix-invalid-memory-access-in-2030.pdf.asan.69.463.patch) to fix
+CVE-2013-1788.
+ * Backport upstream commit b1026b5978c385328f2a15a2185c599a563edf91 to fix
+CVE-2013-1790 (patch upstream_Initialize-refLine-totally.patch).
+ * With the changes above, this upload closes: #702071.
+
+ -- Pino Toscano Mon, 25 Mar 2013 21:43:07 +0100
+
poppler (0.18.4-5) unstable; urgency=low
* Correctly initialize PSOutputDev::fontFileNameLen and
diff -Nru poppler-0.18.4/debian/patches/series poppler-0.18.4/debian/patches/series
--- poppler-0.18.4/debian/patches/series 2013-01-31 13:58:17.0 +0100
+++ poppler-0.18.4/debian/patches/series 2013-03-23 07:48:04.0 +0100
@@ -4,3 +4,7 @@
upstream_Change-nn-to-number.patch
upstream_fix-GooString-insert.diff
psoutputdev-initialize-vars.diff
+upstream_Fix-another-invalid-memory-access-in-1091.pdf.asan.7.patch
+upstream_Fix-invalid-memory-access-in-2030.pdf.asan.69.463.patch
+upstream_Fix-invalid-memory-access-in-1150.pdf.asan.8.69.patch
+upstream_Initialize-refLine-totally.patch
diff -Nru poppler-0.18.4/debian/patches/upstream_Fix-another-invalid-memory-access-in-1091.pdf.asan.7.patch poppler-0.18.4/debian/patches/upstream_Fix-another-invalid-memory-access-in-1091.pdf.asan.7.patch
--- poppler-0.18.4/debian/patches/upstream_Fix-another-invalid-memory-access-in-1091.pdf.asan.7.patch 1970-01-01 01:00:00.0 +0100
+++ poppler-0.18.4/debian/patches/upstream_Fix-another-invalid-memory-access-in-1091.pdf.asan.7.patch 2013-03-23 07:48:04.0 +0100
@@ -0,0 +1,40 @@
+From 0388837f01bc467045164f9ddaff787000a8caaa Mon Sep 17 00:00:00 2001
+From: Albert Astals Cid
+Date: Thu, 10 Jan 2013 20:29:06 +0100
+Subject: [PATCH] Fix another invalid memory access in 1091.pdf.asan.72.42
+
+---
+ poppler/Stream.cc | 10 --
+ 1 file changed, 8 insertions(+), 2 deletions(-)
+
+diff --git a/poppler/Stream.cc b/poppler/Stream.cc
+index d118ddd..4cb3326 100644
+--- a/poppler/Stream.cc
b/poppler/Stream.cc
+@@ -2132,7 +2132,8 @@ GBool CCITTFaxStream::isBinary(GBool last) {
+
+ // clip [-256,511] --> [0,255]
+ #define dctClipOffset 256
+-static Guchar dctClip[768];
++#define dctClipLength 768
++static Guchar dctClip[dctClipLength];
+ static int dctClipInit = 0;
+
+ // zig zag decode map
+@@ -3078,7 +3079,12 @@ void DCTStream::transformDataUnit(Gushort *quantTable,
+
+ // convert to 8-bit integers
+ for (i = 0; i < 64; ++i) {
+-dataOut[i] = dctClip[dctClipOffset + 128 + ((dataIn[i] + 8) >> 4)];
++const int ix = dctClipOffset + 128 + ((dataIn[i] + 8) >> 4);
++if (unlikely(ix < 0 || ix >= dctClipLength)) {
++ dataOut[i] = 0;
++} else {
++ dataOut[i] = dctClip[ix];
++}
+ }
+ }
+
+--
+1.7.10.4
+
diff -Nru poppler-0.18.4/debian/patches/upstream_Fix-invalid-memory-access-in-1150.pdf.asan.8.69.patch poppler-0.18.4/debian/patches/upstream_Fix-invalid-memory-access-in-1150.pdf.asan.8.69.patch
--- poppler-0.18.4/debian/patches/upstream_Fix-invalid-memory-access-in-1150.pdf.asan.8.69.patch 1970-01-01 01:00:00.0 +0100
+++ poppler-0.18.4/debian/patches/upstream_Fix-invalid-memory-access-in-1150.pdf.asan.8.69.patch 2013-03-23 07:48:04.0 +0100
@@ -0,0 +1,27 @@
+From 8b6dc55e530b2f5ede6b9dfb64aafdd1d5836492 Mon Sep 17 00:00:00 2001
+From: Albert Astals Cid
+Date: Thu, 10 Jan 2013 22:31:52 +0100
+Subject: [PATCH] Fix invalid memory access in 1150.pdf.asan.8.69
+
+---
+ splash/Splash.cc |5 -
+ 1 file changed, 4 insertions(+), 1 deletion(-)
+
+--- a/splash/Splash.cc
b/splash/Splash.cc
+@@ -1521,11 +1521,14 @@ SplashPath *Splash::makeDashedPath(Splas
+ lineDashStartPhase -= (SplashCoord)i * lineDashTotal;
+ lineDashStartOn = gTrue;
+ lineDashStartIdx = 0;
+- while (lineDashStartPhase >= state->lineDash[lineDashStartIdx]) {
++ while (lineDashStartIdx < state->lineDashLength && lineDashStartPhase >= state->lineDash[lineDashStartIdx]) {
+ lineDashSt
Bug#686147: Fixing up "no akonadi-mysql" installations
Hi, to eventually fix up akonadi startup issues after a dist-upgrade or a fresh install due to missing mysql, please manually install akonadi-backend-mysql Unfortunately there is no way to have that fixed automatically (other than making akonadi hard-depend always on mysql). Sorry for the issues and thanks for using KDE, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#699443: unblock: poppler/0.18.4-5
Package: release.debian.org Severity: normal User: release.debian@packages.debian.org Usertags: unblock Hi, I would like to upload poppler 0.18.4-5, which would initialize two class members of PSOutputDev to avoid crashing in the destructor if the initialization of the output device is not "ok" (e.g. when the requested output file name cannot be opened, like in #699421). This bug, other than hitting that pdftops case, can hit library users of PSOutputDev, and its poppler-glib wrapping (such as epdfview). Attached the proposed debdiff. Thanks, -- Pino diff -Nru poppler-0.18.4/debian/changelog poppler-0.18.4/debian/changelog --- poppler-0.18.4/debian/changelog 2012-11-27 16:24:20.0 +0100 +++ poppler-0.18.4/debian/changelog 2013-01-31 14:03:45.0 +0100 @@ -1,3 +1,12 @@ +poppler (0.18.4-5) UNRELEASED; urgency=low + + [ Pino Toscano ] + * Correctly initialize PSOutputDev::fontFileNameLen and +PSOutputDev::psFileNames; patch psoutputdev-initialize-vars.diff. +(Closes: #699421) + + -- Pino Toscano Thu, 31 Jan 2013 13:19:35 +0100 + poppler (0.18.4-4) unstable; urgency=low * Backport upstream commits 7ba15d11e56175601104d125d5e4a47619c224bf and diff -Nru poppler-0.18.4/debian/patches/psoutputdev-initialize-vars.diff poppler-0.18.4/debian/patches/psoutputdev-initialize-vars.diff --- poppler-0.18.4/debian/patches/psoutputdev-initialize-vars.diff 1970-01-01 01:00:00.0 +0100 +++ poppler-0.18.4/debian/patches/psoutputdev-initialize-vars.diff 2013-01-31 13:58:06.0 +0100 @@ -0,0 +1,41 @@ +Author: Pino Toscano +Description: initialize PSOutputDev::fontFileNameLen and PSOutputDev::psFileNames + Avoid crashing in ~PSOutputDev when the PSOutputDev instance is not "ok". +Applied-Upstream: not-needed +Last-Update: 2013-01-31 +Bug-Debian: http://bugs.debian.org/699421 + +--- a/poppler/PSOutputDev.cc b/poppler/PSOutputDev.cc +@@ -1012,6 +1012,7 @@ PSOutputDev::PSOutputDev(const char *fil + fontIDs = NULL; + fontFileIDs = NULL; + fontFileNames = NULL; ++ fontFileNameLen = 0; + font8Info = NULL; + font16Enc = NULL; + imgIDs = NULL; +@@ -1022,6 +1023,7 @@ PSOutputDev::PSOutputDev(const char *fil + haveTextClip = gFalse; + haveCSPattern = gFalse; + t3String = NULL; ++ psFileNames = NULL; + + forceRasterize = forceRasterizeA; + +@@ -1077,6 +1079,7 @@ PSOutputDev::PSOutputDev(PSOutputFunc ou + fontIDs = NULL; + fontFileIDs = NULL; + fontFileNames = NULL; ++ fontFileNameLen = 0; + font8Info = NULL; + font16Enc = NULL; + imgIDs = NULL; +@@ -1087,6 +1090,7 @@ PSOutputDev::PSOutputDev(PSOutputFunc ou + haveTextClip = gFalse; + haveCSPattern = gFalse; + t3String = NULL; ++ psFileNames = NULL; + + forceRasterize = forceRasterizeA; + diff -Nru poppler-0.18.4/debian/patches/series poppler-0.18.4/debian/patches/series --- poppler-0.18.4/debian/patches/series 2012-11-27 16:20:00.0 +0100 +++ poppler-0.18.4/debian/patches/series 2013-01-31 13:58:17.0 +0100 @@ -3,3 +3,4 @@ upstream_pdfinfo-decode-utf-16-surrogate-pairs.patch upstream_Change-nn-to-number.patch upstream_fix-GooString-insert.diff +psoutputdev-initialize-vars.diff
Bug#699662: kpat: no menu entry for kpat in Games sub-menu
retitle 699662 kpat: no menu entry in Games sub-menu of GNOME block 699662 by 573865 severity 699662 minor thanks Hi, Alle domenica 3 febbraio 2013, shirish शिरीष ha scritto: > There is/was no entry of kpat in the Custom Menu Bar in > GNOME. I see several entries from KDE including Kajongg, kolf and > many others but not kpat. The gnome-menus package blacklists for GNOME various KDE applications on questionable reasons, see /etc/gnome/menus.blacklist. Just remove the kpat entry from there to get it shown in the GNOME menus. See #573865 for longer explanations and further discussions. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#699680: [kdevelop] Kdevelop is linked against wrong library name
Hi, Alle domenica 3 febbraio 2013, Julien Bigot ha scritto: > kdevelop-4:4.3.1-3+b1 depends on kdevplatform5-libs. > > kdevplatform5-libs provides libs ending in .so.5.0.0 linked by .so.5 > > the kdevelop binary is linked against libraries ending in .so.3 I don't think so: $ readelf -d /usr/bin/kdevelop | grep NEEDED 0x0001 (NEEDED) Shared library: [libkdecore.so.5] 0x0001 (NEEDED) Shared library: [libkdevplatforminterfaces.so.5] 0x0001 (NEEDED) Shared library: [libkdevplatformshell.so.5] 0x0001 (NEEDED) Shared library: [libktexteditor.so.4] 0x0001 (NEEDED) Shared library: [libkdeui.so.5] 0x0001 (NEEDED) Shared library: [libQtDBus.so.4] 0x0001 (NEEDED) Shared library: [libQtCore.so.4] 0x0001 (NEEDED) Shared library: [libQtGui.so.4] 0x0001 (NEEDED) Shared library: [libstdc++.so.6] 0x0001 (NEEDED) Shared library: [libc.so.6] Can you please describe in detail the issue(s) you are noticing? -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#700093: qtwebkit: FTBFS on x32: Tries and fails to build x86_64 JIT compiler
forwarded 700093 https://bugs.webkit.org/show_bug.cgi?id=99153 tag 700093 + upstream thanks Hi, Alle venerdì 8 febbraio 2013, Daniel Schepler ha scritto: > The qtwebkit source package is getting this build failure on the > unofficial Debian x32 port: > [...] > The attached debdiff fixes this by disabling the JIT compiler on x32 > for now since it won't work. It seems it is a bit more complex than that; there's an upstream (webkit) bug tracking all the steps needed for x32. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#700568: pu: package poppler/0.12.4-1.2+squeeze1
Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: pu
Hi,
I would like to upload a squeeze update for poppler, fixing three CVEs
(which were deemed minor, hence with no dsa), and a crasher bug and a
memory handling issue recently fixed in unstable (and wheezy).
The changes are:
* fix CVE-2010-0206:
- patch straight from upstream
* fix CVE-2010-0207:
- patch from upstream adapted to be API-/ABI-compatible, even though
the functions were private
* fix CVE-2010-4653
- patch from upstream adapted to include Object.h instead of
goo/GooLikely.h (non-existent in poppler 0.12.x)
- fix GooString::insert (#693817)
- backport the fix
- fix two uninitialized vars in PSOutputDev (#699421)
- backport the fix
I also added myself as uploader, as I did many months ago.
Let me know whether the proposed change seem okay, and I can upload to
stable.
Thanks,
--
Pino
diff -u poppler-0.12.4/debian/changelog poppler-0.12.4/debian/changelog
--- poppler-0.12.4/debian/changelog
+++ poppler-0.12.4/debian/changelog
@@ -1,3 +1,19 @@
+poppler (0.12.4-1.2+squeeze1) stable; urgency=low
+
+ * Add myself as uploader.
+ * Fix CVE-2010-0206.
+ * Fix CVE-2010-0207; patch adapted to be API-/ABI-compatible.
+ * Fix CVE-2010-4653; patch adapted to include object.h instead
+of goo/GooLikely.h (non-existent in poppler 0.12.x).
+ * Backport upstream commits 7ba15d11e56175601104d125d5e4a47619c224bf and
+55940e989701eb9118015e30f4f48eb654fa34c4 to fix GooString::insert;
+patch upstream_fix-GooString-insert.diff. (Closes: #693817)
+ * Correctly initialize PSOutputDev::fontFileNameLen and
+PSOutputDev::psFileNames; patch psoutputdev-initialize-vars.diff.
+(Closes: #699421)
+
+ -- Pino Toscano Thu, 14 Feb 2013 13:05:25 +0100
+
poppler (0.12.4-1.2) unstable; urgency=medium
* Non-maintainer upload by the Security Team
diff -u poppler-0.12.4/debian/control poppler-0.12.4/debian/control
--- poppler-0.12.4/debian/control
+++ poppler-0.12.4/debian/control
@@ -4,7 +4,8 @@
Maintainer: Loic Minier
Uploaders: Josselin Mouette ,
Dave Beckett ,
- Ross Burton
+ Ross Burton ,
+ Pino Toscano
Build-Depends: cdbs (>= 0.4.52),
debhelper (>= 5),
quilt,
diff -u poppler-0.12.4/debian/patches/series poppler-0.12.4/debian/patches/series
--- poppler-0.12.4/debian/patches/series
+++ poppler-0.12.4/debian/patches/series
@@ -4 +4,6 @@
-04_security.patch
\ No newline at end of file
+04_security.patch
+05_CVE-2010-0206.patch
+06_CVE-2010-0207.patch
+07_CVE-2010-4653.patch
+upstream_fix-GooString-insert.diff
+psoutputdev-initialize-vars.diff
only in patch2:
unchanged:
--- poppler-0.12.4.orig/debian/patches/psoutputdev-initialize-vars.diff
+++ poppler-0.12.4/debian/patches/psoutputdev-initialize-vars.diff
@@ -0,0 +1,41 @@
+Author: Pino Toscano
+Description: initialize PSOutputDev::fontFileNameLen and PSOutputDev::psFileNames
+ Avoid crashing in ~PSOutputDev when the PSOutputDev instance is not "ok".
+Applied-Upstream: not-needed
+Last-Update: 2013-01-31
+Bug-Debian: http://bugs.debian.org/699421
+
+--- a/poppler/PSOutputDev.cc
b/poppler/PSOutputDev.cc
+@@ -1012,6 +1012,7 @@ PSOutputDev::PSOutputDev(const char *fil
+ fontIDs = NULL;
+ fontFileIDs = NULL;
+ fontFileNames = NULL;
++ fontFileNameLen = 0;
+ font8Info = NULL;
+ font16Enc = NULL;
+ imgIDs = NULL;
+@@ -1022,6 +1023,7 @@ PSOutputDev::PSOutputDev(const char *fil
+ haveTextClip = gFalse;
+ haveCSPattern = gFalse;
+ t3String = NULL;
++ psFileNames = NULL;
+
+ forceRasterize = forceRasterizeA;
+
+@@ -1077,6 +1079,7 @@ PSOutputDev::PSOutputDev(PSOutputFunc ou
+ fontIDs = NULL;
+ fontFileIDs = NULL;
+ fontFileNames = NULL;
++ fontFileNameLen = 0;
+ font8Info = NULL;
+ font16Enc = NULL;
+ imgIDs = NULL;
+@@ -1087,6 +1090,7 @@ PSOutputDev::PSOutputDev(PSOutputFunc ou
+ haveTextClip = gFalse;
+ haveCSPattern = gFalse;
+ t3String = NULL;
++ psFileNames = NULL;
+
+ forceRasterize = forceRasterizeA;
+
only in patch2:
unchanged:
--- poppler-0.12.4.orig/debian/patches/05_CVE-2010-0206.patch
+++ poppler-0.12.4/debian/patches/05_CVE-2010-0206.patch
@@ -0,0 +1,56 @@
+From 30ea3ab8a1eecafb3366aef193910098fdb7ccc8 Mon Sep 17 00:00:00 2001
+From: Albert Astals Cid
+Date: Tue, 25 May 2010 23:07:56 +0100
+Subject: [PATCH] Fix crash when parsing pdf in bug 28170
+
+This code is a can of crashing worms :-7
+---
+ poppler/JBIG2Stream.cc | 23 ---
+ 1 file changed, 16 insertions(+), 7 deletions(-)
+
+diff --git a/poppler/JBIG2Stream.cc b/poppler/JBIG2Stream.cc
+index 97994bd..f16ad58 100644
+--- a/poppler/JBIG2Stream.cc
b/poppler/JBIG2Stream.cc
+@@ -742,13 +742,18 @@ JBIG2Bitmap *JBIG2Bitmap::getSlice(Guint x, Guint y, Guint wA, Guint hA) {
+ Guint xx, yy;
+
+ slice = new JBIG2Bitmap(0, wA, hA);
+- slice->clearToZero();
+- for (yy = 0; yy < hA; ++yy) {
+-
Bug#700631: doxygen: please reenable doxygen-gui on Hurd
Source: doxygen Version: 1.8.3.1-1 Severity: wishlist User: debian-h...@lists.debian.org Usertags: hurd Hi, doxygen-gui has been disabled years ago (#301954) due to unavailable qt3 on Hurd at that time; since then, qt3 (and now qt4) have been ported on Hurd, so the reason does not apply anymore. I just tested removing the doxywizard excluding with current doxygen/exp, and it compiled fine; thus, please make doxygen-gui available on Hurd again. Thanks, -- Pino -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#700568: pu: package poppler/0.12.4-1.2+squeeze1
Hi, Alle venerdì 15 febbraio 2013, Adam D. Barratt ha scritto: > On Thu, 2013-02-14 at 13:18 +0100, Pino Toscano wrote: > > +poppler (0.12.4-1.2+squeeze1) stable; urgency=low > > + > > + * Add myself as uploader. > > + * Fix CVE-2010-0206. > > + * Fix CVE-2010-0207; patch adapted to be API-/ABI-compatible. > > + * Fix CVE-2010-4653; patch adapted to include object.h instead > > +of goo/GooLikely.h (non-existent in poppler 0.12.x). > > + * Backport upstream commits > > 7ba15d11e56175601104d125d5e4a47619c224bf and + > > 55940e989701eb9118015e30f4f48eb654fa34c4 to fix GooString::insert; > > +patch upstream_fix-GooString-insert.diff. (Closes: #693817) + > > * Correctly initialize PSOutputDev::fontFileNameLen and + > > PSOutputDev::psFileNames; patch psoutputdev-initialize-vars.diff. > > + (Closes: #699421) > > Please go ahead; thanks. Thanks, uploaded. -- Pino Toscano signature.asc Description: This is a digitally signed message part.
Bug#700911: Fwd: ebook-tools
retitle 700911 simple DEP-8 testcase severity 700911 wishlist thanks Alle martedì 19 febbraio 2013, Vibhav Pant ha scritto: > I'm attaching a diff that adds DEP-8 tests to the package, originally > submitted to Launchpad. Can you please provide the link to the Launchpad bug? > I have also attached the sample epub used for the tests. At http://code.google.com/p/epub-samples/wiki/SamplesListing I see many documents, which one is the "sample.epub" you attached? Please name it in the same way also in the Debian packaging (make it easier to find out which is, and update it in the future). Also, you must mention in debian/copyright which is the license of that document. Regarding the "build" test, there are various things I don't like myself; I'll probably rework the test when I get to upgrade it. Thanks, -- Pino Toscano signature.asc Description: This is a digitally signed message part.
