Bug#687630: unblock: arpack/3.1.1-2.1
Package: release.debian.org
Severity: normal
User: release.debian.org@packages.debian.org
Usertags: unblock
Please unblock package arpack
The only change is the addition of
Breaks: octave3.2
to libarpack2. This fixes some upgrade paths as noticed and verified in
piuparts.
octave3.2 is an obsolete package that is removed during the upgrade, but
apt may choose a bad timing causing a trigger failure (actually a dpkg
bug, but probably can't be fixed for wheezy).
Andreas
unblock arpack/3.1.1-2.1
diffstat for arpack-3.1.1 arpack-3.1.1
changelog | 13 +++++++++++++
control | 1 +
2 files changed, 14 insertions(+)
diff -Nru arpack-3.1.1/debian/changelog arpack-3.1.1/debian/changelog
--- arpack-3.1.1/debian/changelog 2012-06-14 10:45:35.000000000 +0200
+++ arpack-3.1.1/debian/changelog 2012-09-13 21:01:02.000000000 +0200
@@ -1,3 +1,16 @@
+arpack (3.1.1-2.1) unstable; urgency=low
+
+ * Non-maintainer upload.
+ * libarpack2: add Breaks: octave3.2 (Closes: #684773)
+ This fixes some upgrade paths by removing the obsolete octave3.2 (and its
+ triggers) a bit earlier. dpkg may run trigger processing of a package
+ even if its dependencies are not satisfied. The octave3.2 triggers may
+ be run in such a state and will fail, aborting the upgrade, even if
+ octave3.2 would be removed later anyway.
+ This is a dpkg bug, but probably won't be fixed for wheezy (see #678848).
+
+ -- Andreas Beckmann <debian@abeckmann.de> Thu, 13 Sep 2012 21:01:02 +0200
+
arpack (3.1.1-2) unstable; urgency=low
* Update the symbol files. Thanks to Nobuhiro Iwamatsu (Closes: #666038)
diff -Nru arpack-3.1.1/debian/control arpack-3.1.1/debian/control
--- arpack-3.1.1/debian/control 2012-04-05 09:53:57.000000000 +0200
+++ arpack-3.1.1/debian/control 2012-09-13 21:01:54.000000000 +0200
@@ -18,6 +18,7 @@
Depends: ${shlibs:Depends}, ${misc:Depends}, libparpack2
Replaces: arpack2
Conflicts: arpack2
+Breaks: octave3.2
Description: Fortran77 subroutines to solve large scale eigenvalue problems
ARPACK software is capable of solving large scale symmetric,
nonsymmetric, and generalized eigenproblems from significant
Reply to: