Bug#119696: gtml: no problems with compression anymore: please close
Hi,
Bcc: Helmut <[EMAIL PROTECTED]>
Reply-To: Helmut <[EMAIL PROTECTED]>
Package: gtml
Version: 3.5.3-1
with 3.5.3 i have no problems anymore
So I also think this bug can be closed.
Thanks,
Helmut
PS Matej:
Sorry for not providing a sample, i spent christmas by a friend..
Bug#1102853: redland-bindings: add perl-xs-dev to Build-Depends
Source: redland-bindings Version: 1.0.17.1+dfsg-6 Severity: normal User: helm...@debian.org Usertags: libcrypt-perl-xs-dev Hello, please add "perl-xs-dev" to Build-Depends. The package builds a Perl extension module. The perl-xs-dev virtual package ensures that the functionality for building Perl extension modules is available. For the time being, this may work as is, but packages lacking this dependency are known to fail to cross build. There also is an effort[1] to drop libcrypt-dev from the build-essential package set. This will not happen during the trixie freeze. Once libcrypt-dev is no longer build-essential, this package will fail to build from source as Perl's extension module header includes crypt.h. This is expected to happen some months into the forky cycle. Please add this dependency. The additional dependency is safe for inclusion in trixie, but given the freeze, packages should not be uploaded for this change alone anymore. Thank you Helmut [1] https://lists.debian.org/debian-devel/2025/04/msg00113.html
Bug#1102850: libdbd-sybase-perl: add perl-xs-dev to Build-Depends
Source: libdbd-sybase-perl Version: 1.24-3 Severity: normal User: helm...@debian.org Usertags: libcrypt-perl-xs-dev Hello, please add "perl-xs-dev" to Build-Depends. The package builds a Perl extension module. The perl-xs-dev virtual package ensures that the functionality for building Perl extension modules is available. For the time being, this may work as is, but packages lacking this dependency are known to fail to cross build. There also is an effort[1] to drop libcrypt-dev from the build-essential package set. This will not happen during the trixie freeze. Once libcrypt-dev is no longer build-essential, this package will fail to build from source as Perl's extension module header includes crypt.h. This is expected to happen some months into the forky cycle. Please add this dependency. The additional dependency is safe for inclusion in trixie, but given the freeze, packages should not be uploaded for this change alone anymore. Thank you Helmut [1] https://lists.debian.org/debian-devel/2025/04/msg00113.html
Bug#1102788: libapreq2: add perl-xs-dev to Build-Depends
Source: libapreq2 Version: 2.17-7 Severity: normal User: helm...@debian.org Usertags: libcrypt-perl-xs-dev Hello, please add "perl-xs-dev" to Build-Depends. The package builds a Perl extension module. The perl-xs-dev virtual package ensures that the functionality for building Perl extension modules is available. For the time being, this may work as is, but packages lacking this dependency are known to fail to cross build. There also is an effort[1] to drop libcrypt-dev from the build-essential package set. This will not happen during the trixie freeze. Once libcrypt-dev is no longer build-essential, this package will fail to build from source as Perl's extension module header includes crypt.h. This is expected to happen some months into the forky cycle. Please add this dependency. The additional dependency is safe for inclusion in trixie, but given the freeze, packages should not be uploaded for this change alone anymore. Thank you Helmut [1] https://lists.debian.org/debian-devel/2025/04/msg00113.html
Bug#1102799: stfl: add perl-xs-dev to Build-Depends
Source: stfl Version: 0.22-5 Severity: normal User: helm...@debian.org Usertags: libcrypt-perl-xs-dev Hello, please add "perl-xs-dev" to Build-Depends. The package builds a Perl extension module. The perl-xs-dev virtual package ensures that the functionality for building Perl extension modules is available. For the time being, this may work as is, but packages lacking this dependency are known to fail to cross build. There also is an effort[1] to drop libcrypt-dev from the build-essential package set. This will not happen during the trixie freeze. Once libcrypt-dev is no longer build-essential, this package will fail to build from source as Perl's extension module header includes crypt.h. This is expected to happen some months into the forky cycle. Please add this dependency. The additional dependency is safe for inclusion in trixie, but given the freeze, packages should not be uploaded for this change alone anymore. Thank you Helmut [1] https://lists.debian.org/debian-devel/2025/04/msg00113.html
Bug#1102859: icheck: add perl-xs-dev to Build-Depends
Source: icheck Version: 0.9.7-8 Severity: normal User: helm...@debian.org Usertags: libcrypt-perl-xs-dev Hello, please add "perl-xs-dev" to Build-Depends. The package builds a Perl extension module. The perl-xs-dev virtual package ensures that the functionality for building Perl extension modules is available. For the time being, this may work as is, but packages lacking this dependency are known to fail to cross build. There also is an effort[1] to drop libcrypt-dev from the build-essential package set. This will not happen during the trixie freeze. Once libcrypt-dev is no longer build-essential, this package will fail to build from source as Perl's extension module header includes crypt.h. This is expected to happen some months into the forky cycle. Please add this dependency. The additional dependency is safe for inclusion in trixie, but given the freeze, packages should not be uploaded for this change alone anymore. Thank you Helmut [1] https://lists.debian.org/debian-devel/2025/04/msg00113.html
Bug#1102857: razor: add perl-xs-dev to Build-Depends
Source: razor Version: 2.85-9 Severity: normal User: helm...@debian.org Usertags: libcrypt-perl-xs-dev Hello, please add "perl-xs-dev" to Build-Depends. The package builds a Perl extension module. The perl-xs-dev virtual package ensures that the functionality for building Perl extension modules is available. For the time being, this may work as is, but packages lacking this dependency are known to fail to cross build. There also is an effort[1] to drop libcrypt-dev from the build-essential package set. This will not happen during the trixie freeze. Once libcrypt-dev is no longer build-essential, this package will fail to build from source as Perl's extension module header includes crypt.h. This is expected to happen some months into the forky cycle. Please add this dependency. The additional dependency is safe for inclusion in trixie, but given the freeze, packages should not be uploaded for this change alone anymore. Thank you Helmut [1] https://lists.debian.org/debian-devel/2025/04/msg00113.html
Bug#1102867: zbar: add perl-xs-dev to Build-Depends
Source: zbar Version: 0.23.93-7 Severity: normal User: helm...@debian.org Usertags: libcrypt-perl-xs-dev Hello, please add "perl-xs-dev" to Build-Depends. The package builds a Perl extension module. The perl-xs-dev virtual package ensures that the functionality for building Perl extension modules is available. For the time being, this may work as is, but packages lacking this dependency are known to fail to cross build. There also is an effort[1] to drop libcrypt-dev from the build-essential package set. This will not happen during the trixie freeze. Once libcrypt-dev is no longer build-essential, this package will fail to build from source as Perl's extension module header includes crypt.h. This is expected to happen some months into the forky cycle. Please add this dependency. The additional dependency is safe for inclusion in trixie, but given the freeze, packages should not be uploaded for this change alone anymore. Thank you Helmut [1] https://lists.debian.org/debian-devel/2025/04/msg00113.html
Bug#1102870: qdbm: add perl-xs-dev to Build-Depends
Source: qdbm Version: 1.8.78-13 Severity: normal User: helm...@debian.org Usertags: libcrypt-perl-xs-dev Hello, please add "perl-xs-dev" to Build-Depends. The package builds a Perl extension module. The perl-xs-dev virtual package ensures that the functionality for building Perl extension modules is available. For the time being, this may work as is, but packages lacking this dependency are known to fail to cross build. There also is an effort[1] to drop libcrypt-dev from the build-essential package set. This will not happen during the trixie freeze. Once libcrypt-dev is no longer build-essential, this package will fail to build from source as Perl's extension module header includes crypt.h. This is expected to happen some months into the forky cycle. Please add this dependency. The additional dependency is safe for inclusion in trixie, but given the freeze, packages should not be uploaded for this change alone anymore. Thank you Helmut [1] https://lists.debian.org/debian-devel/2025/04/msg00113.html
Bug#1102869: libdigest-whirlpool-perl: add perl-xs-dev to Build-Depends
Source: libdigest-whirlpool-perl Version: 1.09-4 Severity: normal User: helm...@debian.org Usertags: libcrypt-perl-xs-dev Hello, please add "perl-xs-dev" to Build-Depends. The package builds a Perl extension module. The perl-xs-dev virtual package ensures that the functionality for building Perl extension modules is available. For the time being, this may work as is, but packages lacking this dependency are known to fail to cross build. There also is an effort[1] to drop libcrypt-dev from the build-essential package set. This will not happen during the trixie freeze. Once libcrypt-dev is no longer build-essential, this package will fail to build from source as Perl's extension module header includes crypt.h. This is expected to happen some months into the forky cycle. Please add this dependency. The additional dependency is safe for inclusion in trixie, but given the freeze, packages should not be uploaded for this change alone anymore. Thank you Helmut [1] https://lists.debian.org/debian-devel/2025/04/msg00113.html
Bug#1102876: clearsilver: add perl-xs-dev to Build-Depends
Source: clearsilver Version: 0.10.5-6 Severity: normal User: helm...@debian.org Usertags: libcrypt-perl-xs-dev Hello, please add "perl-xs-dev" to Build-Depends. The package builds a Perl extension module. The perl-xs-dev virtual package ensures that the functionality for building Perl extension modules is available. For the time being, this may work as is, but packages lacking this dependency are known to fail to cross build. There also is an effort[1] to drop libcrypt-dev from the build-essential package set. This will not happen during the trixie freeze. Once libcrypt-dev is no longer build-essential, this package will fail to build from source as Perl's extension module header includes crypt.h. This is expected to happen some months into the forky cycle. Please add this dependency. The additional dependency is safe for inclusion in trixie, but given the freeze, packages should not be uploaded for this change alone anymore. Thank you Helmut [1] https://lists.debian.org/debian-devel/2025/04/msg00113.html
Bug#751809: pdnsd: Config in /usr/share/pdnsd violates FHS, POLA
Control: severity -1 wishlist On Mon, Jun 16, 2014 at 08:38:33PM +, Chris Nehren wrote: > I've installed pdnsd on a system in testing for rolling out > to production. The configuration that this package currently > uses, with the configuration in /usr/share/pdnsd, violates Debian > Policy section 9.1.1 (filesystem hierarchy), FHS specifications > for /etc and /usr/share. The configs are definitely not read-only, > and thus do not belong in /usr/share. $ apt-file search -x '^/usr/share.*\.conf$' | wc -l 3230 $ apt-file search -x '^/usr/share.*\.conf$' | cut -d: -f1 | uniq | wc -l 854 $ Are you sure that 850 packages are suddenly rc-buggy without anyone noticing? > Arguably the entire AUTO_MODE configuration system should be > removed because it is a blatant violation of both the FHS (and > Debian Policy by reference) and POLA. In no circumstance would I > expect to find configuration in /usr/share, ever. Also look into /lib/systemd/system, there are tons of configuration files as well. No, this really is not a FHS violation unless the only way to override these configurations is to change /usr, but you can simply change the one in /etc. Not a bug. > This is related to something else I found, which I'm not > sure whether to file as a separate bug. When I use my > own configuration instead of the bizarre /usr/share/pdnsd > configuration, the init scripts makes assumptions that are not > made clear: specifically, that pdnsd needs to drop privileges > to the autocreated pdnsd user and that I must enable the > status socket. The privilege dropping is obvious, but it > should still be made clear to the administrator when using > their own configuration. Without these settings, pdnsd and > its init script don't function properly. If this should be > filed separately, I'll be glad to do so. If you break it, you take the pieces, no? I kinda fail to see a bug here. For the most part it appears to be a documentation issue. Can you point out a use case that is not served by the package or a part of its documentation that you'd like to see improved? Helmut -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140616213543.ga25...@alf.mars
Bug#1062785: planner FTCBFS: fails during gtk-doc-scan
Source: planner Version: 0.14.91-3 Tags: patch User: debian-cr...@lists.debian.org Usertags: ftcbfs planner fails to cross build from source, because it fails running the gtk-doc scanner with an Exec format error. Fortunately, the documentation is split out to an Arch:all package. Hence, we can fix the cross build by disabling gtk-doc in arch-only builds. I'm attaching a patch for your convenience. Helmut --- planner-0.14.91/debian/changelog +++ planner-0.14.91/debian/changelog @@ -1,3 +1,9 @@ +planner (0.14.91-4) UNRELEASED; urgency=medium + + * Fix FTCBFS: Skip gtkdoc in arch-only build. (Closes: #-1) + + -- Helmut Grohne Sat, 03 Feb 2024 08:56:18 +0100 + planner (0.14.91-3) unstable; urgency=medium * QA upload --- planner-0.14.91/debian/rules +++ planner-0.14.91/debian/rules @@ -6,7 +6,7 @@ dh $@ override_dh_auto_configure: - dh_auto_configure -- -Dgtk_doc=true + dh_auto_configure -- -Dgtk_doc=$(if $(filter planner-doc,$(shell dh_listpackages)),true,false) override_dh_shlibdeps: dh_shlibdeps -l$(CURDIR)/debian/planner/usr/lib/$(DEB_HOST_MULTIARCH)/planner
Bug#1063323: libiw30t64: file loss due to /usr-move (DEP17)
Package: libiw30t64
Version: 30~pre9-16.1~exp1
Severity: serious
Tags: patch
User: helm...@debian.org
Usertags: dep17p1
Control: affects -1 + libiw30
X-Debbugs-Cc: vor...@debian.org
Installing libiw30t64 may cause file loss in an upgrade scenario.
Reproducer:
mmdebstrap --variant=apt --verbose bookworm /dev/null "deb
http://deb.debian.org/debian bookworm main" "deb http://deb.debian.org/debian
experimental main" --include=wireless-tools --chrooted-customize-hook="apt-get
download libiw30t64" --chrooted-customize-hook="echo libiw30 deinstall | dpkg
--set-selections" --chrooted-customize-hook="dpkg --auto-deconfigure --unpack
*.deb" --chrooted-customize-hook="dpkg --configure -a"
--chrooted-customize-hook="dpkg -r libiw30" --chrooted-customize-hook="! dpkg
--verify | grep ."
I'm attaching a patch for your convenience. I consider that libiw30 is
not as central as other packages and hence propose employing Conflicts
here. Conflicts allow removing the protective diversion in trixie's
postinst rather than forky's postinst already.
Helmut
diff --minimal -Nru wireless-tools-30~pre9/debian/changelog
wireless-tools-30~pre9/debian/changelog
--- wireless-tools-30~pre9/debian/changelog 2024-02-04 21:34:45.0
+0100
+++ wireless-tools-30~pre9/debian/changelog 2024-02-06 07:33:48.0
+0100
@@ -1,3 +1,9 @@
+wireless-tools (30~pre9-16.1~exp2) UNRELEASED; urgency=medium
+
+ * Fix /usr-move file loss. (Closes: #-1)
+
+ -- Helmut Grohne Tue, 06 Feb 2024 07:33:48 +0100
+
wireless-tools (30~pre9-16.1~exp1) experimental; urgency=medium
* Non-maintainer upload.
diff --minimal -Nru wireless-tools-30~pre9/debian/clean
wireless-tools-30~pre9/debian/clean
--- wireless-tools-30~pre9/debian/clean 1970-01-01 01:00:00.0 +0100
+++ wireless-tools-30~pre9/debian/clean 2024-02-06 07:33:30.0 +0100
@@ -0,0 +1,2 @@
+debian/libiw30t64.preinst
+debian/libiw30t64.postinst
diff --minimal -Nru wireless-tools-30~pre9/debian/control
wireless-tools-30~pre9/debian/control
--- wireless-tools-30~pre9/debian/control 2024-02-04 21:34:45.0
+0100
+++ wireless-tools-30~pre9/debian/control 2024-02-06 07:31:36.0
+0100
@@ -31,8 +31,7 @@
Package: libiw30t64
Provides: ${t64:Provides}
-Replaces: libiw30
-Breaks: libiw30 (<< ${source:Version})
+Conflicts: libiw30
Section: libs
Architecture: linux-any
Multi-Arch: same
diff --minimal -Nru wireless-tools-30~pre9/debian/libiw30t64.lintian-overrides
wireless-tools-30~pre9/debian/libiw30t64.lintian-overrides
--- wireless-tools-30~pre9/debian/libiw30t64.lintian-overrides 2024-02-04
21:34:45.0 +0100
+++ wireless-tools-30~pre9/debian/libiw30t64.lintian-overrides 2024-02-06
07:33:48.0 +0100
@@ -1 +1,5 @@
libiw30t64: package-name-doesnt-match-sonames libiw30
+# begin-remove-after: released:trixie
+# DEP17 protective diversion
+diversion-for-unknown-file lib/x86_64-linux-gnu/libiw.so.30 [preinst:*]
+# end-remove-after
diff --minimal -Nru wireless-tools-30~pre9/debian/libiw30t64.postinst.in
wireless-tools-30~pre9/debian/libiw30t64.postinst.in
--- wireless-tools-30~pre9/debian/libiw30t64.postinst.in1970-01-01
01:00:00.0 +0100
+++ wireless-tools-30~pre9/debian/libiw30t64.postinst.in2024-02-06
07:29:28.0 +0100
@@ -0,0 +1,13 @@
+#!/bin/sh
+
+set -e
+
+# begin-remove-after: released:trixie
+if [ "$1" = configure ]; then
+ dpkg-divert --package libiw30t64 --no-rename --remove --divert
"/lib/#DEB_HOST_MULTIARCH#/libiw.so.30.usr-is-merged"
"/lib/#DEB_HOST_MULTIARCH#/libiw.so.30"
+fi
+# end-remove-after
+
+#DEBHELPER#
+
+exit 0
diff --minimal -Nru wireless-tools-30~pre9/debian/libiw30t64.preinst.in
wireless-tools-30~pre9/debian/libiw30t64.preinst.in
--- wireless-tools-30~pre9/debian/libiw30t64.preinst.in 1970-01-01
01:00:00.0 +0100
+++ wireless-tools-30~pre9/debian/libiw30t64.preinst.in 2024-02-06
07:29:30.0 +0100
@@ -0,0 +1,13 @@
+#!/bin/sh
+
+set -e
+
+# begin-remove-after: released:trixie
+if [ "$1" = install ]; then
+ dpkg-divert --package libiw30t64 --no-rename --add --divert
"/lib/#DEB_HOST_MULTIARCH#/libiw.so.30.usr-is-merged"
"/lib/#DEB_HOST_MULTIARCH#/libiw.so.30"
+fi
+# end-remove-after
+
+#DEBHELPER#
+
+exit 0
diff --minimal -Nru wireless-tools-30~pre9/debian/rules
wireless-tools-30~pre9/debian/rules
--- wireless-tools-30~pre9/debian/rules 2023-11-28 01:03:11.0 +0100
+++ wireless-tools-30~pre9/debian/rules 2024-02-06 07:33:39.0 +0100
@@ -19,3 +19,8 @@
override_dh_installudev:
dh_installudev --priority=19
+
+debian/%:debian/%.in
+ sed -e 's/#DEB_HOST_MULTIARCH#/$(DEB_HOST_MULTIARCH)/g' $< > $@
+
+execute_before_dh_installdeb:debian/libiw30t64.preinst
debian/libiw30t64.postinst
Bug#1063323: libiw30t64: file loss due to /usr-move (DEP17)
On Tue, Feb 06, 2024 at 07:42:49AM +0100, Helmut Grohne wrote:
> I'm attaching a patch for your convenience. I consider that libiw30 is
> not as central as other packages and hence propose employing Conflicts
> here. Conflicts allow removing the protective diversion in trixie's
> postinst rather than forky's postinst already.
Steve made me aware that such Conflicts and Breaks should be versioned
as they may otherwise interact with Provides and multiarch. Updated
patch attached.
Helmut
diff --minimal -Nru wireless-tools-30~pre9/debian/changelog
wireless-tools-30~pre9/debian/changelog
--- wireless-tools-30~pre9/debian/changelog 2024-02-04 21:34:45.0
+0100
+++ wireless-tools-30~pre9/debian/changelog 2024-02-06 07:33:48.0
+0100
@@ -1,3 +1,9 @@
+wireless-tools (30~pre9-16.1~exp2) UNRELEASED; urgency=medium
+
+ * Fix /usr-move file loss. (Closes: #-1)
+
+ -- Helmut Grohne Tue, 06 Feb 2024 07:33:48 +0100
+
wireless-tools (30~pre9-16.1~exp1) experimental; urgency=medium
* Non-maintainer upload.
diff --minimal -Nru wireless-tools-30~pre9/debian/clean
wireless-tools-30~pre9/debian/clean
--- wireless-tools-30~pre9/debian/clean 1970-01-01 01:00:00.0 +0100
+++ wireless-tools-30~pre9/debian/clean 2024-02-06 07:33:30.0 +0100
@@ -0,0 +1,2 @@
+debian/libiw30t64.preinst
+debian/libiw30t64.postinst
diff --minimal -Nru wireless-tools-30~pre9/debian/control
wireless-tools-30~pre9/debian/control
--- wireless-tools-30~pre9/debian/control 2024-02-04 21:34:45.0
+0100
+++ wireless-tools-30~pre9/debian/control 2024-02-06 07:31:36.0
+0100
@@ -31,8 +31,7 @@
Package: libiw30t64
Provides: ${t64:Provides}
-Replaces: libiw30
-Breaks: libiw30 (<< ${source:Version})
+Conflicts: libiw30 (<< ${source:Version})
Section: libs
Architecture: linux-any
Multi-Arch: same
diff --minimal -Nru wireless-tools-30~pre9/debian/libiw30t64.lintian-overrides
wireless-tools-30~pre9/debian/libiw30t64.lintian-overrides
--- wireless-tools-30~pre9/debian/libiw30t64.lintian-overrides 2024-02-04
21:34:45.0 +0100
+++ wireless-tools-30~pre9/debian/libiw30t64.lintian-overrides 2024-02-06
07:33:48.0 +0100
@@ -1 +1,5 @@
libiw30t64: package-name-doesnt-match-sonames libiw30
+# begin-remove-after: released:trixie
+# DEP17 protective diversion
+diversion-for-unknown-file lib/x86_64-linux-gnu/libiw.so.30 [preinst:*]
+# end-remove-after
diff --minimal -Nru wireless-tools-30~pre9/debian/libiw30t64.postinst.in
wireless-tools-30~pre9/debian/libiw30t64.postinst.in
--- wireless-tools-30~pre9/debian/libiw30t64.postinst.in1970-01-01
01:00:00.0 +0100
+++ wireless-tools-30~pre9/debian/libiw30t64.postinst.in2024-02-06
07:29:28.0 +0100
@@ -0,0 +1,13 @@
+#!/bin/sh
+
+set -e
+
+# begin-remove-after: released:trixie
+if [ "$1" = configure ]; then
+ dpkg-divert --package libiw30t64 --no-rename --remove --divert
"/lib/#DEB_HOST_MULTIARCH#/libiw.so.30.usr-is-merged"
"/lib/#DEB_HOST_MULTIARCH#/libiw.so.30"
+fi
+# end-remove-after
+
+#DEBHELPER#
+
+exit 0
diff --minimal -Nru wireless-tools-30~pre9/debian/libiw30t64.preinst.in
wireless-tools-30~pre9/debian/libiw30t64.preinst.in
--- wireless-tools-30~pre9/debian/libiw30t64.preinst.in 1970-01-01
01:00:00.0 +0100
+++ wireless-tools-30~pre9/debian/libiw30t64.preinst.in 2024-02-06
07:29:30.0 +0100
@@ -0,0 +1,13 @@
+#!/bin/sh
+
+set -e
+
+# begin-remove-after: released:trixie
+if [ "$1" = install ]; then
+ dpkg-divert --package libiw30t64 --no-rename --add --divert
"/lib/#DEB_HOST_MULTIARCH#/libiw.so.30.usr-is-merged"
"/lib/#DEB_HOST_MULTIARCH#/libiw.so.30"
+fi
+# end-remove-after
+
+#DEBHELPER#
+
+exit 0
diff --minimal -Nru wireless-tools-30~pre9/debian/rules
wireless-tools-30~pre9/debian/rules
--- wireless-tools-30~pre9/debian/rules 2023-11-28 01:03:11.0 +0100
+++ wireless-tools-30~pre9/debian/rules 2024-02-06 07:33:39.0 +0100
@@ -19,3 +19,8 @@
override_dh_installudev:
dh_installudev --priority=19
+
+debian/%:debian/%.in
+ sed -e 's/#DEB_HOST_MULTIARCH#/$(DEB_HOST_MULTIARCH)/g' $< > $@
+
+execute_before_dh_installdeb:debian/libiw30t64.preinst
debian/libiw30t64.postinst
Bug#1063452: tty-clock FTCBFS: hard codes the build architecture pkg-config
Source: tty-clock
Version: 2.3-4
Tags: patch upstream
User: debian-cr...@lists.debian.org
Usertags: ftcbfs
tty-clock fails to cross build from source, because the upstream
Makefile hard codes the build architecture pkg-config. I'm attaching a
patch that makes pkg-config substitutable. After applying it, tty-clock
becomes cross buildable.
Helmut
--- tty-clock-2.3.orig/Makefile
+++ tty-clock-2.3/Makefile
@@ -4,6 +4,7 @@
SRC = ttyclock.c
CC ?= gcc
+PKG_CONFIG ?= pkg-config
BIN = tty-clock
PREFIX ?= /usr/local
INSTALLPATH = ${DESTDIR}${PREFIX}/bin
@@ -16,8 +17,8 @@
CFLAGS += -Wall -g $$(ncursesw5-config --cflags)
LDFLAGS += $$(ncursesw5-config --libs)
else
- CFLAGS += -Wall -g $$(pkg-config --cflags ncurses)
- LDFLAGS += $$(pkg-config --libs ncurses)
+ CFLAGS += -Wall -g $$($(PKG_CONFIG) --cflags ncurses)
+ LDFLAGS += $$($(PKG_CONFIG) --libs ncurses)
endif
tty-clock : ${SRC}
Bug#1064495: sitecopy FTBFS: incompatible neon version
Source: sitecopy Version: 1:0.16.6-11 Severity: serious Tags: ftbfs sitecopy fails to build from source in unstable. I think the relevant part of the build log is: | checking for neon-config... /usr/bin/neon-config | checking linking against neon... yes | configure: incompatible neon library version 0.33.0: wanted 0.24 25 26 27 28 31 32 | configure: error: could not use external neon library | tail -v -n \+0 config.log Apparently, sitecopy is incompatible with neon >= 0.33. This quite definitely needs a sourceful upload of sitecopy. Helmut
Bug#1064997: tdfsb: fails to trap errors from ./compile.sh
Source: tdfsb Version: 0.0.10-3 Severity: serious Justification: policy 4.6 tdfsb's compile.sh fails to terminate with an error when one of the invoked build commands fails. This violates Debian policy section 4.6. A relatively simple counter measure would be adding "set -e" to the script. Helmut
Bug#1065126: libelk0t64 has an undeclared file conflict
Package: libelk0t64 Version: 3.99.8-5 Severity: serious User: debian...@lists.debian.org Usertags: fileconflict Control: affects -1 + libelk0 libelk0t64 has an undeclared file conflict. This may result in an unpack error from dpkg. The files * /usr/lib/x86_64-linux-gnu/libelk-xlib.so.0 * /usr/lib/x86_64-linux-gnu/libelk-xlib.so.0.0.0 * /usr/lib/x86_64-linux-gnu/libelk-xt.so.0 * /usr/lib/x86_64-linux-gnu/libelk-xt.so.0.0.0 * /usr/lib/x86_64-linux-gnu/libelk.so.0 * /usr/lib/x86_64-linux-gnu/libelk.so.0.0.0 are contained in the packages * libelk0/3.99.8-4.2+b1 as present in bookworm|bullseye|trixie|unstable * libelk0t64/3.99.8-5 as present in unstable These packages can be unpacked concurrently, because there is no relevant Replaces or Conflicts relation. Attempting to unpack these packages concurrently results in an unpack error from dpkg, because none of the packages installs a diversion for the affected files. Kind regards The Debian Usr Merge Analysis Tool This bug report has been automatically filed with no human intervention. The source code is available at https://salsa.debian.org/helmutg/dumat. If the filing is unclear or in error, don't hesitate to contact hel...@subdivi.de for assistance.
Bug#1070125: dmucs FTCBFS: uses the build architecture compiler
Source: dmucs Version: 0.6.1+dfsg-1 Tags: patch User: debian-cr...@lists.debian.org Usertags: ftcbfs dmucs fails to cross build from source, because it uses the build architecture compiler. override_dh_auto_configure now invokes ./configure without passing --host and hence configure uses the wrong compiler. Moreover, the upstream build system fails to forward the detected compiler to the COSMIC subdirectory. I'm attaching a patch to fix both for your convenience. Helmut diff -Nru dmucs-0.6.1+dfsg/debian/changelog dmucs-0.6.1+dfsg/debian/changelog --- dmucs-0.6.1+dfsg/debian/changelog 2024-04-22 02:19:52.0 +0200 +++ dmucs-0.6.1+dfsg/debian/changelog 2024-04-30 13:31:33.0 +0200 @@ -1,3 +1,11 @@ +dmucs (0.6.1+dfsg-2) UNRELEASED; urgency=medium + + * Fix FTCBFS: (Closes: #-1) ++ Pass --host to configure. ++ cross.patch: Forward CC to COSMIC subdir. + + -- Helmut Grohne Tue, 30 Apr 2024 13:31:33 +0200 + dmucs (0.6.1+dfsg-1) unstable; urgency=medium * QA upload. diff -Nru dmucs-0.6.1+dfsg/debian/patches/cross.patch dmucs-0.6.1+dfsg/debian/patches/cross.patch --- dmucs-0.6.1+dfsg/debian/patches/cross.patch 1970-01-01 01:00:00.0 +0100 +++ dmucs-0.6.1+dfsg/debian/patches/cross.patch 2024-04-30 13:31:33.0 +0200 @@ -0,0 +1,19 @@ +--- dmucs-0.6.1+dfsg.orig/COSMIC/Makefile dmucs-0.6.1+dfsg/COSMIC/Makefile +@@ -29,7 +29,7 @@ + # One may also start up the PortMaster (Spm -f firewallfilename). + # Please read the documentation on this. + +-CC = cc ++CC ?= cc + + OBJ = Saccept.o Sprintf.o Stest.ooutofmem.o \ + Sclose.o Sprtskt.o Stimeoutwait.o rdcolor.o \ +--- dmucs-0.6.1+dfsg.orig/Makefile.am dmucs-0.6.1+dfsg/Makefile.am +@@ -1,3 +1,5 @@ ++export CC ++ + SUBDIRS = COSMIC + + bin_PROGRAMS = gethost loadavg monitor remhost diff -Nru dmucs-0.6.1+dfsg/debian/patches/series dmucs-0.6.1+dfsg/debian/patches/series --- dmucs-0.6.1+dfsg/debian/patches/series 2024-04-20 05:31:05.0 +0200 +++ dmucs-0.6.1+dfsg/debian/patches/series 2024-04-30 13:31:33.0 +0200 @@ -3,3 +3,4 @@ 03_gcc-7.patch 40_reproducible.patch 50_fix-FTBS-GCC-13.patch +cross.patch diff -Nru dmucs-0.6.1+dfsg/debian/rules dmucs-0.6.1+dfsg/debian/rules --- dmucs-0.6.1+dfsg/debian/rules 2024-04-22 02:09:44.0 +0200 +++ dmucs-0.6.1+dfsg/debian/rules 2024-04-30 13:31:30.0 +0200 @@ -3,11 +3,13 @@ export DEB_BUILD_MAINT_OPTIONS = hardening=+all export DEB_CFLAGS_MAINT_APPEND = -Wall +include /usr/share/dpkg/architecture.mk + override_dh_auto_install: cp -a remhost addhost override_dh_auto_configure: - ./configure + ./configure --build=$(DEB_BUILD_GNU_TYPE) --host=$(DEB_HOST_GNU_TYPE) override_dh_clean: dh_clean
Bug#983227: coz-profiler: drop unused Build-Depends
Hi Petter, On Thu, May 09, 2024 at 11:07:20AM +0200, Petter Reinholdtsen wrote: > The reason it have these build dependencies, is to make sure the package > only build on architectures where it will actually work, to avoid > earning a release critical bug of providing binaries that do not work. > > I am thus very reluctant to drop build dependencies also needed at > runtime. Your reasoning on this is sound to me. Would you mind adding comments to debian/control documenting this non-obvious aspect such that the next time someone looks for technically unused dependencies they see why they're there? >From my pov, please close this bug when documenting their need. Helmut
Bug#1071008: libx52pro0: installs udev rules twice to /usr and /
Package: libx52pro0 Version: 0.1.1-3 Severity: serious Justification: policy 10.1 Tags: patch X-Debbugs-Cc: Petter Reinholdtsen Hi, since the last upload, libx52pro0 contains both /lib/udev/rules.d/99-x52pro.rules and /usr/lib/udev/rules.d/99-x52pro.rule. Doing so violates Debian policy section 10.1. The former is installed via the upstream build system combined with dh_install and debian/libx52pro0.install while the latter is installed via debian/*.udev with dh_installudev. Given DEP17, the latter is the desired location. I'm attaching a patch for your convenience. Helmut diff --minimal -Nru x52pro-0.1.1/debian/changelog x52pro-0.1.1/debian/changelog --- x52pro-0.1.1/debian/changelog 2024-05-12 10:39:38.0 +0200 +++ x52pro-0.1.1/debian/changelog 2024-05-12 22:59:37.0 +0200 @@ -1,3 +1,9 @@ +x52pro (0.1.1-4) UNRELEASED; urgency=medium + + * Install udev rules only once. (Closes: #-1) + + -- Helmut Grohne Sun, 12 May 2024 22:59:37 +0200 + x52pro (0.1.1-3) unstable; urgency=medium * QA upload. diff --minimal -Nru x52pro-0.1.1/debian/libx52pro0.install x52pro-0.1.1/debian/libx52pro0.install --- x52pro-0.1.1/debian/libx52pro0.install 2024-05-12 10:14:01.0 +0200 +++ x52pro-0.1.1/debian/libx52pro0.install 2024-05-12 22:59:36.0 +0200 @@ -1,4 +1,3 @@ -lib/udev/rules.d usr/bin/x52output usr/lib/lib*.so.* usr/share/man diff --minimal -Nru x52pro-0.1.1/debian/not-installed x52pro-0.1.1/debian/not-installed --- x52pro-0.1.1/debian/not-installed 1970-01-01 01:00:00.0 +0100 +++ x52pro-0.1.1/debian/not-installed 2024-05-12 22:59:37.0 +0200 @@ -0,0 +1,2 @@ +# Installed via debian/*.udev symbolic link +lib/udev/rules.d
Bug#1079960: Should fruit be removed from unstable?
Source: fruit Severity: serious Justification: grab attention of maintainer User: helm...@debian.org Usertags: sidremove Dear maintainer, I suggest removing fruit from Debian for the following reasons: * It accumulated one RC-bug: + #1014187: fruit: Binary file without a source code Last modified: 2 years * It is not part of bookworm or trixie and is not a key package. This bug serves as a pre-removal warning. After one month, the bug will be reassigned to ftp.debian.org to actually request removal of the package. In case the package should be kept in unstable, please evaluate each of the RC-bugs listed above. * If the bug is meant to prevent the package from entering testing or a stable release, but this package should stay part of unstable, please add a usertag: user helm...@debian.org usertags NNN + sidremove-ignore * If the bug no longer applies, please close it. If it is closed, check whether the fixed version is correct and adjust if necessary. * Is the bug really release-critical? If not, please downgrade. * If the bug still applies, please send a status update at least once a year. Once all of the mentioned RC bugs have been acted upon in one way or another, please close this bug. In case the package should be removed from unstable, you may reassign this bug report: Control: severity -1 normal Control: retitle -1 RM: fruit -- RoM; rc-buggy Control: reassign -1 ftp.debian.org Control: affects -1 + src:fruit Alternatively, you may wait a month and have it reassigned. In case you disagree with the above, please downgrade this bug below RC severity. Doing so will also prevent automatic reassignment. Kind regards A tool for automatically removing packages from unstable This bug report has been automatically filed with little human intervention. If the filing is unclear or in error, don't hesitate to contact Helmut Grohne for assistance.
Bug#1080501: Should airport-utils be removed from unstable?
Source: airport-utils Severity: serious Justification: grab attention of maintainer User: helm...@debian.org Usertags: sidremove Dear maintainer, I suggest removing airport-utils from Debian for the following reasons: * It accumulated one RC-bug: + #414092: airport-utils: Tools start and quit immediately without working Last modified: 1 year, 1 month * It is not part of bookworm or trixie and is not a key package. This bug serves as a pre-removal warning. After one month, the bug will be reassigned to ftp.debian.org to actually request removal of the package. In case the package should be kept in unstable, please evaluate each of the RC-bugs listed above. * If the bug is meant to prevent the package from entering testing or a stable release, but this package should stay part of unstable, please add a usertag: user helm...@debian.org usertags NNN + sidremove-ignore * If the bug no longer applies, please close it. If it is closed, check whether the fixed version is correct and adjust if necessary. * Is the bug really release-critical? If not, please downgrade. * If the bug still applies, please send a status update at least once a year. Once all of the mentioned RC bugs have been acted upon in one way or another, please close this bug. In case the package should be removed from unstable, you may reassign this bug report: Control: severity -1 normal Control: retitle -1 RM: airport-utils -- RoM; rc-buggy Control: reassign -1 ftp.debian.org Control: affects -1 + src:airport-utils Alternatively, you may wait a month and have it reassigned. In case you disagree with the above, please downgrade this bug below RC severity. Doing so will also prevent automatic reassignment. Kind regards A tool for automatically removing packages from unstable This bug report has been automatically filed with little human intervention. If the filing is unclear or in error, don't hesitate to contact Helmut Grohne for assistance.
Bug#1022983: obexftp FTCBFS: ruby not found
Source: obexftp Version: 0.24-7 Tags: patch User: debian-cr...@lists.debian.org Usertags: ftcbfs obexftp already cross builds given the noruby build profile. Recently, Antonio Terceiro worked on cross build infrastructure for the ruby side and now we can make the whole obexftp cross build succeed. Please consider applying the attached patch. Helmut diff --minimal -Nru obexftp-0.24/debian/changelog obexftp-0.24/debian/changelog --- obexftp-0.24/debian/changelog 2022-07-11 23:45:14.0 +0200 +++ obexftp-0.24/debian/changelog 2022-10-28 16:30:29.0 +0200 @@ -1,3 +1,9 @@ +obexftp (0.24-8) UNRELEASED; urgency=medium + + * Fix FTCBFS: Set ruby paths. (Closes: #-1) + + -- Helmut Grohne Fri, 28 Oct 2022 16:30:29 +0200 + obexftp (0.24-7) unstable; urgency=medium * QA upload. diff --minimal -Nru obexftp-0.24/debian/rules obexftp-0.24/debian/rules --- obexftp-0.24/debian/rules 2022-07-11 23:42:34.0 +0200 +++ obexftp-0.24/debian/rules 2022-10-28 16:30:27.0 +0200 @@ -20,6 +20,10 @@ endif ifneq ($(filter ruby-obexftp,$(DOPACKAGES)),) RUBY_DEPENDS := -Vruby:Depends="lib$(firstword $(shell dh_ruby --print-supported))" +RUBY_VER=$(shell ruby -e "puts RbConfig::CONFIG['ruby_version']") +ifneq ($(DEB_BUILD_ARCH),$(DEB_HOST_ARCH)) +export RUBYLIB=/usr/lib/$(DEB_HOST_MULTIARCH)/ruby-crossbuild/$(RUBY_VER) +endif endif %:
Bug#1023316: 9base: re-adds /bin/rc to /etc/shells on upgrade
Source: 9base
Version: 1:6-12
Tags: patch
Severity: serious
Justification: Debian policy 10.7.3
If I remove /bin/rc from /etc/shells and then upgrade or reinstall
9base, it will re-add itself to /etc/shells. Such behaviour violates
Debian policy section 10.7.3. I propose moving to dpkg triggers and
update-shells to fix this issue and attach a patch for your convenience.
Helmut
diff --minimal -Nru 9base-6/debian/changelog 9base-6/debian/changelog
--- 9base-6/debian/changelog2022-03-17 21:15:01.0 +0100
+++ 9base-6/debian/changelog2022-11-02 09:45:01.0 +0100
@@ -1,3 +1,9 @@
+9base (1:6-13) UNRELEASED; urgency=medium
+
+ * Update /etc/shells via dpkg triggers. (Closes: #-1)
+
+ -- Helmut Grohne Wed, 02 Nov 2022 09:45:01 +0100
+
9base (1:6-12) unstable; urgency=medium
* QA upload.
diff --minimal -Nru 9base-6/debian/control 9base-6/debian/control
--- 9base-6/debian/control 2022-03-17 21:15:01.0 +0100
+++ 9base-6/debian/control 2022-11-02 09:45:01.0 +0100
@@ -11,7 +11,7 @@
Package: 9base
Architecture: any
-Depends: ${misc:Depends}, ${shlibs:Depends}
+Depends: ${misc:Depends}, ${shlibs:Depends}, debianutils (>= 5.3-1~)
Description: Plan 9 userland tools
9base is a port of following original Plan 9 userland tools to Unix:
ascii, awk, basename, bc, cal, cat, cleanname, cmp, date, dc, dd, diff, du,
diff --minimal -Nru 9base-6/debian/postinst 9base-6/debian/postinst
--- 9base-6/debian/postinst 2022-03-17 21:15:01.0 +0100
+++ 9base-6/debian/postinst 2022-11-02 09:45:01.0 +0100
@@ -5,10 +5,6 @@
# and so we can maintain compatibility with rc scripts expecting /bin/rc
update-alternatives --install /bin/rc rc /usr/lib/plan9/bin/rc 20
-# allow use of 9base' rc as a login shell:
-add-shell /bin/rc
-add-shell /usr/lib/plan9/bin/rc
-
#DEBHELPER#
exit 0
diff --minimal -Nru 9base-6/debian/postrm 9base-6/debian/postrm
--- 9base-6/debian/postrm 2022-03-17 21:15:01.0 +0100
+++ 9base-6/debian/postrm 1970-01-01 01:00:00.0 +0100
@@ -1,9 +0,0 @@
-#!/bin/sh
-set -e
-
-if [ "$1" = "remove" ]; then
-remove-shell /bin/rc
-remove-shell /usr/lib/plan9/bin/rc
-fi
-
-#DEBHELPER#
diff --minimal -Nru 9base-6/debian/rules 9base-6/debian/rules
--- 9base-6/debian/rules2022-03-17 21:15:01.0 +0100
+++ 9base-6/debian/rules2022-11-02 09:45:01.0 +0100
@@ -52,6 +52,8 @@
mv $$MANPAGE $${MANPAGE}$(MANEXT); \
done
+ install -D -m644 debian/shells
$(PKGDIR)$(DATADIR)/debianutils/shells.d/plan9
+
override_dh_fixperms:
dh_fixperms
diff --minimal -Nru 9base-6/debian/shells 9base-6/debian/shells
--- 9base-6/debian/shells 1970-01-01 01:00:00.0 +0100
+++ 9base-6/debian/shells 2022-11-02 09:45:01.0 +0100
@@ -0,0 +1,2 @@
+/bin/rc
+/usr/lib/plan9/bin/rc
Bug#1024258: redland FTCBFS: multiple issues
Source: redland
Version: 1.0.17-3
Tags: patch
User: debian-cr...@lists.debian.org
Usertags: ftcbfs
Control: block -1 by 1024241
X-Debbugs-Cc: Andrej Shadura
redland fails to cross build from source for multiple reasons. The
immediate failure is a missing dependency on pkg-config. As such, it
fails discovering rasqal and fails. Beyond that, it also uses
mysql_config, which is known to not work during cross compilation.
Instead, pkg-config should be used. Unfortunately, the .pc file shipped
by mariadb wrongly reports an ancient version (#1024241). This must be
fixed before using it. Once that has happened, the attached patch makes
redland cross buildable.
Helmut
diff --minimal -Nru redland-1.0.17/debian/changelog
redland-1.0.17/debian/changelog
--- redland-1.0.17/debian/changelog 2022-11-14 06:03:12.0 +0100
+++ redland-1.0.17/debian/changelog 2022-11-16 09:28:10.0 +0100
@@ -1,3 +1,11 @@
+redland (1.0.17-4) UNRELEASED; urgency=medium
+
+ * Fix FTCBFS: (Closes: #-1)
++ Missing Build-Depends: pkg-config.
++ cross.patch: avoid using mysql_config
+
+ -- Helmut Grohne Wed, 16 Nov 2022 09:28:10 +0100
+
redland (1.0.17-3) unstable; urgency=medium
* QA upload.
diff --minimal -Nru redland-1.0.17/debian/control redland-1.0.17/debian/control
--- redland-1.0.17/debian/control 2022-11-14 05:57:56.0 +0100
+++ redland-1.0.17/debian/control 2022-11-16 09:28:08.0 +0100
@@ -15,6 +15,7 @@
libsqlite3-dev,
libtool,
perl:any,
+ pkg-config,
unixodbc-dev,
Rules-Requires-Root: no
Standards-Version: 4.6.1
diff --minimal -Nru redland-1.0.17/debian/patches/cross.patch
redland-1.0.17/debian/patches/cross.patch
--- redland-1.0.17/debian/patches/cross.patch 1970-01-01 01:00:00.0
+0100
+++ redland-1.0.17/debian/patches/cross.patch 2022-11-16 09:28:10.0
+0100
@@ -0,0 +1,81 @@
+--- redland-1.0.17.orig/configure.ac
redland-1.0.17/configure.ac
+@@ -659,25 +659,34 @@
+ fi
+ mysql_config=yes
+ fi
+-if test $mysql_config = yes -a "X$MYSQL_CONFIG" = "X" ; then
+- AC_CHECK_PROGS(MYSQL_CONFIG, mysql_config)
+-fi
+
+ libmysql_min_version=3.23.56
+ # Also tested on 4.0.14
+
+-mysql_cflags=
+-mysql_libs=
++MYSQL_CPPFLAGS=
++MYSQL_LIBS=
++
++AS_IF([test "$mysql_config" = yes],[
++ PKG_CHECK_MODULES([MYSQL],[mysqlclient >= $libmysql_min_version],[
++mysql_config=no
++MYSQL_CPPFLAGS=$MYSQL_CFLAGS
++storages_available="$storages_available mysql"
++mysql_storage_available=yes
++ ],[])
++])
++
++if test $mysql_config = yes -a "X$MYSQL_CONFIG" = "X" ; then
++ AC_CHECK_PROGS(MYSQL_CONFIG, mysql_config)
++fi
++
+ if test "X$MYSQL_CONFIG" != X; then
+ dnl need to change quotes to allow square brackets
+ changequote(<<, >>)dnl
+- mysql_libs=`$MYSQL_CONFIG --libs | sed -e "s/'//g"`
++ MYSQL_LIBS=`$MYSQL_CONFIG --libs | sed -e "s/'//g"`
+ # Stop the MySQL config program adding -Os or -O options to CFLAGS
+- mysql_cflags=`$MYSQL_CONFIG --cflags | sed -e "s/'//g" -e "s/[
\t]-O[A-Za-z0-9]*/ /" `
++ MYSQL_CPPFLAGS=`$MYSQL_CONFIG --cflags | sed -e "s/'//g" -e "s/[
\t]-O[A-Za-z0-9]*/ /" `
+ changequote([, ])dnl
+
+- LIBS="$LIBRDF_LIBS $mysql_libs"
+- CPPFLAGS="$LIBRDF_CPPFLAGS $mysql_cflags"
+ AC_MSG_CHECKING(for mysql library)
+ LIBMYSQL_VERSION=`$MYSQL_CONFIG --version`
+ libmysql_version_dec=`echo $LIBMYSQL_VERSION | $AWK -F. '{printf("%d\n",
1*$1 + 100*$2 + $3)};'`
+@@ -689,7 +698,11 @@
+ storages_available="$storages_available mysql($LIBMYSQL_VERSION)"
+ mysql_storage_available=yes
+ fi
++fi
+
++AS_IF([test -n "$MYSQL_LIBS"],[
++ LIBS="$LIBRDF_LIBS $MYSQL_LIBS"
++ CPPFLAGS="$LIBRDF_CPPFLAGS $MYSQL_CPPFLAGS"
+ AC_MSG_CHECKING(whether MYSQL_OPT_RECONNECT is declared)
+ AC_COMPILE_IFELSE([AC_LANG_SOURCE([#include
+ int x=MYSQL_OPT_RECONNECT;])],
+@@ -699,8 +712,7 @@
+
+ LIBS="$LIBRDF_LIBS"
+ CPPFLAGS="$LIBRDF_CPPFLAGS"
+-fi
+-
++])
+
+
+ AC_ARG_WITH(sqlite, [ --with-sqlite(=yes|no) Enable SQLite store
(default=auto)], with_sqlite="$withval", with_sqlite="auto")
+@@ -1071,9 +1083,9 @@
+ AM_CONDITIONAL(STORAGE_POSTGRESQL, test $postgresql_storage = yes)
+ AM_CONDITIONAL(STORAGE_VIRTUOSO, test $virtuoso_storage = yes)
+
+-if test $mysql_storage = yes; then
+- MYSQL_CPPFLAGS=`$MYSQL_CONFIG --cflags | sed -e "s/'//g" -e
"s/-O[A-Za-z0-9]*//" `
+- MYSQL_LIBS="$LIBRDF_LIBS $mysql_libs"
++if test $mysql_storage != yes; then
++ MYSQL_CPPFLAGS=
++ MYSQL_LIBS=
+ fi
+ if test $tstore_storage = yes; then
+ TSTORE_CPPFLAGS="`$TSTORE_CONFIG --cflags` `$PKG_CONFIG glib --cflags`"
diff --minimal -Nru redland-1.0.17/debian/patches/series
redland-1.0.17/debian/patches/series
--- redland-1.0.17/debian/patches/series1970-01-01 01:00:00.0
+0100
+++ redland-1.0.17/debian/patches/series2022-11-16 09:28:10.0
+0100
@@ -0,0 +1 @@
+cross.patch
Bug#1024927: libtextwrap should support the noudeb build profile
Source: libtextwrap
Version: 0.1-15
Severity: minor
Tags: patch
User: helm...@debian.org
Usertags: rebootstrap
libtextwrap builds a udeb. Therefore, it should support a noudeb build
profile. It's a one-line change and I'm attaching a patch for your
convenience.
Helmut
diff --minimal -Nru libtextwrap-0.1/debian/changelog
libtextwrap-0.1/debian/changelog
--- libtextwrap-0.1/debian/changelog2021-09-14 23:53:23.0 +0200
+++ libtextwrap-0.1/debian/changelog2022-11-27 16:58:09.0 +0100
@@ -1,3 +1,9 @@
+libtextwrap (0.1-16) UNRELEASED; urgency=medium
+
+ * Support the noudeb build profile. (Closes: #-1)
+
+ -- Helmut Grohne Sun, 27 Nov 2022 16:58:09 +0100
+
libtextwrap (0.1-15) unstable; urgency=medium
* QA upload.
diff --minimal -Nru libtextwrap-0.1/debian/control
libtextwrap-0.1/debian/control
--- libtextwrap-0.1/debian/control 2021-09-14 23:34:20.0 +0200
+++ libtextwrap-0.1/debian/control 2022-11-27 16:58:08.0 +0100
@@ -60,4 +60,5 @@
${misc:Depends},
${shlibs:Depends},
Package-Type: udeb
+Build-Profiles:
Description: text-wrapping library with i18n - runtime udeb
Bug#1031777: gkrellmitime FTCBFS: uses the build architecture pkg-config
Source: gkrellmitime Version: 1.0.1-7 Tags: patch upstream User: debian-cr...@lists.debian.org Usertags: ftcbfs gkrellmitime fails to cross build from source, because the upstream Makefile hard codes the build architecture pkg-config. I'm attaching a patch for your convenience. Helmut --- gkrellmitime-1.0.1.orig/Makefile +++ gkrellmitime-1.0.1/Makefile @@ -6,7 +6,8 @@ INCLUDEDIR ?= $(PREFIX)/include # Support for Debian policy -CFLAGS += -I$(INCLUDEDIR) `pkg-config gtk+-2.0 --cflags` +PKG_CONFIG ?= pkg-config +CFLAGS += -I$(INCLUDEDIR) `$(PKG_CONFIG) gtk+-2.0 --cflags` INSTALL = install INSTALL_FILE= $(INSTALL) -p-o root -g root -m 644 INSTALL_DIR = $(INSTALL) -p -d -o root -g root -m 755 @@ -17,7 +18,7 @@ $(TARGET).so: $(TARGET).o - $(CC) $(LDFLAGS) -shared `pkg-config gtk+-2.0 --libs` $< -o $@ + $(CC) $(LDFLAGS) -shared `$(PKG_CONFIG) gtk+-2.0 --libs` $< -o $@ $(TARGET).o: $(TARGET).c $(CC) $(CPPFLAGS) $(CFLAGS) -c $<
Bug#948134: gnome-shell-mailnag FTCBFS: stores valac in CC
Hi Jeremy, On Fri, Feb 24, 2023 at 10:01:09AM -0500, Jeremy Bícha wrote: > This change seems to cause the package to fail the build. Before your > change, this is a build warning not an error. This is due to improperly combining the new version with my patch. Patching the reported version works. > dh_auto_build -- CC=valac > make -j8 "INSTALL=install --strip-program=true" CC=valac > make[2]: Entering directory '/<>' > valac --pkg folks aggregate-avatars.vala > aggregate-avatars.vala:53.3-53.41: warning: unhandled error `GLib.Error' >53 | cache_dir.make_directory_with_parents(); > | ^~~ > Unknown option -I/usr/include/glib-2.0 > Run 'valac --help' to see a full list of available command line options. > error: cc exited with status 256 > Compilation failed: 1 error(s), 1 warning(s) At this time, the Makefile stuffs compiler flags into the CC variable. Passing CC, overrides them and they thus get lost. Please separate them out of CC. Helmut
Bug#1032498: stymulator FTCBFS: does not pass cross tools to make
Source: stymulator Version: 0.21a~dfsg-3 Tags: patch User: debian-cr...@lists.debian.org Usertags: ftcbfs stymulator fails to cross build from source, because it does not pass cross tools to make. The easiest way of doing so - using dh_auto_build - makes stymulator cross buildable. I'm attaching a patch for your convenience. Helmut diff --minimal -Nru stymulator-0.21a~dfsg/debian/changelog stymulator-0.21a~dfsg/debian/changelog --- stymulator-0.21a~dfsg/debian/changelog 2022-11-04 00:37:59.0 +0100 +++ stymulator-0.21a~dfsg/debian/changelog 2023-03-07 20:17:06.0 +0100 @@ -1,3 +1,9 @@ +stymulator (0.21a~dfsg-4) UNRELEASED; urgency=medium + + * Fix FTCBFS: Let dh_auto_build pass cross tools to make. (Closes: #-1) + + -- Helmut Grohne Tue, 07 Mar 2023 20:17:06 +0100 + stymulator (0.21a~dfsg-3) unstable; urgency=medium * QA upload. diff --minimal -Nru stymulator-0.21a~dfsg/debian/rules stymulator-0.21a~dfsg/debian/rules --- stymulator-0.21a~dfsg/debian/rules 2022-11-04 00:37:59.0 +0100 +++ stymulator-0.21a~dfsg/debian/rules 2023-03-07 20:17:04.0 +0100 @@ -14,7 +14,7 @@ build: build-stamp build-stamp: dh_testdir - $(MAKE) CXXFLAGS="$(CXXFLAGS)" CXX="g++" -C src/ + dh_auto_build -D src -- CXXFLAGS="$(CXXFLAGS)" touch $@ clean:
Bug#717825: please allow ghostscript to cross build
On Sat, Mar 18, 2023 at 12:52:57PM +0100, Håvard F. Aasen wrote: > Helmut, do you have an opinion on the importance of this issue? In general, cross build bugs are not release-critical. Regressions affecting architecture bootstrap tend to be considered important and fixed during freeze. Everything else is best effort. Given that ghostscript is a key package, this would need an unblock. Thanks for pointing at the fix however. Helmut
Bug#1033211: clp: configure script misdetects tons of things
Source: clp Version: 1.17.6-3 Severity: important User: debian-cr...@lists.debian.org Usertags: ftcbfs Hi, I happened to look into a clp cross build failure, but what I found was more problems. If you look at configure output, you see this: | checking whether the compiler supports GNU C... no | checking whether gcc accepts -g... no | checking for gcc option to enable C11 features... unsupported | checking for gcc option to enable C99 features... unsupported | checking for gcc option to enable C89 features... unsupported Quite obviously, this is all wrong and autoconf will turn on quite a bit of barely tested workarounds due to these misdetections. So I looked into why this happens and ultimately, all of these use ac_fn_c_try_compile and they all say "$? = 0" followed by "failed program" in config.log. So compilation was successful, but it was treated as failure, because conftest.$ac_objext didn't exist. As it happens, ac_objext is undefined, because _AC_COMPILER_OBJEXT hasn't been run yet. It gets run later and then ac_fn_c_try_compile actually works, but when these tests are rerun, their cached values are used. I looked into the m4 files to see why this would happen, but I ended up not identifying the cause. However, if you export ac_objext=o, ac_fn_c_try_compile actually works and these things get correctly detected. As it happens, this also fixes cross compilation, but really the misdetections are a problem and should be fixed even for native builds. I'm attaching a patch that works around the issue by setting ac_objext, but I hope you can find a better solution. Helmut diff --minimal -Nru clp-1.17.6/debian/changelog clp-1.17.6/debian/changelog --- clp-1.17.6/debian/changelog 2022-11-27 05:27:26.0 +0100 +++ clp-1.17.6/debian/changelog 2023-03-19 19:40:11.0 +0100 @@ -1,3 +1,9 @@ +clp (1.17.6-4) UNRELEASED; urgency=medium + + * Work around severe configure misdetections. (Closes: #-1) + + -- Helmut Grohne Sun, 19 Mar 2023 19:40:11 +0100 + clp (1.17.6-3) unstable; urgency=medium * QA upload. diff --minimal -Nru clp-1.17.6/debian/rules clp-1.17.6/debian/rules --- clp-1.17.6/debian/rules 2022-11-27 05:27:26.0 +0100 +++ clp-1.17.6/debian/rules 2023-03-19 19:40:11.0 +0100 @@ -9,7 +9,10 @@ dh_autoreconf autoreconf -- --force --install --no-recursive --include=BuildTools override_dh_auto_configure: - dh_auto_configure -- --enable-static --enable-dot --enable-dependency-linking + # The configure script is broken as it uses ac_fn_c_try_compile before + # having run _AC_COMPILER_OBJEXT, so supply ac_objext to make it barely + # work. + ac_objext=o dh_auto_configure -- --enable-static --enable-dot --enable-dependency-linking override_dh_auto_build: dh_auto_build
Bug#1034006: libowfat FTCBFS: builds for the build architecture
Source: libowfat Version: 0.32-4 Tags: patch User: debian-cr...@lists.debian.org Usertags: ftcbfs libowfat fails to cross build from source, because it builds for the build architecture. To cross build libowfat one is supposed to supply either CCC or CROSS. Beyond that, it fails to use CCC for the linker step. I'm attaching a patch for your convenience. Helmut diff --minimal -Nru libowfat-0.32/debian/changelog libowfat-0.32/debian/changelog --- libowfat-0.32/debian/changelog 2022-11-10 16:32:49.0 +0100 +++ libowfat-0.32/debian/changelog 2023-04-06 09:31:10.0 +0200 @@ -1,3 +1,11 @@ +libowfat (0.32-5) UNRELEASED; urgency=medium + + * Fix FTCBFS: (Closes: #-1) ++ Pass CCC to make. ++ cross.patch: Use CCC for linking as well. + + -- Helmut Grohne Thu, 06 Apr 2023 09:31:10 +0200 + libowfat (0.32-4) unstable; urgency=high * QA upload diff --minimal -Nru libowfat-0.32/debian/patches/cross.patch libowfat-0.32/debian/patches/cross.patch --- libowfat-0.32/debian/patches/cross.patch1970-01-01 01:00:00.0 +0100 +++ libowfat-0.32/debian/patches/cross.patch2023-04-06 09:31:10.0 +0200 @@ -0,0 +1,9 @@ +--- libowfat-0.32.orig/GNUmakefile libowfat-0.32/GNUmakefile +@@ -394,5 +394,5 @@ + ln -sf $< $@ + + libowfat.so.0: $(ALL_OBJS) +- $(CC) -shared -Wl,-soname=$@ $(LDFLAGS) $^ -o $@ ++ $(CCC) -shared -Wl,-soname=$@ $(LDFLAGS) $^ -o $@ + diff --minimal -Nru libowfat-0.32/debian/patches/series libowfat-0.32/debian/patches/series --- libowfat-0.32/debian/patches/series 2022-11-10 16:31:16.0 +0100 +++ libowfat-0.32/debian/patches/series 2023-04-06 09:31:10.0 +0200 @@ -6,3 +6,4 @@ 07-fix-FD_CLOEXEC.patch 08-clang-ftbfs.diff fix_gcc10.patch +cross.patch diff --minimal -Nru libowfat-0.32/debian/rules libowfat-0.32/debian/rules --- libowfat-0.32/debian/rules 2022-11-10 16:31:16.0 +0100 +++ libowfat-0.32/debian/rules 2023-04-06 09:31:10.0 +0200 @@ -1,5 +1,7 @@ #!/usr/bin/make -f +include /usr/share/dpkg/buildtools.mk + CPPFLAGS:=$(shell dpkg-buildflags --get CPPFLAGS) CFLAGS:=$(shell dpkg-buildflags --get CFLAGS) $(CPPFLAGS) -I. CXXFLAGS:=$(shell dpkg-buildflags --get CXXFLAGS) $(CPPFLAGS) @@ -9,9 +11,9 @@ dh $@ override_dh_auto_build: - $(MAKE) -f $(CURDIR)/GNUmakefile CFLAGS="$(CFLAGS) -fPIC" DIET='' libowfat.so + $(MAKE) -f $(CURDIR)/GNUmakefile CCC=$(CC) CFLAGS="$(CFLAGS) -fPIC" DIET='' libowfat.so $(MAKE) -f $(CURDIR)/GNUmakefile clean - $(MAKE) -f $(CURDIR)/GNUmakefile CFLAGS="$(CFLAGS)" DIET='' + $(MAKE) -f $(CURDIR)/GNUmakefile CCC=$(CC) CFLAGS="$(CFLAGS)" DIET='' rm entities.h override_dh_installchangelogs:
Bug#1035820: 9base: leaves entries in /etc/shells after upgrade from bullseye
Control: forcemerge 1033167 -1 Control: affects 1033167 + 9base Hi Andreas, On Tue, May 09, 2023 at 04:39:21PM +0200, Andreas Beckmann wrote: > during a test with piuparts I noticed your package leaves modifications > in /etc/shells after upgrading from bullseye to bookworm and purging the > package. > > 9base/bullseye called add-shell/remove-shell in its postinst/postrm. > 9base/bookworm no longer does that, but it also does not clean up the > leftover entries from bullseye in its postinst. 9base/bookworm no longer does, because it now uses dpkg-triggers to perform the cleanup. It actually does clean up its entries. > >From the attached log (scroll to the bottom...): > > 0m45.2s ERROR: FAIL: After purging files have been modified: > /etc/shells not owned You should look closer: 0m45.2s DEBUG: Modified(user, group, mode, size, target): /etc/shells expected(root, root, - 100644, 128, None) != found(root, root, - 100644, 140, None) It's a 12 byte difference. That's not 9base's entries. What you see here is "/usr/bin/sh\n". So this is a /usr-merge bug. We already know it. Thus force-merging. > The following (untested) snippet for the postinst should perform the > neccessary cleanup: > > if [ "$1" = "install" ] || [ "$1" = "upgrade" ]; then > if dpkg --compare-versions "$2" lt-nl "1:6-14~" ; then > remove-shell /bin/rc > remove-shell /usr/lib/plan9/bin/rc > fi > fi No. Please continue to use the declarative approach. Helmut
Bug#1040322: debdry: should not ship test data in /usr/lib/python3/dist-packages/control
Package: debdry Version: 0.2.2-1 Severity: serious debdry creates a control file in its test suite. This file happens to get installed into /usr/lib/python3/dist-packages/control and this happens to cause an undeclared file conflict with python3-kombu. I'm filing this with debdry, because it clearly should not be installing this file. Helmut
Bug#1042409: python3-distorm3,libdistorm3-3: undeclared file conflict /usr/lib/x86_64-linux-gnu/libdistorm3.so.3
Package: python3-distorm3,libdistorm3-3 Version: 3.5.2b-1 Severity: serious Hi, both libdistorm3-3 and python3-distorm3 contain /usr/lib/x86_64-linux-gnu/libdistorm3.so.3. When coinstalling these packages, an unpack error results. I suspect that the python one shouldn't contain the shared library. Do you concur? Helmut
Bug#1043245: libdbus-c++-ecore-1-0/experimental: undeclared file conflict with libdbus-c++-1-0v5/unsable
Package: libdbus-c++-ecore-1-0 Version: 0.9.0-13 Severity: serious Control: affects -1 + libdbus-c++-1-0v5 User: debian...@lists.debian.org Usertags: fileconflict Both libdbus-c++-ecore-1-0/experimental and libdbus-c++-1-0v5/unstable contain the files /usr/lib/x86_64-linux-gnu/libdbus-c++-ecore-1.so.0 and /usr/lib/x86_64-linux-gnu/libdbus-c++-ecore-1.so.0.0.0 without resolving this conflict via Conflicts, Replaces or diversions. This may result in an unpack error from dpkg and is a serious issue. I guess that you meant to restructure the library and that these files are supposed to be moved between packages. In this case, please add Breaks and Replaces declarations for libdbus-c++-1-0v5 to libdbus-c++-ecore-1-0. Helmut
Bug#1043246: libdbus-c++-glib-1-0/experimental: undeclared file conflict with libdbus-c++-1-0v5/unsable
Package: libdbus-c++-glib-1-0 Version: 0.9.0-13 Severity: serious Control: affects -1 + libdbus-c++-1-0v5 User: debian...@lists.debian.org Usertags: fileconflict Both libdbus-c++-glib-1-0/experimental and libdbus-c++-1-0v5/unstable contain the files /usr/lib/x86_64-linux-gnu/libdbus-c++-glib-1.so.0 and /usr/lib/x86_64-linux-gnu/libdbus-c++-glib-1.so.0.0.0 without resolving this conflict via Conflicts, Replaces or diversions. This may result in an unpack error from dpkg and is a serious issue. I guess that you meant to restructure the library and that these files are supposed to be moved between packages. In this case, please add Breaks and Replaces declarations for libdbus-c++-1-0v5 to libdbus-c++-glib-1-0. Helmut
Bug#1051364: nvi uses Conflicts in violation of policy 10.1
Package: nvi Version: 1.81.6-18 Severity: serious Justification: policy 10.1 nvi has declared Conflicts with nex due to the duplicated use of /usr/bin/nex. Since nvi has come first, this purely is a bug in the nex package and it is tracked as #1022957. On the nvi side, such a conflict is harmful, because a fixed nex (one that no longer contains /usr/bin/nex) cannot be coinstalled. The conflict is declared in direct violation of policy section 10.1 and should therefore be removed. Helmut
Bug#1051459: fbreader FTCBFS: strips with the build architecture strip
Source: fbreader Version: 0.99.4+dfsg-3 Tags: patch User: debian-cr...@lists.debian.org Usertags: ftcbfs fbreader regressed cross building since 0.99.4+dfsg-3. It now strips with the build architecture strip. Beyond breaking cross building, this also breaks DEB_BUILD_OPTIONS=nostrip as well as generation of -dbgsym packages. It is best to skip any stripping during build and letting debhelper do its job. I'm attaching a patch for your convenience. Helmut diff --minimal -Nru fbreader-0.99.4+dfsg/debian/changelog fbreader-0.99.4+dfsg/debian/changelog --- fbreader-0.99.4+dfsg/debian/changelog 2023-09-07 17:41:50.0 +0200 +++ fbreader-0.99.4+dfsg/debian/changelog 2023-09-08 10:04:51.0 +0200 @@ -1,3 +1,9 @@ +fbreader (0.99.4+dfsg-6) UNRELEASED; urgency=medium + + * Fix FTCBFS: Defer stripping to debhelper. (Closes: #-1) + + -- Helmut Grohne Fri, 08 Sep 2023 10:04:51 +0200 + fbreader (0.99.4+dfsg-5) unstable; urgency=medium * QA upload. diff --minimal -Nru fbreader-0.99.4+dfsg/debian/patches/nostrip.patch fbreader-0.99.4+dfsg/debian/patches/nostrip.patch --- fbreader-0.99.4+dfsg/debian/patches/nostrip.patch 1970-01-01 01:00:00.0 +0100 +++ fbreader-0.99.4+dfsg/debian/patches/nostrip.patch 2023-09-08 10:04:46.0 +0200 @@ -0,0 +1,15 @@ +--- fbreader-0.99.4+dfsg.orig/fbreader/desktop/Makefile fbreader-0.99.4+dfsg/fbreader/desktop/Makefile +@@ -7,7 +7,7 @@ + BINARY = $(DESTDIR)$(BINDIR)/FBReader + + install: +- @strip $(BINARY) ++ @$(STRIP) $(BINARY) + @install -d $(DESTDIR)/usr/share/applications + @install -m 0644 desktop $(DESTDIR)/usr/share/applications/$(TARGET).desktop + @install -d $(DESTDIR)$(IMAGEDIR) +--- /dev/null fbreader-0.99.4+dfsg/makefiles/arch/unix.mk +@@ -0,0 +1 @@ ++STRIP ?= strip diff --minimal -Nru fbreader-0.99.4+dfsg/debian/patches/series fbreader-0.99.4+dfsg/debian/patches/series --- fbreader-0.99.4+dfsg/debian/patches/series 2023-09-07 17:41:50.0 +0200 +++ fbreader-0.99.4+dfsg/debian/patches/series 2023-09-08 10:01:15.0 +0200 @@ -6,3 +6,4 @@ 0009-fbreader-Makefile-do-not-try-to-install-empty-format.patch 0011-zlibrary-unix-curl-avoid-duplicate-case-in-a-switch.patch 0013-fbreader-Makefile-make-linking-order-reproducible.patch +nostrip.patch diff --minimal -Nru fbreader-0.99.4+dfsg/debian/rules fbreader-0.99.4+dfsg/debian/rules --- fbreader-0.99.4+dfsg/debian/rules 2023-09-07 17:41:50.0 +0200 +++ fbreader-0.99.4+dfsg/debian/rules 2023-09-08 10:04:51.0 +0200 @@ -37,7 +37,7 @@ $(MAKE) -C zlibrary/core TARGET_ARCH=desktop UI_TYPE=dummy DESTDIR=$(DEB_DIR)/libzlcore-dev do_install_dev $(MAKE) -C zlibrary/text TARGET_ARCH=desktop UI_TYPE=dummy DESTDIR=$(DEB_DIR)/libzltext-dev do_install_dev $(MAKE) -C zlibrary/ui TARGET_ARCH=desktop UI_TYPE=qt4 DESTDIR=$(DEB_DIR)/fbreader do_install - $(MAKE) -C fbreader TARGET_ARCH=desktop UI_TYPE=dummy DESTDIR=$(DEB_DIR)/fbreader do_install + dh_auto_install -D fbreader -- TARGET_ARCH=desktop UI_TYPE=dummy DESTDIR=$(DEB_DIR)/fbreader do_install override_dh_makeshlibs: dh_makeshlibs -plibzlcore0.13
Bug#1054014: ladvd: defer location of systemd unit to systemd.pc
Source: ladvd
Version: 1.1.2-2
Tags: patch
User: helm...@debian.org
Usertags: dep17m2
We want to move aliased files from / to /usr to finalize the /usr-merge
transition via DEP17. ladvd is affected, because it ships a systemd unit
below /lib. Rather than moving it to /usr, I recommend deferring the
placement to systemd.pc. If ladvd gets backported to bookworm, it'll
automatically revert to the old /lib placement thus honouring the file
move moratorium that still affects bookworm. I'm attaching a patch for
your convenience.
Helmut
diff -Nru ladvd-1.1.2/debian/changelog ladvd-1.1.2/debian/changelog
--- ladvd-1.1.2/debian/changelog2022-10-23 08:10:08.0 +0200
+++ ladvd-1.1.2/debian/changelog2023-10-15 22:06:55.0 +0200
@@ -1,3 +1,9 @@
+ladvd (1.1.2-3) UNRELEASED; urgency=medium
+
+ * Defer location of systemd unit to systemd.pc. (Closes: #-1)
+
+ -- Helmut Grohne Sun, 15 Oct 2023 22:06:55 +0200
+
ladvd (1.1.2-2) unstable; urgency=medium
* QA upload.
diff -Nru ladvd-1.1.2/debian/control ladvd-1.1.2/debian/control
--- ladvd-1.1.2/debian/control 2022-10-23 08:10:08.0 +0200
+++ ladvd-1.1.2/debian/control 2023-10-15 22:00:17.0 +0200
@@ -5,7 +5,8 @@
Build-Depends: debhelper-compat (= 13),
autoconf, automake, libtool, pkg-config,
libbsd-dev (>= 0.6.0),
- libpcap-dev, libcap-ng-dev, libevent-dev, libpci-dev, libmnl-dev, libteam-dev
+ libpcap-dev, libcap-ng-dev, libevent-dev, libpci-dev, libmnl-dev,
libteam-dev,
+ systemd-dev,
Standards-Version: 4.6.1
Rules-Requires-Root: no
Homepage: https://github.com/sspans/ladvd/
diff -Nru ladvd-1.1.2/debian/rules ladvd-1.1.2/debian/rules
--- ladvd-1.1.2/debian/rules2022-10-23 08:10:08.0 +0200
+++ ladvd-1.1.2/debian/rules2023-10-15 22:00:38.0 +0200
@@ -39,7 +39,7 @@
../configure $(CONFFLAGS) \
--disable-silent-rules \
--prefix=/usr --mandir=\$${prefix}/share/man \
- --with-systemdsystemunitdir=/lib/systemd/system \
+ --with-systemdsystemunitdir=$(shell pkg-config
--variable=systemdsystemunitdir systemd) \
--with-pid-dir=/run/ladvd
build: build-stamp
Bug#1054476: yaskkserv: will install yaskkserv.service twice once dh_installsystemd installs to /usr
Source: yaskkserv Version: 1.1.1-2 Severity: normal Tags: patch User: helm...@debian.org Usertags: dep17m2 We want to modify dh_installsystemd such that units are installed below /usr in order to finalize the /usr-merge transition via DEP17. When doing so, yaskkserv.service will be installed to both /lib (via dh_install) and /usr/lib (via dh_installsystemd). Such behaviour constitutes a policy violation and therefore this bug will become release-critical once I upload debhelper. I'm attaching a patch for your convenience. Helmut diff -Nru yaskkserv-1.1.1/debian/changelog yaskkserv-1.1.1/debian/changelog --- yaskkserv-1.1.1/debian/changelog2023-10-03 10:19:38.0 +0200 +++ yaskkserv-1.1.1/debian/changelog2023-10-24 13:36:09.0 +0200 @@ -1,3 +1,9 @@ +yaskkserv (1.1.1-3) UNRELEASED; urgency=medium + + * Install yaskkserv.service only once. (Closes: #-1) + + -- Helmut Grohne Tue, 24 Oct 2023 13:36:09 +0200 + yaskkserv (1.1.1-2) unstable; urgency=medium * QA upload. diff -Nru yaskkserv-1.1.1/debian/yaskkserv.install yaskkserv-1.1.1/debian/yaskkserv.install --- yaskkserv-1.1.1/debian/yaskkserv.install2023-10-03 10:17:47.0 +0200 +++ yaskkserv-1.1.1/debian/yaskkserv.install2023-10-24 13:36:06.0 +0200 @@ -1,2 +1 @@ debian/update-skkdic-yaskkserv usr/sbin -debian/yaskkserv.socket lib/systemd/system
Bug#1059392: eboard FTCBFS: builds for the build architecture
Source: eboard Version: 1.1.3-2 Tags: patch User: debian-cr...@lists.debian.org Usertags: ftcbfs eboard fails to cross build from source, because it uses the build architecture compiler as a make default. Consider applying the attached patch to let dpkg's buildtools.mk initialize CC correctly for cross compilation. Helmut diff --minimal -Nru eboard-1.1.3/debian/changelog eboard-1.1.3/debian/changelog --- eboard-1.1.3/debian/changelog 2023-09-04 20:21:10.0 +0200 +++ eboard-1.1.3/debian/changelog 2023-12-22 11:19:12.0 +0100 @@ -1,3 +1,9 @@ +eboard (1.1.3-3) UNRELEASED; urgency=medium + + * Fix FTCBFS: Let dpkg's buildtools.mk initialize CC. (Closes: #-1) + + -- Helmut Grohne Fri, 22 Dec 2023 11:19:12 +0100 + eboard (1.1.3-2) unstable; urgency=medium * QA upload. diff --minimal -Nru eboard-1.1.3/debian/rules eboard-1.1.3/debian/rules --- eboard-1.1.3/debian/rules 2023-09-04 20:21:10.0 +0200 +++ eboard-1.1.3/debian/rules 2023-12-22 11:19:12.0 +0100 @@ -2,6 +2,7 @@ export DEB_BUILD_MAINT_OPTIONS = hardening=+all +-include /usr/share/dpkg/buildtools.mk include /usr/share/dpkg/buildflags.mk %:
Bug#613494: bacula-sd: bsock.c: 135 Unable to connect to Storage daemon on ...
Package: bacula Version: 5.0.2-2.2 Severity: wishlist I got no answer if I use the command >status sto<. Solution: SDAdress 127.0.0.1 must be changed to FQDN in bacula-sd.conf -- System Information: Debian Release: 6.0 APT prefers squeeze-updates APT policy: (500, 'squeeze-updates'), (500, 'stable') Architecture: amd64 (x86_64) Kernel: Linux 2.6.32-5-amd64 (SMP w/2 CPU cores) Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages bacula depends on: ii bacula-client 5.0.2-2.2network backup, recovery and verif ii bacula-common 5.0.2-2.2+b1 network backup, recovery and verif ii bacula-server 5.0.2-2.2network backup, recovery and verif bacula recommends no packages. Versions of packages bacula suggests: pn bacula-doc (no description available) -- no debconf information -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110215084159.1530.48616.reportbug@keyeu-lnfs9-svr.keyeu.keyplastics.local
Bug#674913: sgml-data: do not call update-catalog
Package: sgml-data Version: 2.0.6 Severity: important Your package calls update-catalog on /etc/sgml/$PACKAGE.cat. The latter file will transition to a conffile, so changing that file induces questions to the user. Please remove the old and no longer needed call to update-catalog. Helmut -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120528150618.ga4...@alf.mars
Bug#675488: sgml-data: please rebuild to fix your copy of #477751
Source: sgml-data Severity: serious Dear maintainer(s) of sgml-data, TL;DR: Please upload a new version of this package closing this bug. Problem ~~~ Your package uses the dh_installcatalogs helper from debhelper. This helper added code to the postinst that unconditionally overwrites files in /etc which is a policy violation. The corresponding bug #477751 is now solved in debhelper. Nevertheless the code overwriting files in /etc is still present in a binary package built from this source package, so your package needs a rebuild. Unfortunately the binary package in question is Architecture: all, so a binNMU is not enough. How to solve This bug tracks the progress of the rebuild and should be closed by any upload of this package. Before building, please ensure that your debhelper version is at least 9.20120528 which should be the case if you are running sid. Is my package really/still affected? Any binary package using the dh_installcatalogs helper will add a versioned dependency on sgml-base. If the depended upon version is at least 1.26+nmu2, your package is not affected. In that case, just close this bug. If you have any further questions concerning this issue, please don't hesitate to contact me. Thanks for your help Helmut -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/debian-qa-packages
Bug#674913: sgml-data: do not call update-catalog
tags 674913 + patch
tags 675488 + patch
thanks
On Mon, May 28, 2012 at 05:06:18PM +0200, Helmut Grohne wrote:
> Your package calls update-catalog on /etc/sgml/$PACKAGE.cat. The latter
> file will transition to a conffile, so changing that file induces
> questions to the user. Please remove the old and no longer needed call
> to update-catalog.
I attached a patch for both issues.
Helmut
diff -Nru sgml-data-2.0.6/debian/changelog sgml-data-2.0.6+nmu1/debian/changelog
--- sgml-data-2.0.6/debian/changelog2011-05-29 18:04:05.0 +0200
+++ sgml-data-2.0.6+nmu1/debian/changelog 2012-06-12 11:47:46.0
+0200
@@ -1,3 +1,11 @@
+sgml-data (2.0.6+nmu1) unstable; urgency=low
+
+ * Non-maintainer upload. (Closes: #675488)
+ * Remove calls to update-catalog from postinst. They are noops with the new
+sgml-base. (Closes: #674913)
+
+ -- Helmut Grohne Tue, 12 Jun 2012 11:46:23 +0200
+
sgml-data (2.0.6) unstable; urgency=low
* QA upload.
diff -Nru sgml-data-2.0.6/debian/postinst sgml-data-2.0.6+nmu1/debian/postinst
--- sgml-data-2.0.6/debian/postinst 2004-02-15 18:34:00.0 +0100
+++ sgml-data-2.0.6+nmu1/debian/postinst2012-06-12 11:44:47.0
+0200
@@ -1,16 +1,8 @@
#!/bin/sh
# sgml-data postinst
-# jobs: remove old catalog entry in transitional.cat
set -e
-PACKAGE=sgml-data
-
-if [ "$1" = configure ]; then
-# remove legacy SGML catalog
-update-catalog --quiet --remove "/etc/sgml/${PACKAGE}.cat"
/usr/share/sgml/html/dtd/xml/1.0/xhtml.soc
-fi
-
# automatically generated debhelper commands
#DEBHELPER#
Bug#687812: vnstat --live: min > avg
Package: vnstat Version: 1.11-1 Severity: normal Here is a sample output of vnstat --live. |rx | tx | --+-- | bytes 1.97 MiB |2.05 MiB | --+-- | max 126.50 KiB/s | 88.00 KiB/s | average80.72 KiB/s | 84.00 KiB/s | min38.50 KiB/s | 87.00 KiB/s | --+-- | packets 2454 |2401 | --+-- | max123 p/s | 125 p/s | average 98 p/s | 96 p/s | min 84 p/s | 81 p/s | --+-- | time25 seconds Notice that it lists average tx as 84 KiB and min tx as 87 KiB. From a mathematical point of view this sounds wrong. This behaviour occurs "sometimes". I cannot tell a reliable way to reproduce. Obviously you need very constant traffic to exhibit this behaviour. A possible cause may be the use of integer arithmetics when computing the timespent. There is room for one second error there and indeed 84 / 25 * 26 > 87. Helmut -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120916073858.ga14...@alf.mars
Bug#703053: lastmp: switch from python-libmpdclient to python-mpd
Package: lastmp Version: 1.0.6-4 Severity: wishlist Dear lastmp maintainer, The lastmp package is using the python-libmpdclient (mpdclient2) module. This library has been abandoned upstream[1] in favour of python-mpd. Would it be possible to switch this tool over to the other library? Helmut [1] http://incise.org/py-libmpdclient2.html -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130314172236.ga22...@alf.mars
Bug#703054: pygmy: switch from python-libmpdclient to python-mpd
Package: pygmy Version: 0.48-4 Severity: wishlist Dear pygmy maintainer, The pygmy package is using the python-libmpdclient (mpdclient2) module. This library has been abandoned upstream[1] in favour of python-mpd. Would it be possible to switch this tool over to the other library? Helmut [1] http://incise.org/py-libmpdclient2.html -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130314172133.ga21...@alf.mars
Bug#703053: lastmp: switch from python-libmpdclient to python-mpd
Control: tags -1 + patch
On Thu, Mar 14, 2013 at 06:22:37PM +0100, Helmut Grohne wrote:
> The lastmp package is using the python-libmpdclient (mpdclient2) module.
> This library has been abandoned upstream[1] in favour of python-mpd.
> Would it be possible to switch this tool over to the other library?
I attached a patch for this. Note that this patch is based on the
library documentation and completely untested (due to the lack of a
last.fm account). Could anyone test the patch and report back a result
to me and this bug?
The patch applies to the lastmp python script and you will have to
install python-mpd to make it work.
Thanks
Helmut
--- lastmp 2010-04-19 03:41:31.0 +0200
+++ lastmp.new 2013-03-15 09:48:42.0 +0100
@@ -5,8 +5,9 @@
import time
import getopt
import signal
+import socket
-import mpdclient2
+import mpd
import lastfm
import lastfm.client
import lastfm.config
@@ -26,11 +27,11 @@
class Song:
def __init__(self, sobj):
-self.artist = getattr(sobj, 'artist', '')
-self.title = getattr(sobj, 'title', '')
-self.album = getattr(sobj, 'album', '')
-self.length = int(getattr(sobj, 'time', 0))
-self.file = getattr(sobj, 'file', '')
+self.artist = sobj.get('artist', '')
+self.title = sobj.get('title', '')
+self.album = sobj.get('album', '')
+self.length = int(sobj.get('time', 0))
+self.file = sobj.get('file', '')
def __eq__(self, other):
if other == None:
@@ -64,6 +65,13 @@
else:
return d
+def mpd_connect(mpd_args):
+mpd = mpd.MPDClient()
+mpd.connect(mpd_args["host"], mpd_args["port"])
+if mpd_args["password"]:
+mpd.password(mpd_args["password"])
+return mpd
+
class MPDMonitor:
def __init__(self, cli):
self.cli = cli
@@ -74,25 +82,25 @@
self.mpd = None
def wake(self):
-status = self.mpd.do.status()
-song = Song(self.mpd.do.currentsong())
+status = self.mpd.status()
+song = Song(self.mpd.currentsong())
-if not hasattr(status, 'state'):
+if 'state' not in status:
raise MPDAuthError
-if not self.prevstatus or status.state != self.prevstatus.state:
-self.cli.log.debug('Changed state: %s' % status.state)
+if not self.prevstatus or status["state"] != self.prevstatus["state"]:
+self.cli.log.debug('Changed state: %s' % status["state"])
-if status.state in ('play', 'pause'):
-pos, length = map(float, status.time.split(':'))
+if status["state"] in ('play', 'pause'):
+pos, length = map(float, status["time"].split(':'))
if length == 0: length = lastfm.MAX_LEN
-if status.state == 'play':
+if status["state"] == 'play':
if song != self.prevsong or \
-self.prevstatus.state == 'stop':
+self.prevstatus["state"] == 'stop':
self.cli.log.info(u'New song: %s' % song)
if (self.prevsong and pos > self.sleep +
-FUZZ + int(status.xfade)) or \
+FUZZ + int(status["xfade"])) or \
(self.prevsong is None and
pos/length > lastfm.SUB_PERCENT or
pos > lastfm.SUB_SECONDS):
@@ -141,7 +149,7 @@
while True:
try:
if not self.mpd:
-self.mpd = mpdclient2.connect(**self.mpd_args)
+self.mpd = mpd_connect(self.mpd_args)
self.cli.log.info('Connected to MPD')
self.prevstatus = None
self.prevsong = None
@@ -151,12 +159,12 @@
self.played_enough = False
if self.mpd:
self.wake()
-except (EOFError, mpdclient2.socket.error):
+except (EOFError, socket.error):
if not failed:
self.cli.log.error("Can't connect or lost connection to MPD")
self.mpd = None
failed = True
-except MPDAuthError:
+except (MPDAuthError, mpd.CommandError):
if not failed:
self.cli.log.error("Can't read info from MPD (bad password?)")
failed = True
Bug#703054: upgrading to RC to warn about pending removal
severity 703053 serious severity 703054 serious thanks Since python-mpdclient has no upstream and no maintainer there is no justification to keeping it any longer. As such it will be removed at some point and lastmp and pygmy will be have unsatisfiable dependencies. Being maintained by the QA group as well the most likely action is to remove them as well unless someone steps up and takes care of them. The upgrade in severity serves as a warning to potential users. popcon atm: inst vote pygmy55 8 lastmp 50 29 python-mpdclient 170 37 All of the numbers are falling for at least two years. So if anyone finds an actual user, let them speak up! Helmut -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130513111859.ga10...@alf.mars
Bug#561089: ispell: segfaults on checking any file
Package: ispell Version: 3.1.20.0-7 Severity: grave Justification: renders package unusable When I try to spell check any file ispell simply segfaults. When I invoke it without arguments it prints the help text. To find out whether this was a recent regression I downgraded the package, but that did not help. This indicates that the cause may be unrelated to ispell. I also ran gdb on a core file from ispell, but the traceback did not reveal anything useful. Running strace in ispell shows that it uses curses to set up the terminal and then segfaults right after reading the file to be checked. Is there anything else I can do to help diagnose this? Helmut -- System Information: Architecture: amd64 (x86_64) Locale: LANG=C, LC_CTYPE=de_DE (charmap=ISO-8859-1) Shell: /bin/sh linked to /bin/dash Versions of packages ispell depends on: ii dictionaries-common 1.4.0 Common utilities for spelling dict ii dpkg 1.15.5.4 Debian package management system ii ingerman [ispell-dictiona 20091006-2 New German orthography dictionary ii install-info 4.13a.dfsg.1-5 Manage installed documentation in ii libc6 2.10.2-2 GNU C Library: Shared libraries ii libncurses5 5.7+20090803-2 shared libraries for terminal hand Versions of packages ispell recommends: ii wamerican [wordlist] 6-3American English dictionary words ii wngerman [wordlist] 20091006-2 New German orthography wordlist -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#561089: ispell: segfaults on checking any file
severity 561089 important thanks Thanks for all your work on this bug. I also observed that languages other than german work well with ispell. (Answer to some question in this thread.) On Mon, Dec 14, 2009 at 08:49:32PM +0100, Roland Rosenfeld wrote: > In the meantime I found out, that my recent 20091006-2 package did not > run munchlist over the mwl file (don't ask me why). After doing this > again, buildhash no longer has problems building the hash. > I just uploaded a fixed igerman98 20091006-3 to the archive. Thanks! > So I think that the severity of this bug report can be decreased. > Nevertheless it is IMHO a bug that buildhash can run into such a > trouble that ispell segfaults when it uses the dictionaries... Done. Helmut -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#569227: ncurses-base: break handling of ctrl-c in xterm and rxvt using bash
Package: ncurses-base Version: 5.7+20090803-2 Severity: critical Justification: breaks unrelated software ctrl-c does no longer cause SIGINT. Debugging this issue: When I create a new xterm (or rxvt) as a fork from my windowmanager (awesome) the terminal shows the broken behaviour. I have still some xterms that are not affected open. Those were started in 2009. Using strace on both working and a broken xterm show that both send \3 to the terminal fd when I press ctrl-c. stty on a working xterm looks like: speed 38400 baud; line = 0; On a broken xterm it looks like: speed 38400 baud; line = 0; -brkint -imaxbel Using stty brkint imaxbel or stty sane does not solve the issue. Also stty intr ^C does not help. Starting a xterm from an working xterm results in a working xterm. Starting a xterm from a broken xterm results in a broken xterm. Starting a xterm, starting vim within it and then doing :!xterm produces a working xterm. Saving the environment of a working xterm, loading it in a broken xterm and then starting a new xterm results in a broken xterm. Do you have any other ideas for debugging the issue? If you feel that I have assigned the bug report to the wrong package, please reassign it to the correct package. Helmut -- System Information: Debian Release: squeeze/sid APT prefers testing APT policy: (500, 'testing') Architecture: i386 (i686) Kernel: Linux 2.6.24.3 Locale: LANG=C, LC_CTYPE=de_DE (charmap=ISO-8859-1) Shell: /bin/sh linked to /bin/dash Versions of packages ncurses-base depends on: ii libncurses5 5.7+20090803-2 shared libraries for terminal hand Other packages of interest: ii bash 4.1-1 The GNU Bourne Again SHell ii rxvt 1:2.6.4-14 VT102 terminal emulator for the X Window System ii xterm 253-1 X terminal emulator ncurses-base recommends no packages. ncurses-base suggests no packages. -- no debconf information -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#569227: ncurses-base: break handling of ctrl-c in xterm and rxvt using bash
Hi Sven, thanks for your very quick reply. On Wed, Feb 10, 2010 at 11:26:07PM +0100, Sven Joachim wrote: > > I have still some xterms that are not affected open. Those were started > > in 2009. > Before or after you upgraded ncurses-base? That package has not been > touched for more than five months. Unfortunately I don't have any idea when I started the last working xterm. But I probably restarted some xterm within the past four months, so we can assume that ncurses-base is not the causing package. > > On a broken xterm it looks like: > > speed 38400 baud; line = 0; > > -brkint -imaxbel > > Here it looks exactly the same as in your "broken" xterm, and ^C works > fine anyway. Ok. This information is of no help then. > > Using stty brkint imaxbel or stty sane does not solve the issue. Also > > stty intr ^C does not help. > > stty sane should remove the "-brkint -imaxbel". Does it? It does. > > Do you have any other ideas for debugging the issue? > > Not really, but you could send the output of 'env' and Uhm. I don't really like to show my complete environment. I therefore give a diff from working to broken: -SHLVL=5 +SHLVL=20 -WINDOWID=20971535 +WINDOWID=12582927 -XTERM_LOCALE=de_DE +XTERM_LOCALE=C Here are some variables that might be of interest: LANG=C LANGUAGE=C LC_CTYPE=de_DE (no other LC_* is set) TERM=xterm WINDOWPATH=7 XTERM_SHELL=/bin/bash XTERM_VERSION=XTerm(253) > 'xrdb -query | grep -i xterm'. Do you see any differences between > working and broken xterms? The xrdb -query | grep -i xterm command has no output at all for me. > > If you feel that I have assigned the bug report to the wrong package, > > please reassign it to the correct package. > > I feel it is assigned to the wrong package, but I have no idea what the > right package could be. Actually I didn't feel it was the right package either. However packages like xterm, rxvt or bash seemed even more wrong. > > Kernel: Linux 2.6.24.3 > > Ever thought of upgrading this two years old, totally unsupported kernel? Yes, it is on my todo list for about a year. Unfortunately I really rely on that system (well I shouldn't be running testing then ;-). Additionally I experience bugs that cause data loss on reboots, so I avoid the latter as hard as possible. Helmut PS: I will be unable to read mail for the extended weekend starting in 14 hours. -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#569227: ncurses-base: break handling of ctrl-c in xterm and rxvt using bash
Hi Joachim, On Wed, Feb 10, 2010 at 11:26:07PM +0100, Sven Joachim wrote: > Ever thought of upgrading this two years old, totally unsupported kernel? I followed your advise and upgraded my kernel to vanilla 2.6.32.8. The problem persists in all detail (only that I lost all my "old" xterms). On the other hand the kernel upgrade only worsened the situation in many other aspects (#545163, #569314, ...). Do you have any other ideas on how to debug this problem? Helmut -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20100216074612.ga23...@alf.mars
Re: Bug#569227: ncurses-base: break handling of ctrl-c in xterm and rxvt using bash
reassign 569227 bash thanks Hi Thomas, On Tue, Feb 16, 2010 at 06:42:27AM -0500, Thomas Dickey wrote: > Your shell is not likely using the parts of ncurses which would change > the terminal modes. For instance, bash uses only the termcap interface. > I'd look at the shell. I finally found some time to verify this. Two observations: When I start an xterm and instantly hit ctrl-c I will see some ^C^C^C until the bash is fully loaded. Afterwards ctrl-c no longer produces any reaction. Invoking a zsh from that bash gives me a shell, that handles ctrl-c as it should. My conclusion is that this must be a fault of bash. Unfortunately downgrading bash to 4.0-4 did not solve the problem. Helmut -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20100408200025.ga13...@alf.mars
Re: Bug#569227: ncurses-base: break handling of ctrl-c in xterm and rxvt using bash
I can no longer reproduce the issue at all after upgrading various packages including Xorg. Helmut -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20100718155124.ga8...@alf.mars
Bug#589570: error: Unable to migrate to dependency based boot sequencing.
Package: timidity Version: 2.13.2-39 Severity: important # dpkg-reconfigure sysv-rc ... error: Unable to migrate to dependency based boot sequencing. error: Problems detected: package timidity left obsolete init.d script behind ... # Helmut -- System Information: Debian Release: squeeze/sid APT prefers unstable APT policy: (500, 'unstable'), (500, 'testing'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 2.6.33.2 (SMP w/2 CPU cores) Locale: LANG=C, LC_CTYPE=de_DE (charmap=ISO-8859-1) Shell: /bin/sh linked to /bin/dash Versions of packages timidity depends on: ii libasound2 1.0.23-1 shared library for ALSA applicatio ii libaudio2 1.9.2-3 Network Audio System - shared libr ii libc6 2.11.2-2 Embedded GNU C Library: Shared lib ii libesd00.2.41-7 Enlightened Sound Daemon - Shared ii libflac8 1.2.1-2+b1Free Lossless Audio Codec - runtim ii libice62:1.0.6-1 X11 Inter-Client Exchange library ii libjack0 1:0.118+svn3796-7 JACK Audio Connection Kit (librari ii libncurses55.7+20100313-2shared libraries for terminal hand ii libogg01.2.0~dfsg-1 Ogg bitstream library ii libpng12-0 1.2.44-1 PNG library - runtime ii libsm6 2:1.1.1-1 X11 Session Management library ii libvorbis0a1.3.1-1 The Vorbis General Audio Compressi ii libvorbisenc2 1.3.1-1 The Vorbis General Audio Compressi ii libx11-6 2:1.3.3-3 X11 client-side library ii libxaw72:1.0.7-1 X11 Athena Widget library ii libxext6 2:1.1.2-1 X11 miscellaneous extension librar ii libxmu62:1.0.5-1 X11 miscellaneous utility library ii libxt6 1:1.0.7-1 X11 toolkit intrinsics library ii lsb-base 3.2-23.1 Linux Standard Base 3.2 init scrip ii zlib1g 1:1.2.3.4.dfsg-3 compression library - runtime Versions of packages timidity recommends: ii freepats 20060219-1 Free patch set for MIDI audio synt pn timidity-daemon(no description available) Versions of packages timidity suggests: pn fluid-soundfont-gm (no description available) pn fluid-soundfont-gs (no description available) pn pmidi (no description available) -- no debconf information -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20100718184442.ga7...@alf.mars
Bug#589570: error: Unable to migrate to dependency based boot sequencing.
Hi René, On Tue, Jul 20, 2010 at 11:29:27AM +0200, Rene Engelhard wrote: > I bet you removed timidity and didn't purge it. This is expected behaviour > because /etc/init.d/* are conffiles. You are wrong. $ dpkg -l timidity ... ii timidity 2.13.2-39 Software sound renderer (MIDI sequencer, MOD $ > How is this a bug in timidity? Since your assumption was wrong, this question has vanished. Helmut -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20100720115507.gc20...@alf.mars
Bug#507617: vim-latexsuite: latex-suite/packages/ngerman.vim should let g:TeX_package_ngerman = ...
Package: vim-latexsuite Version: 20060325-5 Severity: normal In /usr/share/vim/addons/ftplugin/latex-suite/packages there are files for latex packages and these files seem to let g:TeX_package_ be something. However ngerman.vim does not let such a value. So when I use ngerman and do "proof" I am presented with a warning message. This is really annoying for users of the ngerman package. Helmut -- Package-specific info: Vim related packages installed on this system: - vim-latexsuite - vim-runtime -- System Information: Debian Release: lenny/sid APT prefers unstable APT policy: (500, 'unstable'), (500, 'testing'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 2.6.23.14 (SMP w/2 CPU cores) Locale: LANG=C, LC_CTYPE=de_DE (charmap=ISO-8859-1) Shell: /bin/sh linked to /bin/dash Versions of packages vim-latexsuite depends on: ii python 2.5.2-3 An interactive high-level object-o ii vim 2:7.2.049-2 Vi IMproved - enhanced vi editor ii vim-common 2:7.2.049-2 Vi IMproved - Common files Versions of packages vim-latexsuite recommends: ii texlive-base-bin 2007.dfsg.2-4 TeX Live: Essential binaries ii vim-addon-manager 0.4 manager of addons for the Vim edit Versions of packages vim-latexsuite suggests: ii ghostscript [postscript- 8.62.dfsg.1-3.1 The GPL Ghostscript PostScript/PDF ii gv [postscript-viewer] 1:3.6.5-2 PostScript and PDF viewer for X ii xpdf 3.02-1.4Portable Document Format (PDF) sui ii xpdf-reader [pdf-viewer] 3.02-1.4Portable Document Format (PDF) sui ii xpdf-utils [pdf-viewer] 3.02-1.4Portable Document Format (PDF) sui -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#441024: eukleides: euk2eps depends on texlive-pstricks, but only texlive is suggested
Package: eukleides Version: 1.0.3-1 Severity: minor Tags: patch The package suggests texlive. However texlive just isn't enough. It should also suggest texlive-pstricks for euk2eps requires that package. Helmut -- System Information: Debian Release: lenny/sid APT prefers unstable APT policy: (500, 'unstable'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 2.6.20.1 (SMP w/2 CPU cores) Locale: LANG=C, LC_CTYPE=de_DE (charmap=ISO-8859-1) Shell: /bin/sh linked to /bin/dash Versions of packages eukleides depends on: ii libc6 2.6.1-2GNU C Library: Shared libraries eukleides recommends no packages. -- no debconf information signature.asc Description: Digital signature
Bug#442147: xmp: manpage has two explanations for -l
Package: xmp Version: 2.0.4d-12+b1 Severity: minor man xmp states: -l, --load-only Exit after loading the module&. -l, --loop Enable module looping and backward pattern jumps. How can -l stand for two contradicting options? I suggest removing the first in accordance with xmp -h. Helmut -- System Information: Debian Release: lenny/sid APT prefers unstable APT policy: (500, 'unstable'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 2.6.20.1 (SMP w/2 CPU cores) Locale: LANG=C, LC_CTYPE=de_DE (charmap=ISO-8859-1) Shell: /bin/sh linked to /bin/dash Versions of packages xmp depends on: ii libc6 2.6.1-3 GNU C Library: Shared libraries ii libxmp2 2.0.4d-12+b1 Shared library files for xmp, xxmp ii xmp-common 2.0.4d-12+b1 Common files for xmp, xxmp and the xmp recommends no packages. -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#825002: sgml-base: broken super catalog after removing docutils-common
Package: sgml-base Version: 1.26+nmu2 Severity: serious Justification: breaks reverse dependencies like openjade Owner: helm...@debian.org User: helm...@debian.org Usertags: rebootstrap In dpkg 1.17.0, trigger processing was changed. Starting with that version, dpkg no longer activates triggers for conffiles on package removal (because conffiles are only removed during purge). Since that version the following sequence in a pristine sid chroot leaves a broken super catalog: apt-get install docutils-common apt-get remove docutils-common The latter invocation does not trigger sgml-base. Thus docutils-common remains in the super catalog and e.g. openjade fails. After discussion this with Guillem Jover, we concluded that this particular use of sgml-base was not intentional, so dpkg will not revert to the 1.16.10 behaviour. Guillem Jover suggested that sgml-base should provide a conffile-independent trigger (e.g. "update-catalog") and that packages providing catalogs should invoke it explicitly. Implementing this implies changing sgml-base, debhelper and reuploading 24 source packages shipping catalogs. A simpler approach is to trigger on the referenced files. For all relevant cases, they reside in /usr/share/sgml or /usr/share/xml. Declaring interest on both locations means triggering on 70 binary packages (i.e. 46 excess). This approach practically fixes the bug with one upload. I intend to pursue a combination of the above: Fix the immediate bug with the second method and update dh_installcatalogs to issue the trigger but skip updating those 24 source packages. I intend to upload the attached patch after gaining more confidence in not causing more breakage. Feedback welcome. Helmut diff --minimal -Nru sgml-base-1.27/debian/changelog sgml-base-1.28/debian/changelog --- sgml-base-1.27/debian/changelog 2016-05-17 04:19:42.0 +0200 +++ sgml-base-1.28/debian/changelog 2016-05-22 12:21:57.0 +0200 @@ -1,3 +1,12 @@ +sgml-base (1.28) unstable; urgency=medium + + * QA upload. + * Fix missing catalog update introduced in dpkg 1.17.0. (Closes: #-1) ++ Declare trigger interest on /usr/share/sgml, /usr/share/xml and + update-catalog (for pending debhelper change). + + -- Helmut Grohne Sat, 21 May 2016 23:46:45 +0200 + sgml-base (1.27) unstable; urgency=medium * QA upload. diff --minimal -Nru sgml-base-1.27/debian/sgml-base.triggers sgml-base-1.28/debian/sgml-base.triggers --- sgml-base-1.27/debian/sgml-base.triggers2012-05-28 13:58:23.0 +0200 +++ sgml-base-1.28/debian/sgml-base.triggers2016-05-22 07:58:48.0 +0200 @@ -1 +1,4 @@ +interest update-catalog interest /etc/sgml +interest /usr/share/sgml +interest /usr/share/xml
Bug#825002: sgml-base: broken super catalog after removing docutils-common
Control: clone -1 -2
Control: reassign -2 debhelper
Control: retitle -2 dh_installcatalogs: please activate sgml-base explicitly
Control: severity -2 wishlist
Control: block -2 by -1
Control: noowner -2
On Sun, May 22, 2016 at 12:27:32PM +0200, Helmut Grohne wrote:
> In dpkg 1.17.0, trigger processing was changed. Starting with that
> version, dpkg no longer activates triggers for conffiles on package
> removal (because conffiles are only removed during purge). Since that
> version the following sequence in a pristine sid chroot leaves a broken
> super catalog:
>
> apt-get install docutils-common
> apt-get remove docutils-common
>
> The latter invocation does not trigger sgml-base. Thus docutils-common
> remains in the super catalog and e.g. openjade fails.
>
> After discussion this with Guillem Jover, we concluded that this
> particular use of sgml-base was not intentional, so dpkg will not revert
> to the 1.16.10 behaviour.
>
> Guillem Jover suggested that sgml-base should provide a
> conffile-independent trigger (e.g. "update-catalog") and that packages
> providing catalogs should invoke it explicitly. Implementing this
> implies changing sgml-base, debhelper and reuploading 24 source
> packages shipping catalogs.
You can find the implementation for debhelper attached. It basically
consists of adding an autotrigger invocation and bumps the sgml-base
version.
In addition I am removing the transition script as it is now part of
both wheezy and jessie. Did we ever support skipping two releases?
This is a request for review and should not be included in debhelper
before the relevant sgml-base changes have been uploaded.
Helmut
diff --minimal -Nru debhelper-9.20160403/autoscripts/preinst-sgmlcatalog
debhelper-9.20160403+nmu1/autoscripts/preinst-sgmlcatalog
--- debhelper-9.20160403/autoscripts/preinst-sgmlcatalog2014-11-16
16:43:09.0 +0100
+++ debhelper-9.20160403+nmu1/autoscripts/preinst-sgmlcatalog 1970-01-01
01:00:00.0 +0100
@@ -1,8 +0,0 @@
-if test -f #CENTRALCAT# -a "(" "$1" = "upgrade" -o "$1" = "install" -a -n "$2"
")" &&
- ! dpkg-query -S #CENTRALCAT# >/dev/null 2>&1; then
- # If the dpkg-query command returns non-zero, the central catalog is
- # not owned by any package. This is due to an old behaviour of
- # debhelper. Now that file becomes a conffile. In order to avoid a
- # question during installation, we remove the old non-conffile.
- mv #CENTRALCAT# #CENTRALCAT#.old
-fi
diff --minimal -Nru debhelper-9.20160403/debian/changelog
debhelper-9.20160403+nmu1/debian/changelog
--- debhelper-9.20160403/debian/changelog 2016-04-03 10:56:12.0
+0200
+++ debhelper-9.20160403+nmu1/debian/changelog 2016-05-22 12:47:08.0
+0200
@@ -1,3 +1,13 @@
+debhelper (9.20160403+nmu1) UNRELEASED; urgency=medium
+
+ * Non-maintainer upload.
+ * dh_installcatalogs: Explicitly trigger update-catalog since
+dpkg does not trigger conffiles on package removal.
+ * dh_installcatalogs: Drop autoscript for transition that completed in
+wheezy.
+
+ -- Helmut Grohne Sun, 22 May 2016 12:35:27 +0200
+
debhelper (9.20160403) unstable; urgency=medium
* d/control: Requre dh-autoreconf (>= 12) to ensure
diff --minimal -Nru debhelper-9.20160403/dh_installcatalogs
debhelper-9.20160403+nmu1/dh_installcatalogs
--- debhelper-9.20160403/dh_installcatalogs 2016-01-14 20:49:48.0
+0100
+++ debhelper-9.20160403+nmu1/dh_installcatalogs2016-05-22
12:46:36.0 +0200
@@ -10,7 +10,7 @@
use warnings;
use Debian::Debhelper::Dh_Lib;
-my $sgmlbasever = "1.26+nmu2";
+my $sgmlbasever = "1.28";
=head1 SYNOPSIS
@@ -108,8 +108,7 @@
close CENTRALCAT;
if (! $dh{NOSCRIPTS}) {
- autoscript($package, "preinst", "preinst-sgmlcatalog",
- "s%#CENTRALCAT#%$centralcat%g;");
+ autotrigger($package, "activate", "update-catalog");
autoscript($package, "postrm", "postrm-sgmlcatalog",
"s%#CENTRALCAT#%$centralcat%g;");
}
Bug#837877: bogofilter FTCBFS: fails to pass --host to ./configure
Source: bogofilter
Version: 1.2.4+dfsg1-7
Tags: patch
User: helm...@debian.org
Usertags: rebootstrap
bogofilter fails to cross build from source, because it doesn't pass
--host to configure and thus uses the build architecture toolchain.
Ultimately, the failure occurs when it misses required dependencies. I
am proposing to simply switch the explicit configure invocations to
dh_auto_configure which knows the right flags (including --host) to
pass. After applying the attached patch, bogofilter cross builds just
fine.
Helmut
diff --minimal -Nru bogofilter-1.2.4+dfsg1/debian/changelog
bogofilter-1.2.4+dfsg1/debian/changelog
--- bogofilter-1.2.4+dfsg1/debian/changelog 2016-07-28 17:37:33.0
+0200
+++ bogofilter-1.2.4+dfsg1/debian/changelog 2016-09-15 05:46:08.0
+0200
@@ -1,3 +1,10 @@
+bogofilter (1.2.4+dfsg1-7.1) UNRELEASED; urgency=medium
+
+ * Non-maintainer upload.
+ * Fix FTCBFS: Let dh_auto_configure pass the right flags. (Closes: #-1)
+
+ -- Helmut Grohne Thu, 15 Sep 2016 05:45:40 +0200
+
bogofilter (1.2.4+dfsg1-7) unstable; urgency=medium
* QA upload. (Closes: #832291). Thanks Andreas Beckmann.
diff --minimal -Nru bogofilter-1.2.4+dfsg1/debian/rules
bogofilter-1.2.4+dfsg1/debian/rules
--- bogofilter-1.2.4+dfsg1/debian/rules 2016-05-15 15:35:55.0 +0200
+++ bogofilter-1.2.4+dfsg1/debian/rules 2016-09-15 05:55:09.0 +0200
@@ -18,21 +18,17 @@
dh $@ --with autotools-dev
override_dh_auto_configure:
- mkdir -p $(CURDIR)/obj-doc
- cd $(CURDIR)/obj-doc && ../configure
- mkdir $(CURDIR)/obj-db $(CURDIR)/obj-sqlite
- cd $(CURDIR)/obj-db && ../configure --with-database=db \
- --prefix=/usr --mandir=\$${prefix}/share/man --sysconfdir=/etc \
+ dh_auto_configure --builddirectory=$(CURDIR)/obj-doc
+ dh_auto_configure --builddirectory=$(CURDIR)/obj-db --
--with-database=db \
AWK=awk $(shell dpkg-buildflags --export=configure)
- cd $(CURDIR)/obj-sqlite && ../configure --with-database=sqlite \
- --program-suffix=-sqlite --prefix=/usr
--mandir=\$${prefix}/share/man \
- --sysconfdir=/etc AWK=awk $(shell dpkg-buildflags
--export=configure) && \
- sed -i 's/^INTEGRITY_TESTS.*/INTEGRITY_TESTS=t.lock1/'
src/tests/Makefile
+ dh_auto_configure --builddirectory=$(CURDIR)/obj-sqlite --
--with-database=sqlite \
+ --program-suffix=-sqlite \
+ AWK=awk $(shell dpkg-buildflags --export=configure)
+ sed -i 's/^INTEGRITY_TESTS.*/INTEGRITY_TESTS=t.lock1/'
$(CURDIR)/obj-sqlite/src/tests/Makefile
if [ $(DO_TOKYOCABINET) = yes ]; then \
- mkdir $(CURDIR)/obj-tokyocabinet; \
- cd $(CURDIR)/obj-tokyocabinet && ../configure \
+ dh_auto_configure --builddirectory=$(CURDIR)/obj-tokyocabinet
-- \
--with-database=tokyocabinet
--program-suffix=-tokyocabinet \
- --prefix=/usr --mandir=\$${prefix}/share/man
--sysconfdir=/etc AWK=awk \
+ AWK=awk \
$(shell dpkg-buildflags --export=configure); \
fi
Bug#838205: psutils FTCBFS: uses build architecture compiler
Source: psutils Version: 1.17.dfsg-3 Tags: patch User: helm...@debian.org Usertags: rebootstrap psutils fails to cross build from source, because it uses the build architecture compiler. Even though cdbs passes a suitable CC environment variable, the upstream build system chooses to override that. After passing CC as a command line variable, the cross build succeeds. Please consider applying the attached patch. Helmut diff --minimal -Nru psutils-1.17.dfsg/debian/changelog psutils-1.17.dfsg/debian/changelog --- psutils-1.17.dfsg/debian/changelog 2016-08-27 03:59:27.0 +0200 +++ psutils-1.17.dfsg/debian/changelog 2016-09-18 14:41:23.0 +0200 @@ -1,3 +1,10 @@ +psutils (1.17.dfsg-3.1) UNRELEASED; urgency=medium + + * Non-maintainer upload. + * Fix FTCBFS: Pass CC as a cmdline option to make. (Closes: #-1) + + -- Helmut Grohne Sun, 18 Sep 2016 14:41:00 +0200 + psutils (1.17.dfsg-3) unstable; urgency=medium * Orphan package diff --minimal -Nru psutils-1.17.dfsg/debian/rules psutils-1.17.dfsg/debian/rules --- psutils-1.17.dfsg/debian/rules 2016-08-27 03:59:27.0 +0200 +++ psutils-1.17.dfsg/debian/rules 2016-09-18 14:40:56.0 +0200 @@ -4,6 +4,10 @@ export DEB_BUILD_MAINT_OPTIONS = hardening=+all DPKG_EXPORT_BUILDFLAGS = 1 include /usr/share/dpkg/buildflags.mk +include /usr/share/dpkg/architecture.mk +ifeq ($(origin CC),default) +CC = $(DEB_HOST_GNU_TYPE)-gcc +endif # Include cdbs rules files. include /usr/share/cdbs/1/class/makefile.mk @@ -11,7 +15,7 @@ CURDIR := $(shell pwd) DEB_MAKE_CLEAN_TARGET = realclean -DEB_MAKE_BUILD_TARGET = prefix=/usr all +DEB_MAKE_BUILD_TARGET = prefix=/usr CC=$(CC) all DEB_MAKE_INSTALL_TARGET = prefix=$(CURDIR)/debian/psutils/usr \ MANDIR=$(CURDIR)/debian/psutils/usr/share/man/man1 \ install
Bug#846437: linux-atm: add libfl-dev to Build-Depends
Source: linux-atm Severity: important User: helm...@debian.org Usertags: libfldep linux-atm will soon fail to build from source, beause flex drops its dependency on libfl-dev. Since linux-atm uses parts of libfl-dev (e.g. libl.a, libfl.a or FlexLexer.h), it should add libfl-dev to its Build-Depends. This change was previously announced[1] to debian-devel in accordance with DevRef 7.1.1. Please add the missing dependency. Helmut [1] https://lists.debian.org/debian-devel/2016/03/msg00162.html
Bug#846448: splint: add libfl-dev to Build-Depends
Source: splint Severity: important User: helm...@debian.org Usertags: libfldep splint will soon fail to build from source, beause flex drops its dependency on libfl-dev. Since splint uses parts of libfl-dev (e.g. libl.a, libfl.a or FlexLexer.h), it should add libfl-dev to its Build-Depends. This change was previously announced[1] to debian-devel in accordance with DevRef 7.1.1. Please add the missing dependency. Helmut [1] https://lists.debian.org/debian-devel/2016/03/msg00162.html
Bug#851294: fbset FTCBFS: uses build architecture compiler
Source: fbset Version: 2.1-28 Tags: patch User: helm...@debian.org Usertags: rebootstrap fbset fails to cross build from source, because it uses the build architecture compiler. Using dh_auto_build makes it cross build successfully, because debhelper places cross compilers into CC and CXX. Please consider applying the attached patch. Helmut diff --minimal -Nru fbset-2.1/debian/changelog fbset-2.1/debian/changelog --- fbset-2.1/debian/changelog 2014-03-02 20:55:01.0 +0100 +++ fbset-2.1/debian/changelog 2017-01-13 19:45:01.0 +0100 @@ -1,3 +1,10 @@ +fbset (2.1-28.1) UNRELEASED; urgency=medium + + * Non-maintainer upload. + * Fix FTCBFS: Let dh_auto_build pass cross compilers. Closes: #-1. + + -- Helmut Grohne Fri, 13 Jan 2017 19:45:01 +0100 + fbset (2.1-28) unstable; urgency=low * Orphan package, set maintainer to Debian QA Group. diff --minimal -Nru fbset-2.1/debian/rules fbset-2.1/debian/rules --- fbset-2.1/debian/rules 2014-03-02 20:39:24.0 +0100 +++ fbset-2.1/debian/rules 2017-01-13 19:44:58.0 +0100 @@ -26,12 +26,12 @@ mkdir -p build-udeb ln -f Makefile build-udeb/Makefile - $(MAKE) -C build-udeb $(DIR_FLAGS) CPPFLAGS="-I.. $(CPPFLAGS)" \ + dh_auto_build --builddirectory=build-udeb -- $(DIR_FLAGS) CPPFLAGS="-I.. $(CPPFLAGS)" \ CFLAGS="$(CFLAGS_UDEB)" LDFLAGS="$(LDFLAGS)" mkdir -p build-tree ln -f Makefile build-tree/Makefile - $(MAKE) -C build-tree $(DIR_FLAGS) CPPFLAGS="-I.. $(CPPFLAGS)" \ + dh_auto_build --builddirectory=build-tree -- $(DIR_FLAGS) CPPFLAGS="-I.. $(CPPFLAGS)" \ CFLAGS="$(CFLAGS)" LDFLAGS="$(LDFLAGS)" build: build-indep build-arch
Bug#852366: aumix FTCBFS: uses the build architecture pkg-config
Source: aumix Version: 2.9.1-4 Tags: patch User: helm...@debian.org Usertags: rebootstrap aumix fails to cross build from source, because it uses the build architecture pkg-config and thus fails finding gtk. By using the PKG_PROG_PKG_CONFIG macro, $ac_tool_prefix is properly considered and the cross build succeeds. Please consider applying the attached patch. Helmut diff --minimal -Nru aumix-2.9.1/debian/changelog aumix-2.9.1/debian/changelog --- aumix-2.9.1/debian/changelog2016-07-19 23:15:12.0 +0200 +++ aumix-2.9.1/debian/changelog2017-01-23 21:50:38.0 +0100 @@ -1,3 +1,10 @@ +aumix (2.9.1-4.1) UNRELEASED; urgency=medium + + * Non-maintainer upload. + * Fix FTCBFS: 19_cross.patch (Closes: #-1) + + -- Helmut Grohne Mon, 23 Jan 2017 21:50:38 +0100 + aumix (2.9.1-4) unstable; urgency=medium * QA upload. diff --minimal -Nru aumix-2.9.1/debian/patches/19_cross.patch aumix-2.9.1/debian/patches/19_cross.patch --- aumix-2.9.1/debian/patches/19_cross.patch 1970-01-01 01:00:00.0 +0100 +++ aumix-2.9.1/debian/patches/19_cross.patch 2017-01-23 21:50:38.0 +0100 @@ -0,0 +1,27 @@ +From: Helmut Grohne +Subject: use triplet-prefixed pkg-config + +Index: aumix-2.9.1/configure.ac +=== +--- aumix-2.9.1.orig/configure.ac aumix-2.9.1/configure.ac +@@ -155,8 +155,8 @@ + gtk_must=off, gtk_must=on) + if test $gtk_must = on; then + dnl from gftp +- AC_CHECK_PROG(PKG_CONFIG, pkg-config, yes, no) +- if test "$PKG_CONFIG" = "no"; then ++ PKG_PROG_PKG_CONFIG ++ if test -z "$PKG_CONFIG"; then + echo "pkg-config not found--compiling without GTK+ 2.0." ; + else + echo "pkg-config found--compiling with GTK+ 2.0." ; +@@ -181,7 +181,7 @@ + else + AC_MSG_RESULT([Compiling without GTK+ 2.0.]) + fi +-AM_CONDITIONAL(GTK, test "$gtk_must" = on && test "$PKG_CONFIG" != "no") ++AM_CONDITIONAL(GTK, test "$gtk_must" = on && test -n "$PKG_CONFIG") + + AC_MSG_CHECKING([whether dummy mixer is requested]) + AC_ARG_ENABLE(dummy-mixer, diff --minimal -Nru aumix-2.9.1/debian/patches/series aumix-2.9.1/debian/patches/series --- aumix-2.9.1/debian/patches/series 2014-04-30 01:07:31.0 +0200 +++ aumix-2.9.1/debian/patches/series 2017-01-23 21:49:12.0 +0100 @@ -7,3 +7,4 @@ 16_potfiles.patch 17_zh-tw-po.patch 18_ncursesw.patch +19_cross.patch
Bug#852741: wvstreams FTCBFS: uses build architecture pkg-config, fails to detect required libraries
Source: wvstreams Version: 4.6.1-11 Tags: patch User: helm...@debian.org Usertags: rebootstrap wvstreams fails to cross build from source, because it uses the build architecture pkg-config. It thus fails finding required libraries. Using a triplet-prefixed pkg-config fixes the issue and makes cross builds succeed. Please consider applying the attached patch. Helmut diff --minimal -Nru wvstreams-4.6.1/debian/changelog wvstreams-4.6.1/debian/changelog --- wvstreams-4.6.1/debian/changelog2016-12-17 11:02:02.0 +0100 +++ wvstreams-4.6.1/debian/changelog2017-01-26 21:44:34.0 +0100 @@ -1,3 +1,10 @@ +wvstreams (4.6.1-11.1) UNRELEASED; urgency=medium + + * Non-maintainer upload. + * Fix FTCBFS: 12_cross.patch: Use triplet-prefixed pkg-config (Closes: #-1). + + -- Helmut Grohne Thu, 26 Jan 2017 21:44:34 +0100 + wvstreams (4.6.1-11) unstable; urgency=medium * QA upload. diff --minimal -Nru wvstreams-4.6.1/debian/patches/12_cross.diff wvstreams-4.6.1/debian/patches/12_cross.diff --- wvstreams-4.6.1/debian/patches/12_cross.diff1970-01-01 01:00:00.0 +0100 +++ wvstreams-4.6.1/debian/patches/12_cross.diff2017-01-26 21:44:34.0 +0100 @@ -0,0 +1,53 @@ +From: Helmut Grohne +Subject: prefix pkg-config with $ac_tool_prefix to support cross compilation + +Index: wvstreams-4.6.1/configure.ac +=== +--- wvstreams-4.6.1.orig/configure.ac wvstreams-4.6.1/configure.ac +@@ -345,8 +345,8 @@ + LDFLAGS="$LDFLAGS_save" + + # Detect pkg-config +-AC_PATH_PROG([PKGCONFIG], [pkg-config], [no]) +-if test "$PKGCONFIG" = "no"; then ++PKG_PROG_PKG_CONFIG ++if test "x$PKG_CONFIG" = "x"; then + AC_MSG_WARN([pkg-config is not installed]) + fi + +@@ -370,10 +370,10 @@ + if test "$with_dbus" != "no"; then + if test "$with_dbus" = "" -o "$with_dbus" = "yes"; then + AC_MSG_CHECKING([Checking that D-Bus version greater than 1.2.14 installed.]) +-if pkg-config --atleast-version 1.2.14 dbus-1; then +- WV_APPEND(CPPFLAGS, [`pkg-config --cflags dbus-1`]) +-WV_APPEND(LDFLAGS, [`pkg-config --libs-only-L dbus-1`]) +- LIBS_DBUS=`pkg-config --libs-only-l dbus-1` ++if $PKG_CONFIG --atleast-version 1.2.14 dbus-1; then ++ WV_APPEND(CPPFLAGS, [`$PKG_CONFIG --cflags dbus-1`]) ++WV_APPEND(LDFLAGS, [`$PKG_CONFIG --libs-only-L dbus-1`]) ++ LIBS_DBUS=`$PKG_CONFIG --libs-only-l dbus-1` + AC_MSG_RESULT([yes]) + else + with_dbus=no +@@ -497,7 +497,7 @@ + CPPFLAGS_save="$CPPFLAGS" + LDFLAGS_save="$LDFLAGS" + LIBS_save="$LIBS" +-for wv_qtdir in $with_qt $QTDIR $QT_SEARCH_PATH $(pkg-config --variable=prefix qt-mt); do ++for wv_qtdir in $with_qt $QTDIR $QT_SEARCH_PATH $($PKG_CONFIG --variable=prefix qt-mt); do + eval wv_qtdir="$wv_qtdir" + CPPFLAGS="$CPPFLAGS_save -I$wv_qtdir/include -I$wv_qtdir/include/qt3" + LDFLAGS="$LDFLAGS_save -L$wv_qtdir/lib" +Index: wvstreams-4.6.1/autogen.sh +=== +--- wvstreams-4.6.1.orig/autogen.sh wvstreams-4.6.1/autogen.sh +@@ -1,5 +1,6 @@ + #!/bin/sh + set -e ++aclocal + autoheader + autoconf + (cd argp && autoconf) diff --minimal -Nru wvstreams-4.6.1/debian/patches/series wvstreams-4.6.1/debian/patches/series --- wvstreams-4.6.1/debian/patches/series 2016-07-11 13:41:28.0 +0200 +++ wvstreams-4.6.1/debian/patches/series 2017-01-26 21:43:10.0 +0100 @@ -9,3 +9,4 @@ 10_readline.diff 11_typos.diff gcc-6.patch +12_cross.diff
Bug#852768: ocrad FTCBFS: uses the build architecture C++ compiler
Source: ocrad
Version: 0.25-2
Tags: patch
User: helm...@debian.org
Usertags: rebootstrap
ocrad fails to cross build from source, because it uses the build
architecture C++ compiler. Simply passing a triplet-prefixed CXX
variable to configure makes the cross build succeed. Please consider
applying the attached patch.
Helmut
diff --minimal -Nru ocrad-0.25/debian/changelog ocrad-0.25/debian/changelog
--- ocrad-0.25/debian/changelog 2017-01-17 18:36:35.0 +0100
+++ ocrad-0.25/debian/changelog 2017-01-27 07:16:16.0 +0100
@@ -1,3 +1,10 @@
+ocrad (0.25-2.1) UNRELEASED; urgency=medium
+
+ * Non-maintainer upload.
+ * Fix FTCBFS: Pass triplet-prefixed CXX to configure (Closes: #-1).
+
+ -- Helmut Grohne Fri, 27 Jan 2017 07:16:16 +0100
+
ocrad (0.25-2) unstable; urgency=medium
* QA upload.
diff --minimal -Nru ocrad-0.25/debian/rules ocrad-0.25/debian/rules
--- ocrad-0.25/debian/rules 2016-07-12 11:52:24.0 +0200
+++ ocrad-0.25/debian/rules 2017-01-27 07:16:15.0 +0100
@@ -4,7 +4,11 @@
CPPFLAGS = -W -Wall $(shell dpkg-buildflags --get CPPFLAGS)
-D_FILE_OFFSET_BITS=64
LDFLAGS = $(shell dpkg-buildflags --get LDFLAGS)
+DEB_HOST_GNU_TYPE ?= $(shell dpkg-architectur -qDEB_HOST_GNU_TYPE)
DEB_HOST_MULTIARCH ?= $(shell dpkg-architecture -qDEB_HOST_MULTIARCH)
+ifeq ($(origin CXX),default)
+CXX = $(DEB_HOST_GNU_TYPE)-g++
+endif
.PHONY: build build-indep build-arch
build: build-indep build-arch
@@ -23,6 +27,7 @@
--libdir=\$${prefix}/lib/$(DEB_HOST_MULTIARCH) \
--mandir=\$${prefix}/share/man \
--infodir=\$${prefix}/share/info \
+ CXX="$(CXX)" \
CXXFLAGS="$(CXXFLAGS)" CPPFLAGS="$(CPPFLAGS)"
LDFLAGS="$(LDFLAGS)"
touch $(@)
Bug#853957: mpage FTCBFS: make install strips with the build architecture strip
Source: mpage Version: 2.5.6+dfsg-1 Tags: patch User: helm...@debian.org Usertags: rebootstrap mpage fails to cross build from source, because make install uses "install -s", which uses the build architecture strip and thus fails operating on host architecture objects. Stripping before dh_strip also prevents generation of a -dbgsym package. Not stripping makes both cross compilation an generation of a -dbgsym succeed. Please consider applying the attached patch after stretch is released. Helmut diff --minimal -Nru mpage-2.5.6+dfsg/debian/changelog mpage-2.5.6+dfsg/debian/changelog --- mpage-2.5.6+dfsg/debian/changelog 2015-10-19 03:36:01.0 +0200 +++ mpage-2.5.6+dfsg/debian/changelog 2017-02-02 10:11:49.0 +0100 @@ -1,3 +1,10 @@ +mpage (2.5.6+dfsg-1.1) UNRELEASED; urgency=medium + + * Non-maintainer upload. + * Fix FTCBFS: Do not strip during make install. (Closes: #-1) + + -- Helmut Grohne Thu, 02 Feb 2017 10:11:49 +0100 + mpage (2.5.6+dfsg-1) unstable; urgency=medium * QA upload. diff --minimal -Nru mpage-2.5.6+dfsg/debian/patches/50_no_strip.patch mpage-2.5.6+dfsg/debian/patches/50_no_strip.patch --- mpage-2.5.6+dfsg/debian/patches/50_no_strip.patch 1970-01-01 01:00:00.0 +0100 +++ mpage-2.5.6+dfsg/debian/patches/50_no_strip.patch 2017-02-02 10:11:49.0 +0100 @@ -0,0 +1,25 @@ +From: Helmut Grohne +Subject: allow substituting install + +This can be used to make stripping work with cross tools or to disable +stripping. + +Index: mpage-2.5.6+dfsg/Makefile +=== +--- mpage-2.5.6+dfsg.orig/Makefile mpage-2.5.6+dfsg/Makefile +@@ -25,6 +25,7 @@ + # See also system specific settings below + #CC=cc + CC=gcc ++INSTALL ?= install + + + # +@@ -182,5 +183,5 @@ + if [ ! -d $(DESTDIR)/$(LIBDIR)/mpage ] ; then mkdir -p $(DESTDIR)/$(LIBDIR)/mpage ; fi + if [ ! -d $(DESTDIR)/$(BINDIR) ] ; then mkdir -p $(DESTDIR)/$(BINDIR) ; fi + if [ ! -d $(DESTDIR)/$(MANDIR) ] ; then mkdir -p $(DESTDIR)/$(MANDIR) ; fi +- install -s -m 755 -o root -g root mpage$(E) $(DESTDIR)/usr/bin ++ $(INSTALL) -s -m 755 -o root -g root mpage$(E) $(DESTDIR)/usr/bin + -cp Encodings/* $(DESTDIR)/usr/lib/mpage diff --minimal -Nru mpage-2.5.6+dfsg/debian/patches/series mpage-2.5.6+dfsg/debian/patches/series --- mpage-2.5.6+dfsg/debian/patches/series 2015-10-17 20:38:29.0 +0200 +++ mpage-2.5.6+dfsg/debian/patches/series 2017-02-02 10:09:46.0 +0100 @@ -3,3 +3,4 @@ 20_bts416573_manpage_fixes.patch 30_bts443280_libdir_manpage.patch 40_add_GCC_hardening.patch +50_no_strip.patch diff --minimal -Nru mpage-2.5.6+dfsg/debian/rules mpage-2.5.6+dfsg/debian/rules --- mpage-2.5.6+dfsg/debian/rules 2015-10-17 23:41:59.0 +0200 +++ mpage-2.5.6+dfsg/debian/rules 2017-02-02 10:11:49.0 +0100 @@ -6,6 +6,10 @@ %: dh $@ +override_dh_auto_install: + # #844077 + dh_auto_install -- INSTALL='install --strip-program=true' + override_dh_install: dh_install chmod 0755 $(CURDIR)/debian/mpage/usr/share/doc/mpage/Contrib/mp \
Bug#853957: mpage FTCBFS: make install strips with the build architecture strip
On Thu, Feb 02, 2017 at 04:14:08PM +0200, Adrian Bunk wrote: > On Thu, Feb 02, 2017 at 01:59:40PM +0100, Helmut Grohne wrote: > >... > > +- install -s -m 755 -o root -g root mpage$(E) $(DESTDIR)/usr/bin > > ++ $(INSTALL) -s -m 755 -o root -g root mpage$(E) $(DESTDIR)/usr/bin > >... > > Wouldn't just removing the -s here be the oneliner change to fix the > whole bug? >From a Debian pov, yes. My reasoning for not just removing it here is that it seemed to be a deliberate upstream choice. Thus removing it would make the patch non-upstreamable. The indirection however should be less controversial. Helmut
Bug#861423: libstroke FTCBFS: configures for the build architecture
Source: libstroke Version: 0.5.1-8 Tags: patch User: helm...@debian.org Usertags: rebootstrap libstroke fails to cross build from source, because it configures for the build architecture. The attached patch explicitly passes the required --build and --host flags to configure. Alternatively, using dh_auto_configure they will be supplied automatically. After applying the patch, libstroke cross builds successfully. Please consider applying it after stretch is released. Helmut diff --minimal -Nru libstroke-0.5.1/debian/changelog libstroke-0.5.1/debian/changelog --- libstroke-0.5.1/debian/changelog2016-07-05 11:01:49.0 +0200 +++ libstroke-0.5.1/debian/changelog2017-04-28 22:07:43.0 +0200 @@ -1,3 +1,10 @@ +libstroke (0.5.1-8.1) UNRELEASED; urgency=medium + + * Non-maintainer upload. + * Fix FTCBFS: Pass --build and --host to ./configure (Closes: #-1). + + -- Helmut Grohne Fri, 28 Apr 2017 22:07:43 +0200 + libstroke (0.5.1-8) unstable; urgency=medium * QA upload. diff --minimal -Nru libstroke-0.5.1/debian/rules libstroke-0.5.1/debian/rules --- libstroke-0.5.1/debian/rules2016-07-05 09:43:17.0 +0200 +++ libstroke-0.5.1/debian/rules2017-04-28 22:07:41.0 +0200 @@ -1,5 +1,7 @@ #!/usr/bin/make -f +include /usr/share/dpkg/architecture.mk + build: build-stamp build-arch: build build-indep: build @@ -7,7 +9,7 @@ build-stamp: dh_testdir dh_autotools-dev_updateconfig - ./configure --prefix=/usr --x-includes=/usr/X11R6/include --x-libraries=/usr/X11R6/lib + ./configure --build=$(DEB_BUILD_GNU_TYPE) --host=$(DEB_HOST_GNU_TYPE) --prefix=/usr --x-includes=/usr/X11R6/include --x-libraries=/usr/X11R6/lib # Patch the generated libtool to avoid passing -rpath when linking, # and to explicitly link libraries against the libraries they
Bug#862196: mark xmltoman Multi-Arch: foreign
Package: xmltoman
Version: 0.4-4
Tags: patch
User: helm...@debian.org
Usertags: rebootstrap
Control: affects -1 + src:avahi src: src:seccure src:shairport-sync
src:syrep
The packages listed under affects cannot satisfy their cross build
dependencies, because their dependency on xmltoman is not satisfiable.
In general, Architecture: all packages can never satisfy cross build
dependencies unless marked Multi-Arch: foreign. In this case, such a
marking is reasonable, because xmltoman provides command line utilities,
whose output and interaction does not depend on the architecture of the
perl interpreter used to run xmltoman. Please consider applying the
attached patch after stretch is released.
Helmut
diff --minimal -Nru xmltoman-0.4/debian/changelog xmltoman-0.4/debian/changelog
--- xmltoman-0.4/debian/changelog 2017-01-16 19:08:47.0 +0100
+++ xmltoman-0.4/debian/changelog 2017-05-09 19:28:48.0 +0200
@@ -1,3 +1,10 @@
+xmltoman (0.4-4.1) UNRELEASED; urgency=medium
+
+ * Non-maintainer upload.
+ * Mark xmltoman Multi-Arch: foreign. (Closes: #-1)
+
+ -- Helmut Grohne Tue, 09 May 2017 19:28:48 +0200
+
xmltoman (0.4-4) unstable; urgency=medium
* QA upload.
diff --minimal -Nru xmltoman-0.4/debian/control xmltoman-0.4/debian/control
--- xmltoman-0.4/debian/control 2017-01-16 19:08:47.0 +0100
+++ xmltoman-0.4/debian/control 2017-05-09 19:27:54.0 +0200
@@ -8,6 +8,7 @@
Package: xmltoman
Architecture: all
+Multi-Arch: foreign
Depends: ${misc:Depends}, libxml-parser-perl
Description: simple XML to man converter
xmltoman and xmlmantohtml are two small scripts to convert xml to man
Bug#869593: ccontrol FTBFS: error: conflicting types for 'find_extension'
Source: ccontrol
Version: 1.0-1
Severity: serious
User: helm...@debian.org
Usertags: rebootstrap
ccontrol fails to build from source in unstable amd64 (presumably due to
the gperf 3.1 upload):
| gcc -O3 -Wall -g -O2 -fdebug-prefix-map=/<>=.
-fstack-protector-strong -Wformat -Werror=format-security -Wdate-time
-D_FORTIFY_SOURCE=2 -Wl,-z,relro -DVERSION='"1.0 (David Gibson)"' -o ccontrol
ccontrol.c ccontrol-parse.c ccontrol-lock.c ccontrol-identify.c stdrusty.c
| ccontrol.c: In function 'write_string':
| ccontrol.c:27:2: warning: ignoring return value of 'write', declared with
attribute warn_unused_result [-Wunused-result]
| write(fd, str, strlen(str));
| ^~~
| ccontrol.c: In function 'main':
| ccontrol.c:325:2: warning: ignoring return value of 'getcwd', declared with
attribute warn_unused_result [-Wunused-result]
| getcwd(dirname, sizeof(dirname));
| ^~~~
| In file included from ccontrol-parse.c:40:0:
| keywords.gperf:56:1: error: conflicting types for 'find_keyword'
| In file included from ccontrol-parse.c:40:0:
| keywords.gperf:4:1: note: previous declaration of 'find_keyword' was here
| find_keyword (register const char *str, register unsigned int len);
| ^~~~
| In file included from ccontrol-identify.c:4:0:
| extensions.gperf:63:1: error: conflicting types for 'find_extension'
| extensions.gperf:6:1: note: previous declaration of 'find_extension' was here
| find_extension (register const char *str, register unsigned int len);
| ^~
| stdrusty.c: In function 'fatal':
| stdrusty.c:45:2: warning: ignoring return value of 'write', declared with
attribute warn_unused_result [-Wunused-result]
| write(STDERR_FILENO, "ccontrol error: ", strlen("ccontrol error: "));
| ^~~~
| stdrusty.c:46:2: warning: ignoring return value of 'write', declared with
attribute warn_unused_result [-Wunused-result]
| write(STDERR_FILENO, msg, strlen(msg));
| ^~
| stdrusty.c:50:3: warning: ignoring return value of 'write', declared with
attribute warn_unused_result [-Wunused-result]
|write(STDERR_FILENO, str, strlen(str));
|^~
| stdrusty.c:54:3: warning: ignoring return value of 'write', declared with
attribute warn_unused_result [-Wunused-result]
|write(STDERR_FILENO, ": ", strlen(": "));
|^~~~
| stdrusty.c:55:3: warning: ignoring return value of 'write', declared with
attribute warn_unused_result [-Wunused-result]
|write(STDERR_FILENO, str, strlen(str));
|^~
| stdrusty.c:57:2: warning: ignoring return value of 'write', declared with
attribute warn_unused_result [-Wunused-result]
| write(STDERR_FILENO, "\n", 1);
| ^
| Makefile:19: recipe for target 'ccontrol' failed
| make[1]: *** [ccontrol] Error 1
| rm ccontrol.1.xml
| make[1]: Leaving directory '/<>'
| debian/rules:39: recipe for target 'build-stamp' failed
| make: *** [build-stamp] Error 2
| dpkg-buildpackage: error: debian/rules build-arch gave error exit status 2
Helmut
Bug#870727: malaga FTCBFS: uses the build architecture pkg-config
Source: malaga Version: 7.12-7 Tags: patch User: helm...@debian.org Usertags: rebootstrap malaga fails to cross build from source, because it uses the build architecture pkg-config and thus fails finding gtk2 and glib2.0. After switching configure.in to use PKG_CHECK_MODULES, which considers $ac_tool_prefix, it fails building the documentation, because gtk2 and glib2.0 are only requested for the host architecture. Since cross builds typically skip building architecture-independent packages, moving the documentation build to -indep overrides makes the build proceed. make install tries to strip the resulting binaries with the build architecture strip. This not only breaks cross building, but also generating -dbgsym packages. It can be nullified by telling install to strip with /bin/true. The attached patch combines all of that and makes malaga cross build successfully. Please consider applying it. Helmut diff --minimal -Nru malaga-7.12/debian/changelog malaga-7.12/debian/changelog --- malaga-7.12/debian/changelog2016-04-07 11:51:00.0 +0200 +++ malaga-7.12/debian/changelog2017-08-04 13:28:11.0 +0200 @@ -1,3 +1,13 @@ +malaga (7.12-7.1) UNRELEASED; urgency=medium + + * Non-maintainer upload. + * Fix FTCBFS: Closes: #-1 ++ Add cross.patch. ++ Do not build the documentation during arch-only. ++ Do not strip during install. + + -- Helmut Grohne Fri, 04 Aug 2017 13:28:11 +0200 + malaga (7.12-7) unstable; urgency=low * QA upload. diff --minimal -Nru malaga-7.12/debian/control malaga-7.12/debian/control --- malaga-7.12/debian/control 2016-04-07 10:00:00.0 +0200 +++ malaga-7.12/debian/control 2017-08-04 13:22:23.0 +0200 @@ -2,7 +2,7 @@ Maintainer: Debian QA Group Section: misc Priority: optional -Build-Depends: libgtk2.0-dev, libglib2.0-dev, texinfo, texlive, libreadline-dev, debhelper (>= 9), autotools-dev, libcairo-dev, dpkg (>= 1.15.4) | install-info, dh-autoreconf +Build-Depends: libgtk2.0-dev, libglib2.0-dev, texinfo, texlive, libreadline-dev, debhelper (>= 9), autotools-dev, libcairo-dev, dpkg (>= 1.15.4) | install-info, dh-autoreconf, pkg-config Standards-Version: 3.9.2 Package: malaga-bin diff --minimal -Nru malaga-7.12/debian/patches/cross.diff malaga-7.12/debian/patches/cross.diff --- malaga-7.12/debian/patches/cross.diff 1970-01-01 01:00:00.0 +0100 +++ malaga-7.12/debian/patches/cross.diff 2017-08-04 13:28:11.0 +0200 @@ -0,0 +1,53 @@ +From: Helmut Grohne +Subject: let PKG_CHECK_MODULES consider $ac_tool_prefix + +Index: malaga-7.12/configure.in +=== +--- malaga-7.12.orig/configure.in malaga-7.12/configure.in +@@ -58,12 +58,7 @@ + use_malshow=$enableval, use_malshow=yes) + if test $use_malshow = yes + then +- AC_MSG_CHECKING(for GTK+ 2.8 or later) +- if ! $SHELL -c 'pkg-config --atleast-version=2.8 gtk+-2.0' >/dev/null 2>/dev/null +- then +-use_malshow=no +- fi +- AC_MSG_RESULT($use_malshow) ++ PKG_CHECK_MODULES([GTK], [gtk+-2.0 >= 2.8], [use_malshow=yes], [use_malshow=no]) + if test $use_malshow = no + then + echo 1>&2 "*** No GTK+, version 2.8 or later, found. Disabling malshow." +@@ -72,29 +67,15 @@ + fi + if test $use_malshow = yes + then +- GTK_LIBS=`pkg-config gtk+-2.0 --libs` +- GTK_CFLAGS=`pkg-config gtk+-2.0 --cflags` + MALSHOW="malshow" + else +- GTK_LIBS="" +- GTK_CFLAGS="" + MALSHOW="" + fi + + dnl Check for GLib, version 2 +-AC_MSG_CHECKING(for GLib) +-if $SHELL -c 'pkg-config glib-2.0' >/dev/null 2>/dev/null +-then +- have_glib=yes +-else +- have_glib=no +-fi +-AC_MSG_RESULT($have_glib) +-if test $have_glib = yes ++PKG_CHECK_MODULES([GLIB], [glib-2.0], [have_glib=yes], [have_glib=no]) ++if test $have_glib != yes + then +- GLIB_LIBS=`pkg-config glib-2.0 --libs` +- GLIB_CFLAGS=`pkg-config glib-2.0 --cflags` +-else + echo 1>&2 "*** No GLib, version 2.0 or later, found. Please install it." + echo 1>&2 "*** You may get GLib from <http://www.gtk.org>." + exit 1 diff --minimal -Nru malaga-7.12/debian/patches/series malaga-7.12/debian/patches/series --- malaga-7.12/debian/patches/series 2016-04-07 10:00:00.0 +0200 +++ malaga-7.12/debian/patches/series 2017-08-04 13:27:29.0 +0200 @@ -4,3 +4,4 @@ no-ncurses.diff link-libm.diff texinfo.diff +cross.diff diff --minimal -Nru malaga-7.12/debian/rules malaga-7.12/debian/rules --- malaga-7.12/debian/rules2016-04-07 10:00:00.0 +0200 +++ malaga-7.12/debian/rules2017-08-04 13:28:11.0 +0200 @@ -1,5 +1,7 @@ #!/usr/bin/make -f +export INSTALL=install --strip-program=true + %: dh $@ --with autoreconf @@ -7,13 +9,11 @@ rm -rf build-doc dh_auto_clean -override_dh_auto_configure: - dh_auto_conf
Bug#872691: worklog FTCBFS: uses the build architecture compiler
Source: worklog
Version: 1.8-6
Tags: patch
User: helm...@debian.org
Usertags: rebootstrap
worklog fails to cross build from source, because it uses the build
architecture compiler. The upstream makefile bluntly hardcodes the build
architecture compiler and the packaging does not pass any cross
compilers to it either. After fixing both, worklog cross compiles
successfully. Please consider applying the attached patch.
Helmut
diff --minimal -Nru worklog-1.8/debian/changelog worklog-1.8/debian/changelog
--- worklog-1.8/debian/changelog2012-02-18 20:11:04.0 +0100
+++ worklog-1.8/debian/changelog2017-08-20 07:33:26.0 +0200
@@ -1,3 +1,12 @@
+worklog (1.8-6.1) UNRELEASED; urgency=medium
+
+ * Non-maintainer upload.
+ * Fix FTCBFS: Closes: #-1
++ Let dh_auto_build pass cross compilers to make.
++ makefile_fix.patch: Honour variable CC.
+
+ -- Helmut Grohne Sun, 20 Aug 2017 07:33:26 +0200
+
worklog (1.8-6) unstable; urgency=low
* QA upload.
diff --minimal -Nru worklog-1.8/debian/patches/makefile_fix.patch
worklog-1.8/debian/patches/makefile_fix.patch
--- worklog-1.8/debian/patches/makefile_fix.patch 2012-02-18
20:06:21.0 +0100
+++ worklog-1.8/debian/patches/makefile_fix.patch 2017-08-20
07:33:26.0 +0200
@@ -1,4 +1,4 @@
-Description: Fix hardcoded installation path from the Makefile
+Description: Fix hardcoded installation path and compiler from the Makefile
--- worklog-1.8.orig/Makefile
+++ worklog-1.8/Makefile
@@ -27,15 +27,15 @@
+ #gzip $(MAN)/man1/worklog.1
worklog: worklog.c Makefile
- gcc $(CFLAGS) $(USE_ENV_DIR) worklog.c $(LIBS) -o worklog
-
+- gcc $(CFLAGS) $(USE_ENV_DIR) worklog.c $(LIBS) -o worklog
++ $(CC) $(CFLAGS) $(USE_ENV_DIR) worklog.c $(LIBS) -o worklog
++
+clean:
+ rm -f worklog
-+
+
tar:
- rm -rf worklog-${VERSION}
- mkdir worklog-${VERSION}
@@ -37,3 +40,4 @@ tar:
cp worklog.lsm distr/worklog-${VERSION}.lsm
tar -clzvf distr/worklog-${VERSION}.tar.gz worklog-${VERSION}
- rm -rf worklog-${VERSION}
diff --minimal -Nru worklog-1.8/debian/rules worklog-1.8/debian/rules
--- worklog-1.8/debian/rules2012-02-18 19:39:00.0 +0100
+++ worklog-1.8/debian/rules2017-08-20 07:33:22.0 +0200
@@ -14,7 +14,7 @@
build-stamp: configure-stamp
dh_testdir
- $(MAKE)
+ dh_auto_build
touch build-stamp
clean:
Bug#874183: xmlrpc-c FTCBFS: runs host architecture binaries
Source: xmlrpc-c Version: 1.33.14-4 Tags: patch User: helm...@debian.org Usertags: rebootstrap xmlrpc-c fails to cross build from source, because it fails running gennmtab, which was built with the host architecture compiler. gennmtab is part of an embedded code copy of expat. In any case, it is built with BUILDTOOL_CC which strangely ends up being the host architecture compiler. Properly setting BUILDTOOL_CC (and BUILDTOOL_CCLD) fixes that. Further down the road, debian/rules unconditionally runs "make check" (even when DEB_BUILD_OPTIONS=nocheck) and naturally fails doing so. After also honouring DEB_BUILD_OPTIONS=nocheck, xmlrpc-c cross builds successfully. Please consider applying the attached patch. Also moving "make check" to override_dh_auto_test might make sense conceptually. Helmut diff --minimal -Nru xmlrpc-c-1.33.14/debian/changelog xmlrpc-c-1.33.14/debian/changelog --- xmlrpc-c-1.33.14/debian/changelog 2017-03-10 23:41:13.0 +0100 +++ xmlrpc-c-1.33.14/debian/changelog 2017-09-03 22:38:02.0 +0200 @@ -1,3 +1,12 @@ +xmlrpc-c (1.33.14-4.1) UNRELEASED; urgency=medium + + * Non-maintainer upload. + * Fix FTCBFS: (Closes: #-1) ++ Set BUILDTOOL_CC. ++ Do not run make check when DEB_BUILD_OPTIONS=nocheck. + + -- Helmut Grohne Sun, 03 Sep 2017 22:38:02 +0200 + xmlrpc-c (1.33.14-4) unstable; urgency=medium * QA upload. diff --minimal -Nru xmlrpc-c-1.33.14/debian/rules xmlrpc-c-1.33.14/debian/rules --- xmlrpc-c-1.33.14/debian/rules 2017-03-10 23:41:13.0 +0100 +++ xmlrpc-c-1.33.14/debian/rules 2017-09-03 22:38:02.0 +0200 @@ -19,9 +19,11 @@ --disable-wininet-client override_dh_auto_build: - dh_auto_build + dh_auto_build -- BUILDTOOL_CC=gcc 'BUILDTOOL_CCLD=$$(BUILDTOOL_CC)' ( cd tools && make ) +ifeq ($(filter nocheck,$(DEB_BUILD_OPTIONS)),) make check +endif override_dh_auto_clean: rm -f build-arch-stamp build-indep-stamp
Bug#875705: zoo FTCBFS: uses the build architecture compiler
Source: zoo Version: 2.10-28 Tags: patch User: helm...@debian.org Usertags: rebootstrap zoo fails to cross build from source, because it uses the build architecture compiler. debian/rules fails to pass cross compilers to make and the linux and linux64 targets fail to forward cross compilers. debian/rules also fails to select the correct target. After fixing all of the above, zoo cross builds successfully. Please consider applying the attached patch. Helmut diff --minimal -Nru zoo-2.10/debian/changelog zoo-2.10/debian/changelog --- zoo-2.10/debian/changelog 2015-12-16 16:31:10.0 +0100 +++ zoo-2.10/debian/changelog 2017-09-13 21:28:51.0 +0200 @@ -1,3 +1,13 @@ +zoo (2.10-28.1) UNRELEASED; urgency=medium + + * Non-maintainer upload. + * Fix FTCBFS: (Closes: #-1) ++ Let dh_auto_build pass cross compilers to make. ++ Fix build/host confusion. ++ Fix forwarding of CC. + + -- Helmut Grohne Wed, 13 Sep 2017 21:28:51 +0200 + zoo (2.10-28) unstable; urgency=low * QA upload. diff --minimal -Nru zoo-2.10/debian/patches/20-makefile.patch zoo-2.10/debian/patches/20-makefile.patch --- zoo-2.10/debian/patches/20-makefile.patch 2012-06-13 09:16:15.0 +0200 +++ zoo-2.10/debian/patches/20-makefile.patch 2017-09-13 21:28:51.0 +0200 @@ -1,6 +1,8 @@ From: Jari Aalto Subject: Enabled hardened build flags through dpkg-buildflags +Also enable forwarding CC for cross compilation. + --- a/makefile +++ b/makefile @@ -23,7 +23,11 @@ @@ -21,12 +23,12 @@ # Linux linux: - $(MAKE) CC="gcc" CFLAGS="-c $(OPTIM) $(LINTFLAGS) -DLINUX -DANSI_HDRS" $(TARGETS) -+ $(MAKE) CC="gcc" CFLAGS="-c $(OPTIM) $(LINTFLAGS) -DLINUX -DANSI_HDRS" LDFLAGS="$(LDFLAGS)" $(TARGETS) ++ $(MAKE) CFLAGS="-c $(OPTIM) $(LINTFLAGS) -DLINUX -DANSI_HDRS" LDFLAGS="$(LDFLAGS)" $(TARGETS) # Linux64 linux64: - $(MAKE) CC="gcc" CFLAGS="-c $(OPTIM) $(LINTFLAGS) -DLINUX -DLONG64 -DANSI_HDRS" $(TARGETS) -+ $(MAKE) CC="gcc" CFLAGS="-c $(OPTIM) $(LINTFLAGS) -DLINUX -DLONG64 -DANSI_HDRS" LDFLAGS="$(LDFLAGS)" $(TARGETS) ++ $(MAKE) CFLAGS="-c $(OPTIM) $(LINTFLAGS) -DLINUX -DLONG64 -DANSI_HDRS" LDFLAGS="$(LDFLAGS)" $(TARGETS) # ULTRIX 4.1 ultrix: diff --minimal -Nru zoo-2.10/debian/rules zoo-2.10/debian/rules --- zoo-2.10/debian/rules 2012-06-12 08:57:21.0 +0200 +++ zoo-2.10/debian/rules 2017-09-13 21:28:48.0 +0200 @@ -1,20 +1,21 @@ #!/usr/bin/make -f +include /usr/share/dpkg/architecture.mk + export DEB_BUILD_MAINT_OPTIONS = hardening=+all export DEB_CFLAGS_MAINT_APPEND = -Wall -pedantic export DEB_LDFLAGS_MAINT_APPEND = -Wl,--as-needed TARGET = linux -BITS = $(shell dpkg-architecture -qDEB_BUILD_ARCH_BITS) -ifeq ($(BITS),64) +ifeq ($(DEB_HOST_ARCH_BITS),64) TARGET = linux64 endif override_dh_auto_build: cp --archive makefile makefile.orig makedepend - $(MAKE) CFLAGS="$(CFLAGS) $(CPPFLAGS)" LDFLAGS="$(LDFLAGS)" $(TARGET) + dh_auto_build -- CFLAGS="$(CFLAGS) $(CPPFLAGS)" LDFLAGS="$(LDFLAGS)" $(TARGET) # Restore original; changed by makedepend mv makefile.orig makefile
Bug#875888: gdome2 FTCBFS: configures for the build architecture
Source: gdome2
Version: 0.8.1+debian-6
Tags: patch
User: helm...@debian.org
Usertags: rebootstrap
gdome2 fails to cross build from source, because it configures for the
build architecture. Indirecting ./configure through dh_auto_configure is
not trivial, because dh_auto_configure also passes dpkg-buildflags and a
multiarch --libdir. Still moving to dh_auto_configure fixes the cross
build issue. Please consider applying the attached patch.
Helmut
diff --minimal -Nru gdome2-0.8.1+debian/debian/changelog
gdome2-0.8.1+debian/debian/changelog
--- gdome2-0.8.1+debian/debian/changelog2013-07-02 13:02:53.0
+0200
+++ gdome2-0.8.1+debian/debian/changelog2017-09-15 17:42:41.0
+0200
@@ -1,3 +1,13 @@
+gdome2 (0.8.1+debian-6.1) UNRELEASED; urgency=medium
+
+ * Non-maintainer upload.
+ * Fix FTCBFS: (closes: #-1)
++ Let dh_auto_configure pass --host to ./configure.
++ Disable -Werror=format-security as dh_auto_configure now passes
+ dpkg-buildflags and that makes the build fail.
+
+ -- Helmut Grohne Fri, 15 Sep 2017 17:42:41 +0200
+
gdome2 (0.8.1+debian-6) unstable; urgency=low
* QA upload.
diff --minimal -Nru gdome2-0.8.1+debian/debian/rules
gdome2-0.8.1+debian/debian/rules
--- gdome2-0.8.1+debian/debian/rules2013-04-03 16:24:55.0 +0200
+++ gdome2-0.8.1+debian/debian/rules2017-09-15 17:42:41.0 +0200
@@ -1,6 +1,9 @@
#!/usr/bin/make -f
# GNU copyright 1997 to 1999 by Joey Hess.
+# does not work with -Werror=format-security
+export DEB_BUILD_MAINT_OPTIONS=hardening=-format
+
CFLAGS = -Wall -g
ifneq (,$(findstring noopt,$(DEB_BUILD_OPTIONS)))
CFLAGS += -O0
@@ -17,9 +20,8 @@
cp /usr/share/misc/config.sub /usr/share/misc/config.guess .
libtoolize -c -f
autoreconf -f -i
- ./configure --prefix=/usr \
- --mandir=\$${prefix}/share/man \
- --infodir=\$${prefix}/share/info \
+ dh_auto_configure -- \
+ --libdir=\$${prefix}/lib \
--with-html-dir=\$${prefix}/share/doc/libgdome2-dev
touch configure-stamp
Bug#877591: hfsutils FTCBFS: uses the build architecture compiler
Source: hfsutils Version: 3.2.6-13 Tags: patch User: helm...@debian.org Usertags: rebootstrap hfsutils fails to cross build from source, because it uses the build architecture compiler. After exporting a suitable CC, hfsutils cross builds successfully. Please consider applying the attached patch. Helmut diff --minimal -Nru hfsutils-3.2.6/debian/changelog hfsutils-3.2.6/debian/changelog --- hfsutils-3.2.6/debian/changelog 2014-08-10 17:52:16.0 +0200 +++ hfsutils-3.2.6/debian/changelog 2017-10-03 10:52:31.0 +0200 @@ -1,3 +1,10 @@ +hfsutils (3.2.6-13.1) UNRELEASED; urgency=medium + + * Non-maintainer upload. + * Fix FTCBFS: Export a triplet-prefixed CC. Closes: #-1. + + -- Helmut Grohne Tue, 03 Oct 2017 10:52:31 +0200 + hfsutils (3.2.6-13) unstable; urgency=medium * QA upload. diff --minimal -Nru hfsutils-3.2.6/debian/rules hfsutils-3.2.6/debian/rules --- hfsutils-3.2.6/debian/rules 2013-03-14 21:45:13.0 +0100 +++ hfsutils-3.2.6/debian/rules 2017-10-03 10:52:27.0 +0200 @@ -7,6 +7,12 @@ MAKE += -o configure.in +include /usr/share/dpkg/architecture.mk + +ifeq ($(origin CC),default) +CC := $(DEB_HOST_GNU_TYPE)-gcc +endif +export CC CFLAGS = -Wall $(shell dpkg-buildflags --get CPPFLAGS; dpkg-buildflags --get CFLAGS) LDFLAGS = $(shell dpkg-buildflags --get LDFLAGS)
Bug#878358: dnprogs FTCBFS: many reasons
Source: dnprogs Version: 2.65 Tags: patch User: helm...@debian.org Usertags: rebootstrap dnprogs fails to cross build from source for many reasons. The packaging runs plain make without passing any cross toolchain, so it ends up using the build architecture toolchain. Indirecting that through dh_auto_build fixes that. Then dapfs/Makefile hard codes the build architecture compiler g++ and that of course fails processing host architecture object files. During installation, it passes -s to install, which uses the build architecture strip and thus fails. After fixing all of the above, dnprogs cross builds successfully. Please consider applying the attached patch. Helmut diff --minimal -Nru dnprogs-2.65/Makefile.common dnprogs-2.65+nmu1/Makefile.common --- dnprogs-2.65/Makefile.common2017-07-06 21:42:10.0 +0200 +++ dnprogs-2.65+nmu1/Makefile.common 2017-10-13 06:57:30.0 +0200 @@ -157,12 +157,6 @@ DEPLIBDAP=$(TOP)/libdap/libdnet-dap.so endif -ifneq (,$(findstring nostrip,$(DEB_BUILD_OPTIONS))) -STRIPBIN= -else -STRIPBIN=-s -endif - LIBS=$(LIBDAP) $(LIBDNET) DEPLIBS=$(DEPLIBDAP) $(DEPLIBDNET) diff --minimal -Nru dnprogs-2.65/dapfs/Makefile dnprogs-2.65+nmu1/dapfs/Makefile --- dnprogs-2.65/dapfs/Makefile 2011-12-09 15:42:32.0 +0100 +++ dnprogs-2.65+nmu1/dapfs/Makefile2017-10-13 06:57:30.0 +0200 @@ -13,7 +13,7 @@ CFLAGS=-I../include -I ../librms -Wall $(DFLAGS) -fdollars-in-identifiers $(PROG1): $(PROG1OBJS) $(DEPLIBS) - g++ -o$(PROG1) $(LDFLAGS) $(PROG1OBJS) $(LIBDAP) -L../librms -lrms $(LIBDNET) -lfuse -lpthread + $(CXX) -o$(PROG1) $(LDFLAGS) $(PROG1OBJS) $(LIBDAP) -L../librms -lrms $(LIBDNET) -lfuse -lpthread install: install -d $(rootprefix)/sbin diff --minimal -Nru dnprogs-2.65/debian/changelog dnprogs-2.65+nmu1/debian/changelog --- dnprogs-2.65/debian/changelog 2017-07-06 21:42:10.0 +0200 +++ dnprogs-2.65+nmu1/debian/changelog 2017-10-13 06:57:30.0 +0200 @@ -1,3 +1,13 @@ +dnprogs (2.65+nmu1) UNRELEASED; urgency=medium + + * Non-maintainer upload. + * Fix FTCBFS: (Closes: #-1) ++ Let dh_auto_build pass cross compilers to make. ++ Do not hard code g++ in dapfs/Makefile. ++ Remove -s from install as dh_strip takes care. + + -- Helmut Grohne Fri, 13 Oct 2017 06:57:30 +0200 + dnprogs (2.65) unstable; urgency=low * QA upload. diff --minimal -Nru dnprogs-2.65/debian/rules dnprogs-2.65+nmu1/debian/rules --- dnprogs-2.65/debian/rules 2014-07-28 12:19:45.0 +0200 +++ dnprogs-2.65+nmu1/debian/rules 2017-10-13 06:57:27.0 +0200 @@ -4,7 +4,7 @@ build: $(checkdir) - make prefix=/usr RELEASE=true BUILDING_DEB=true + dh_auto_build -- prefix=/usr RELEASE=true BUILDING_DEB=true touch build build-arch: build
Bug#878365: tuxpaint-config FTCBFS: uses the build architecture toolchain
Source: tuxpaint-config Version: 0.0.13-3 Tags: patch User: helm...@debian.org Usertags: rebootstrap tuxpaint-config fails to cross build from source, because it uses the build architecture toolchain (compiler and strip). Letting dh_auto_build pass cross compilers to make fixes half of that. Then there still is the -s option to install that breaks cross compilation and -dbgsym packages. By removing it and thus deferring to dh_strip that is fixed as well. After applying the attached patch, tuxpaint-config cross builds successfully. Please consider using it. Helmut diff --minimal -Nru tuxpaint-config-0.0.13/debian/changelog tuxpaint-config-0.0.13/debian/changelog --- tuxpaint-config-0.0.13/debian/changelog 2017-08-30 14:13:37.0 +0200 +++ tuxpaint-config-0.0.13/debian/changelog 2017-10-13 10:23:10.0 +0200 @@ -1,3 +1,12 @@ +tuxpaint-config (0.0.13-3.1) UNRELEASED; urgency=medium + + * Non-maintainer upload. + * Fix FTCBFS: (Closes: #-1) ++ Let dh_auto_build pass cross compilers to make. ++ Let dh_strip perform the stripping. (Also fixes -dbgsym.) + + -- Helmut Grohne Fri, 13 Oct 2017 10:23:10 +0200 + tuxpaint-config (0.0.13-3) unstable; urgency=medium * QA upload. diff --minimal -Nru tuxpaint-config-0.0.13/debian/rules tuxpaint-config-0.0.13/debian/rules --- tuxpaint-config-0.0.13/debian/rules 2017-08-30 14:13:37.0 +0200 +++ tuxpaint-config-0.0.13/debian/rules 2017-10-13 10:23:08.0 +0200 @@ -11,10 +11,6 @@ DPKG_EXPORT_BUILDFLAGS = 1 include /usr/share/dpkg/buildflags.mk -ifeq (,$(findstring nostrip,$(DEB_BUILD_OPTIONS))) - INSTALL_PROGRAM += -s -endif - configure: configure-stamp configure-stamp: dh_testdir @@ -29,7 +25,7 @@ dh_testdir # Add here commands to compile the package. - $(MAKE) PREFIX=/usr CFLAGS="$(CFLAGS)" CXXFLAGS="$(CXXFLAGS)" CPPFLAGS="$(CPPFLAGS)" LDFLAGS="$(LDFLAGS)" + dh_auto_build -- PREFIX=/usr CFLAGS="$(CFLAGS)" CXXFLAGS="$(CXXFLAGS)" CPPFLAGS="$(CPPFLAGS)" LDFLAGS="$(LDFLAGS)" touch build-stamp
Bug#943780: rc FTCBFS: fails running ./mksignal
Source: rc Version: 1.7.4+97.gceb59bb-2 User: debian-cr...@lists.debian.org Usertags: ftcbfs rc fails to cross build from source, because it fails running ./mksignals. This tool seems to be intended towards creating a lookup table from signal number to signal name. It's not going to work this way for cross compilation at all. I'm not sure exactly whether we need rc to be cross buildable given that most embedded users seem to prefer systemd, but let me explain what it would take to fix this: The signal numbers in mksignal.c should be looked up using AC_COMPUTE_INT in configure.ac. Then given the discovered numbers, one can create the source code without relying on host architecture execution. One way would be to write a shell script creating sigmsgs.c and sigmsgs.h. Alternatively, mksignal.c could be retained and instead of relying on the signal macros, it would use the results from AC_COMPUTE_INT and be compiled with CC_FOR_BUILD as discovered by AX_CC_FOR_BUILD (from the autoconf-archive). Do you think this is worth the effort? If not, please tag this bug wontfix. Helmut
Bug#943852: tstools FTCBFS: uses the build architecture ld
Source: tstools Version: 1.13~git20151030-1 Tags: patch User: debian-cr...@lists.debian.org Usertags: ftcbfs tstools fails to cross build from source, because the upstream makefile links with $(LD), which happens get default initialized to the build architecture linker. Exporting it from dpkg's buildtools.mk makes tstools cross buildable. Please consider applying the attached patch. Helmut diff --minimal -Nru tstools-1.13~git20151030/debian/changelog tstools-1.13~git20151030/debian/changelog --- tstools-1.13~git20151030/debian/changelog 2019-10-01 17:34:56.0 +0200 +++ tstools-1.13~git20151030/debian/changelog 2019-10-30 20:06:41.0 +0100 @@ -1,3 +1,9 @@ +tstools (1.13~git20151030-2) UNRELEASED; urgency=medium + + * Fix FTCBFS: Export a suitable LD. (Closes: #-1) + + -- Helmut Grohne Wed, 30 Oct 2019 20:06:41 +0100 + tstools (1.13~git20151030-1) unstable; urgency=medium * QA upload. diff --minimal -Nru tstools-1.13~git20151030/debian/rules tstools-1.13~git20151030/debian/rules --- tstools-1.13~git20151030/debian/rules 2019-10-01 17:23:17.0 +0200 +++ tstools-1.13~git20151030/debian/rules 2019-10-30 20:06:40.0 +0100 @@ -10,6 +10,8 @@ export DEB_LDFLAGS_MAINT_APPEND = -Wl,--as-needed include /usr/share/dpkg/default.mk +-include /usr/share/dpkg/buildtools.mk +export LD DEB_CFLAGS = $(shell dpkg-buildflags --get CFLAGS) DEB_LDFLAGS = $(shell dpkg-buildflags --get LDFLAGS)
Bug#944872: libindicator FTBFS: error: G_ADD_PRIVATE [-Werror]
Source: libindicator Version: 0.5.0-4 Severity: serious Tags: ftbfs upstream libindicator fails to build from source in unstable. https://tests.reproducible-builds.org/debian/rbuild/unstable/amd64/libindicator_0.5.0-4.rbuild.log.gz | libtool: compile: gcc -DHAVE_CONFIG_H -I. -I/build/1st/libindicator-0.5.0/./libindicator -I.. -Wdate-time -D_FORTIFY_SOURCE=2 -pthread -I/usr/include/gtk-2.0 -I/usr/lib/x86_64-linux-gnu/gtk-2.0/include -I/usr/include/pango-1.0 -I/usr/include/atk-1.0 -I/usr/include/cairo -I/usr/include/pixman-1 -I/usr/include/gdk-pixbuf-2.0 -I/usr/include/pango-1.0 -I/usr/include/harfbuzz -I/usr/include/pango-1.0 -I/usr/include/fribidi -I/usr/include/uuid -I/usr/include/freetype2 -I/usr/include/libpng16 -I/usr/include/gio-unix-2.0 -I/usr/include/libmount -I/usr/include/blkid -I/usr/include/glib-2.0 -I/usr/lib/x86_64-linux-gnu/glib-2.0/include -DG_LOG_DOMAIN=\"libindicator\" -Wall -Werror -Wno-deprecated-declarations -g -O2 -ffile-prefix-map=/build/1st/libindicator-0.5.0=. -fstack-protector-strong -Wformat -Werror=format-security -c /build/1st/libindicator-0.5.0/./libindicator/indicator-object.c -fPIC -DPIC -o .libs/libindicator_la-indicator-object.o | /build/1st/libindicator-0.5.0/./libindicator/indicator-object.c: In function 'indicator_object_init': | /build/1st/libindicator-0.5.0/./libindicator/indicator-object.c:307:13: error: G_ADD_PRIVATE [-Werror] | 307 | IndicatorObjectPrivate * priv = G_TYPE_INSTANCE_GET_PRIVATE (self, INDICATOR_OBJECT_TYPE, IndicatorObjectPrivate); | | ^~~ | cc1: all warnings being treated as errors | make[4]: *** [Makefile:643: libindicator_la-indicator-object.lo] Error 1 | make[4]: Leaving directory '/build/1st/libindicator-0.5.0/build/gtk2/libindicator' | make[3]: *** [Makefile:501: all] Error 2 | make[3]: Leaving directory '/build/1st/libindicator-0.5.0/build/gtk2/libindicator' | make[2]: *** [Makefile:434: all-recursive] Error 1 | make[2]: Leaving directory '/build/1st/libindicator-0.5.0/build/gtk2' | make[1]: *** [Makefile:365: all] Error 2 | make[1]: Leaving directory '/build/1st/libindicator-0.5.0/build/gtk2' | make: *** [/usr/share/cdbs/1/class/makefile.mk:77: debian/stamp-makefile-build/gtk2] Error 2 | dpkg-buildpackage: error: debian/rules build subprocess returned exit status 2 Also reproduced by crossqa: http://crossqa.debian.net/build/libindicator_0.5.0-4_s390x_20191027214029.log This is likely due to changes in glib2.0. If you search for error and G_ADD_PRIVATE, you'll find similar bugs. Helmut
Bug#944871: docbook-xsl: readds catalogs to the super catalog on every upgrade
Source: docbook-xsl Version: 1.79.1+dfsg-2 Severity: serious Justification: Policy 10.7.3 Consider the following interaction: | # apt install docbook-xsl | # grep delegateURI.*xsl /etc/xml/catalog | http://docbook.sourceforge.net/release/xsl/"; catalog="file:///etc/xml/docbook-xsl.xml"/> | # update-xmlcatalog --del --type uri --id "http://docbook.sourceforge.net/release/xsl/"; --root | # grep delegateURI.*xsl /etc/xml/catalog | # apt reinstall docbook-xsl | # grep delegateURI.*xsl /etc/xml/catalog | http://docbook.sourceforge.net/release/xsl/"; catalog="file:///etc/xml/docbook-xsl.xml"/> | # We can see that apt reinstall docbook-xsl changed the contents of /etc/xml/catalog. This also happens during package upgrades. The prerm script removes docbook-xsl from the root catalog and the postinst script adds it back. Thus, removal by a user is not preserved. This is a violation of Debian policy section 10.7.3, which says "local changes must be preserved during a package upgrade". We had a very similar issue with the sgml super catalog. Read up on #88010 and #477751. Fixing that one was very painful. It required a lot of uploads and we ran into bugs with dpkg triggers. I guess that this bug is similarly painful. On the bright side, fixing this has the potential to remove a lot of maintainer scripts and move us towards more declarative packaging. Unfortunately, it really is a violation of the policy and we should either fix this bug or policy. I'm unsure how to proceed here, so I just report it. Good luck. Helmut
Bug#944873: mark docbook-xsl-ns Multi-Arch: foreign
Package: docbook-xsl-ns
Version: 1.79.1+dfsg-2
Tags: patch
User: debian-cr...@lists.debian.org
Usertags: cross-satisfiability
Control: affects -1 + src:colord src:gdcm src:iputils src:kildclient src:nsis
src:package-update-indicator src:xmlstarlet src:xwrited
The affected packages fail to satisfy their cross build dependencies,
because their dependency on docbook-xsl-ns is unsatisfiable. In general,
Architecture: all packages can never satisfy cross build dependencies
unless marked Multi-Arch: foreign or annotated :native. In this case,
the foreign marking is reasonable:
* docbook-xsl is already thus marked and docbook-xsl-ns is very
similar.
* docbook-xsl-ns is not proposed by the multiarch hinter, because it
has maintainer scripts and the hinter cannot understand them.
Inspecting them reveals that they only update the super catalog,
which is an architecture-independent operation.
Please consider applying the attached patch.
Helmut
diff --minimal -Nru docbook-xsl-1.79.1+dfsg/debian/changelog
docbook-xsl-1.79.1+dfsg/debian/changelog
--- docbook-xsl-1.79.1+dfsg/debian/changelog2016-09-07 01:41:50.0
+0200
+++ docbook-xsl-1.79.1+dfsg/debian/changelog2019-11-16 18:16:14.0
+0100
@@ -1,3 +1,9 @@
+docbook-xsl (1.79.1+dfsg-3) UNRELEASED; urgency=medium
+
+ * Also mark docbook-xsl-ns Multi-Arch: foreign. (Closes: #-1)
+
+ -- Helmut Grohne Sat, 16 Nov 2019 18:16:14 +0100
+
docbook-xsl (1.79.1+dfsg-2) unstable; urgency=medium
* QA upload.
diff --minimal -Nru docbook-xsl-1.79.1+dfsg/debian/control
docbook-xsl-1.79.1+dfsg/debian/control
--- docbook-xsl-1.79.1+dfsg/debian/control 2016-09-07 01:41:50.0
+0200
+++ docbook-xsl-1.79.1+dfsg/debian/control 2019-11-16 18:13:12.0
+0100
@@ -40,6 +40,7 @@
Package: docbook-xsl-ns
Architecture: all
+Multi-Arch: foreign
Depends: ${misc:Depends}
Recommends: docbook5-xml (>> 5)
Suggests: dbtoepub,
Bug#945545: xpp FTBFS: error: invalid use of incomplete type 'ipp_t'
Source: xpp
Version: 1.5-cvs20081009-4
Severity: serious
Tags: ftbfs
xpp fails to build from source in unstable.
https://tests.reproducible-builds.org/debian/rbuild/unstable/amd64/xpp_1.5-cvs20081009-4.rbuild.log.gz
| cupshelper.cxx: In static member function 'static ipp_t*
CupsHelper::newIppRequest()':
| cupshelper.cxx:97:10: error: invalid use of incomplete type 'ipp_t' {aka
'struct _ipp_s'}
|97 | request->request.op.request_id = 1;
| | ^~
| In file included from /usr/include/cups/cups.h:27,
| from cupshelper.h:36,
| from cupshelper.cxx:35:
| /usr/include/cups/ipp.h:731:16: note: forward declaration of 'ipp_t' {aka
'struct _ipp_s'}
| 731 | typedef struct _ipp_s ipp_t; / IPP request/response data /
| |^~
Also seen by crossqa since about two months:
http://crossqa.subdivi.de/build/xpp_1.5-cvs20081009-4_armel_20190914094237.log
Helmut
Bug#946276: mark docbook-website Multi-Arch: foreign
Package: docbook-website
Version: 2.5.0.0-8
Tags: patch
User: debian-cr...@lists.debian.org
Usertags: cross-satisfiability
Control: affects -1 + src:libzeep
libzeep fails to satisfy its cross Build-Depends, because its dependency
on docbook-website is not satisfiable. In general, Architecture: all
packages can never satisfy cross Build-Depends unless marked Multi-Arch:
foreign or annotated :native. In this case, the multiarch hinter does
not suggest the Multi-Arch: foreign marking, because docbook-website has
maintainer scripts. I checked the maintainer scripts to be architecture
agnostic. Many other docbook stylesheet packages are already thus
marked. Please consider applying the attached patch.
Helmut
diff -u docbook-website-2.5.0.0/debian/changelog
docbook-website-2.5.0.0/debian/changelog
--- docbook-website-2.5.0.0/debian/changelog
+++ docbook-website-2.5.0.0/debian/changelog
@@ -1,3 +1,9 @@
+docbook-website (2.5.0.0-9) UNRELEASED; urgency=medium
+
+ * Mark docbook-website Multi-Arch: foreign. (Closes: #-1)
+
+ -- Helmut Grohne Fri, 06 Dec 2019 06:11:52 +0100
+
docbook-website (2.5.0.0-8) unstable; urgency=low
* QA upload.
diff -u docbook-website-2.5.0.0/debian/control
docbook-website-2.5.0.0/debian/control
--- docbook-website-2.5.0.0/debian/control
+++ docbook-website-2.5.0.0/debian/control
@@ -9,6 +9,7 @@
Package: docbook-website
Architecture: all
+Multi-Arch: foreign
Replaces: docbook-xml-website
Depends: ${misc:Depends}, docbook-xml (>= 4.2-7), docbook-xsl
Conflicts: docbook-xml-website
Bug#840782: closed by Hideki Yamane (Bug#840782: fixed in fontforge 1:20190801~dfsg-1)
Control: reopen -1
Control: affects -1 - src:fontforge
On Sat, Nov 09, 2019 at 07:03:03PM +, Debian Bug Tracking System wrote:
> Changes:
> fontforge (1:20190801~dfsg-1) experimental; urgency=medium
> .
>* New upstream version 20190801~dfsg (Closes: #866690, #912062)
>* debian/control{,.in}
> - remove Christian Perrier from Uploaders
>(Closes: #894873, #927590)
> - add Rules-Requires-Root: no
> - bump up libfontforge3 from 2
> - migrate to python3 (Closes: #936538)
> - set Multi-Arch: foreign for -common,doc
> - drop gnulib (Closes: #840782)
While your change to fontforge is gnulib-related. The bug you closed is
with gnulib, not fontforge. It no longer affects fontforge, but it's
still there.
Helmut
Bug#947138: udo FTCBFS: builds for the build architecture
Source: udo Version: 6.4.1-5 Tags: patch User: debian-cr...@lists.debian.org Usertags: ftcbfs udo fails to cross build from source, because it does not pass cross tools to make. The easiest way of fixing that - using dh_auto_build - fails when the built udo is run to generate the documentation. Given that udo is small and has few dependencies, we can simply build it twice though. Please consider applying the attached patch. Helmut diff --minimal -Nru udo-6.4.1/debian/changelog udo-6.4.1/debian/changelog --- udo-6.4.1/debian/changelog +++ udo-6.4.1/debian/changelog @@ -1,3 +1,9 @@ +udo (6.4.1-6) UNRELEASED; urgency=medium + + * Fix FTCBFS: Build a second time for cross building. (Closes: #-1) + + -- Helmut Grohne Tue, 21 Dec 2019 17:14:56 +0100 + udo (6.4.1-5) unstable; urgency=medium * QA upload. diff --minimal -Nru udo-6.4.1/debian/rules udo-6.4.1/debian/rules --- udo-6.4.1/debian/rules +++ udo-6.4.1/debian/rules @@ -1,6 +1,8 @@ #!/usr/bin/make -f #export DH_VERBOSE = 1 +include /usr/share/dpkg/architecture.mk + export DEB_BUILD_MAINT_OPTIONS=hardening=+all,-format %: @@ -23,6 +25,11 @@ Source/udo --no-logfile --html -q -o eng/html/index.html $(CURDIR)/Guide/eng/manual/index.u cp Guide/ger/manual/images/*.gif ger/html/images/ Source/udo --no-logfile --html -q -o ger/html/index.html $(CURDIR)/Guide/ger/manual/index.u +ifneq ($(DEB_BUILD_ARCH),$(DEB_HOST_ARCH)) + dh_auto_clean --sourcedirectory=Source --buildsystem=makefile -- -f Makefile.debian + dh_auto_build --sourcedirectory=Source --buildsystem=makefile -- -f Makefile.debian \ + CFLAGS="$(CFLAGS) $(CPPFLAGS) -D__LINUX__" LDFLAGS="$(LDFLAGS)" +endif override_dh_auto_install: $(MAKE) -C Source -f Makefile.debian install DESTDIR=$(CURDIR)/debian/tmp
Bug#947576: adplay FTCBFS: misdetects ao support
Source: adplay Version: 1.8-1 Tags: patch User: debian-cr...@lists.debian.org Usertags: ftcbfs adplay fails to cross build from source, because its configure script misdetects availability of libao-dev. It's not included in Build-Depends, yet assumed ok during cross building. An easy solution is to just --disable-output-ao explicitly. Please consider applying the attached patch. Helmut diff --minimal -Nru adplay-1.8/debian/changelog adplay-1.8/debian/changelog --- adplay-1.8/debian/changelog 2019-10-17 16:23:04.0 +0200 +++ adplay-1.8/debian/changelog 2019-12-28 09:13:30.0 +0100 @@ -1,3 +1,9 @@ +adplay (1.8-2) UNRELEASED; urgency=medium + + * Fix FTCBFS: Explicitly disable ao output. (Closes: #-1) + + -- Helmut Grohne Sat, 28 Dec 2019 09:13:30 +0100 + adplay (1.8-1) unstable; urgency=medium * QA upload. diff --minimal -Nru adplay-1.8/debian/rules adplay-1.8/debian/rules --- adplay-1.8/debian/rules 2019-10-17 16:16:08.0 +0200 +++ adplay-1.8/debian/rules 2019-12-28 09:13:29.0 +0100 @@ -6,6 +6,9 @@ %: dh $@ +override_dh_auto_configure: + dh_auto_configure -- --disable-output-ao + # src/getopt.h conflicts with the one in glibc >= 2.26 override_dh_clean: dh_clean src/getopt.h
Bug#947972: transcalc: broken, outdated, embedded copy of AM_PATH_GTK_2_0
Source: transcalc Version: 0.14-6 Tags: fixed-upstream User: debian-cr...@lists.debian.org Usertags: ftcbfs transcalc fails to cross build from source, because it uses a broken, outdated, embedded copy of AM_PATH_GTK_2_0 in aclocal.m4. The actual bug is already fixed, see #895002. transcalc just happens to ship a copy of the bug. The Debian policy discourages embedded code copies. Please remove your copy. Failing that, please update your copy and register it with the security tracker. See https://wiki.debian.org/EmbeddedCodeCopies for how to do that. This bug report comes without a patch, because the actual issue is already fixed. Helmut
Bug#947973: fspanel FTCBFS: does not pass cross tools to make
Source: fspanel
Version: 0.7-15
Tags: patch
User: debian-cr...@lists.debian.org
Usertags: ftcbfs
fspanel fails to cross build from source, because it does not pass cross
tools to make. The easiest way of fixing that - using dh_auto_build -
makes fspanel cross buildable. Please consider applying the attached
patch.
Helmut
diff --minimal -Nru fspanel-0.7/debian/changelog fspanel-0.7/debian/changelog
--- fspanel-0.7/debian/changelog2019-12-31 16:10:52.0 +0100
+++ fspanel-0.7/debian/changelog2020-01-02 22:24:12.0 +0100
@@ -1,3 +1,9 @@
+fspanel (0.7-16) UNRELEASED; urgency=medium
+
+ * Fix FTCBFS: Let dh_auto_build pass cross tools to make. (Closes: #-1)
+
+ -- Helmut Grohne Thu, 02 Jan 2020 22:24:12 +0100
+
fspanel (0.7-15) unstable; urgency=medium
[ Giovani Augusto Ferreira ]
diff --minimal -Nru fspanel-0.7/debian/rules fspanel-0.7/debian/rules
--- fspanel-0.7/debian/rules2019-12-31 16:10:52.0 +0100
+++ fspanel-0.7/debian/rules2020-01-02 22:24:11.0 +0100
@@ -13,7 +13,7 @@
# Do nothing. See files debian/{install,manpages}
override_dh_auto_build:
- $(MAKE) EXTRA_CFLAGS="$(CFLAGS) $(CPPFLAGS) $(LDFLAGS)"
+ dh_auto_build -- EXTRA_CFLAGS="$(CFLAGS) $(CPPFLAGS) $(LDFLAGS)"
%:
dh $@
