Bug#979493: Error: Rule failed to generate the following targets: _doc/_html/highlight.pack.js

2021-01-07 Thread Johannes 'josch' Schauer
Package: ocaml-odoc
Version: 1.5.1+dfsg-2+b1
Severity: grave
Justification: renders package unusable

Hi Ralf,

as you probably remember, I brought up the odoc problem before here:

https://gitlab.com/irill/dose3/-/merge_requests/1#note_475481814

I now investigated further and it seems that the package has an
autopkgtest (yay!) so I triggered that and it failed with the same error
message that I got:

https://ci.debian.net/data/autopkgtest/unstable/amd64/o/ocaml-odoc/9469763/log.gz

Since the logs are removed after some time, here the last lines from the
log:

autopkgtest [02:11:21]: test odoc-on-odoc: [---
File series fully applied, ends at patch debian/patches/no-vendored-js-highlight
File "_doc/_html/_unknown_", line 1, characters 0-0:
Error: Rule failed to generate the following targets:
- _doc/_html/highlight.pack.js
autopkgtest [02:11:26]: test odoc-on-odoc: ---]
autopkgtest [02:11:26]: test odoc-on-odoc:  - - - - - - - - - - results - - - - 
- - - - - -
odoc-on-odoc FAIL non-zero exit status 1
autopkgtest [02:11:27]:  summary
odoc-on-odoc FAIL non-zero exit status 1

[ I also took the liberty to add salsaci to the ocaml-odoc repository on
salsa and it seems there is also a FTBFS which I was unable to trigger
locally using sbuild:
https://salsa.debian.org/ocaml-team/ocaml-odoc/-/pipelines/216353 ]

This makes me believe that this is probably not a problem with my system
but with the ocaml-odoc package itself.

A minimal test that triggers this error message is this shell script:

--%<--
#!/bin/sh
mkdir "$1/test"
cat << END > "$1/test/dune-project"
(lang dune 2.0)
(name test)
(generate_opam_files true)
(package (name test))
END
chroot "$1" sh -c 'cd /test && dune build @doc'
-->%--

Which I passed to mmdebstrap with the first timestamp from
snapshot.debian.org in which the ocaml-odoc package existed to see
whether maybe this is a regression and this used to work in the
beginning:

$ mmdebstrap --aptopt='Acquire::Check-Valid-Until "false"' \
--variant=apt unstable --include=ocaml-odoc,ocaml-dune,ocaml-nox \
--customize-hook=/tmp/test.sh \
/dev/null http://snapshot.debian.org/archive/debian/20200624T145103Z/

But even back in June 2020 I got the same error message as from the
autopkgtest on debci run above.

I didn't try with the opam version yet. Maybe it works there. Nobody
else (build-)depends on ocaml-odoc yet, so this problem is probably not
super important unless we want to generate documentation for the next
dose3 release.

Thanks!

cheers, josch



Bug#958621: marked as pending in schroot

2020-09-06 Thread Johannes 'josch' Schauer
Control: tag -1 pending

Hello,

Bug #958621 in schroot reported by you has been fixed in the
Git repository and is awaiting an upload. You can see the commit
message below and you can check the diff of the fix at:

https://salsa.debian.org/debian/schroot/-/commit/e06a55cf5a24aa78daa3fb3423a49ab09de866ae


Remove Build-Depends on deprecated dh-systemd (Closes: #958621)


(this message was generated automatically)
-- 
Greetings

https://bugs.debian.org/958621



Bug#969331: rviz: [rospack] Error: no such package media_export

2020-08-31 Thread Johannes 'josch' Schauer
Package: rviz
Version: 1.14.1+dfsg-1
Severity: grave
Justification: renders package unusable

Hi,

starting rviz on Debian unstable currently yields the following output:

[ INFO] [1598885512.042478573]: rviz version 1.14.1
[ INFO] [1598885512.042531596]: compiled against Qt version 5.14.2
[ INFO] [1598885512.042547093]: compiled against OGRE version 1.9.0 (Ghadamon)
[ INFO] [1598885512.052974842]: Forcing OpenGl version 0.
[ INFO] [1598885512.592152405]: Stereo is NOT SUPPORTED
[ INFO] [1598885512.592246596]: OpenGl version: 4.6 (GLSL 4.6).
[rospack] Error: no such package media_export
[librospack]: error while executing command
zsh: segmentation fault  rviz

Maybe triggered by this line?

https://sources.debian.org/src/ros-rviz/1.14.1+dfsg-1/src/rviz/ogre_helpers/render_system.cpp/?hl=316#L318

According to the package.xml, rviz depends on media_export but that one
is not packaged yet?

Thanks!

cheers, josch


-- System Information:
Debian Release: bullseye/sid
  APT prefers stable
  APT policy: (990, 'stable'), (500, 'unstable-debug'), (500, 
'stable-updates'), (500, 'unstable'), (500, 'testing'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386, armhf

Kernel: Linux 4.19.0-9-amd64 (SMP w/4 CPU cores)
Kernel taint flags: TAINT_CRAP
Locale: LANG=en_US.utf8, LC_CTYPE=en_US.utf8 (charmap=UTF-8), LANGUAGE=en_US:en 
(charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

Versions of packages rviz depends on:
ii  fonts-liberation   1:1.07.4-9
ii  libboost-filesystem1.71.0  1.71.0-6+b2
ii  libboost-thread1.71.0  1.71.0-6+b2
ii  libc6  2.31-2
ii  libclass-loader1d  0.5.0-1
ii  libconsole-bridge0.4   0.4.3+dfsg-1
ii  libgcc-s1  10.2.0-3
ii  libimage-transport0d   1.12.0-1+b1
ii  libinteractive-markers2d   1.12.0-3
ii  liblaser-geometry0d1.6.5-1+b1
ii  libmessage-filters1d   1.15.8+ds1-1
ii  libogre-1.9.0v51.9.0+dfsg1-12+b1
ii  libopengl0 1.3.1-1
ii  libqt5core5a   5.14.2+dfsg-4
ii  libqt5gui5 5.14.2+dfsg-4
ii  libqt5svg5 5.14.2-2
ii  libqt5widgets5 5.14.2+dfsg-4
ii  librosconsole3d1.14.2-1
ii  libroscpp-serialization0d  0.7.2-2
ii  libroscpp3d1.15.8+ds1-1
ii  libroslib0d1.15.4-2
ii  librostime1d   0.7.2-4
ii  librviz5d  1.14.1+dfsg-1
ii  libstdc++6 10.2.0-3
ii  libtf2-2d  0.7.2-3
ii  libtinyxml2-8  8.0.0+dfsg-2
ii  liburdf1d  1.13.2-2
ii  liburdfdom-world   1.0.3-1

rviz recommends no packages.

rviz suggests no packages.

-- no debconf information



Bug#968217: python is priority:standard but not installable

2020-08-10 Thread Johannes 'josch' Schauer
Package: python
Version: 2.7.17-2
Severity: grave
Justification: renders package unusable

Hi,

the python package is currently uninstallable in Debian unstable:

output-version: 1.2
report:
 -
  package: python
  version: 2.7.17-2
  architecture: amd64
  status: broken
  reasons:
   -
missing:
 pkg:
  package: libpython-stdlib
  version: 2.7.17-2
  architecture: amd64
  unsat-dependency: libpython2-stdlib (= 2.7.17-2)
 depchains:
  -
   depchain:
-
 package: python
 version: 2.7.17-2
 architecture: amd64
 depends: libpython-stdlib (= 2.7.17-2)

Maybe this is intentional because of the imminent removal of python 2.
What is problematic here is, that python is also priority:standard. My
package mmdebstrap is a debootstrap replacement and in addition to being
able to install packages with priorities required and important, it can
also install all packages with priority standard. If a package from the
priority:standard set is not installable, then this also breaks the unit
tests and autopkgtest of mmdebstrap and thus, python not being
installable is currently blocking my mmdebstrap development. I see
several ways to resolve this:

 - fix the installability problem (there seems to be a version mismatch)
 - remove the python package
 - adjust the priority to something less important than "standard"
 - add a rule to mmdebstrap to ignore a package named python

Since the uninstallability has now been a thing for over a week, I'm
filing this bug to inquire about the best way forward.

Thanks!

cheers, josch



Bug#965111: closing 965111

2020-07-27 Thread Johannes 'josch' Schauer
close 965111 0.17.0
thanks

the autopkgtests pass just fine on testing and unstable



Bug#965050: marked as pending in flask-babel

2020-07-15 Thread Johannes 'josch' Schauer
Control: tag -1 pending

Hello,

Bug #965050 in flask-babel reported by you has been fixed in the
Git repository and is awaiting an upload. You can see the commit
message below and you can check the diff of the fix at:

https://salsa.debian.org/python-team/modules/flask-babel/-/commit/a4e9ce6a08b7322c92c6170b9782f748577828ee


fix import of ImmutableDict from werkzeug (closes: #965050)


(this message was generated automatically)
-- 
Greetings

https://bugs.debian.org/965050



Bug#964026: marked as pending in dose3

2020-07-12 Thread Johannes 'josch' Schauer
Control: tag -1 pending

Hello,

Bug #964026 in dose3 reported by you has been fixed in the
Git repository and is awaiting an upload. You can see the commit
message below and you can check the diff of the fix at:

https://salsa.debian.org/ocaml-team/dose3/-/commit/760996a450bf2000bc66cbb8eddc08e8bf99e081


add patch to handle m-a:same conflicts with virtual packages (closes: #964026)


(this message was generated automatically)
-- 
Greetings

https://bugs.debian.org/964026



Bug#958362: pdfrw: fails with python 3.7+ -- abandoned upstream

2020-04-20 Thread Johannes 'josch' Schauer
Source: pdfrw
Version: 0.4-2
Severity: grave
Justification: renders package unusable

Hi,

the package fails with Python 3.7 and newer. Starting from Python 3.7,
StopIteration is a RuntimeError (see PEP 479). But pdfrw still uses
StopIteration:

https://github.com/pmaupin/pdfrw/issues/145
https://github.com/pmaupin/pdfrw/issues/199

This package recently got its python2 package removed but since tests
are disabled, nobody noticed that its tests fail with Python 3.7+:

https://github.com/pmaupin/pdfrw/issues/198
https://github.com/pmaupin/pdfrw/issues/197

Other upstream bugs indicating that the package is broken on Python3:

https://github.com/pmaupin/pdfrw/issues/193
https://github.com/pmaupin/pdfrw/issues/170

Upstream didn't close bugs for two years and the last commit was more
than two years ago. Upstream admits not having time at the moment here:

https://github.com/pmaupin/pdfrw/issues/195

And is looking for additional maintainers here:

https://github.com/pmaupin/pdfrw/issues/191

In this state, pdfrw should not be included in the next Debian stable
release.

Thanks!

cheers, josch



-- System Information:
Debian Release: bullseye/sid
  APT prefers stable
  APT policy: (990, 'stable'), (500, 'unstable-debug'), (500, 
'stable-updates'), (500, 'unstable'), (500, 'testing'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386, armhf

Kernel: Linux 4.19.0-8-amd64 (SMP w/4 CPU cores)
Kernel taint flags: TAINT_WARN
Locale: LANG=en_US.utf8, LC_CTYPE=en_US.utf8 (charmap=UTF-8), LANGUAGE=en_US:en 
(charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled



Bug#953404: debootstrapping testing and stable results in exit 100

2020-03-08 Thread Johannes 'josch' Schauer
Package: debootstrap
Version: 1.0.120
Severity: grave
Justification: renders package unusable

Hi,

steps to reproduce:

$ sudo debootstrap --variant=minbase stable debian-stable
[...]
$ echo $?
100

It works for unstable though.

Thanks!

cheers, josch


-- System Information:
Debian Release: bullseye/sid
  APT prefers stable
  APT policy: (990, 'stable'), (500, 'unstable-debug'), (500, 
'stable-updates'), (500, 'unstable'), (500, 'testing'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386, armhf

Kernel: Linux 4.19.0-8-amd64 (SMP w/4 CPU cores)
Kernel taint flags: TAINT_WARN
Locale: LANG=en_US.utf8, LC_CTYPE=en_US.utf8 (charmap=UTF-8), LANGUAGE=en_US:en 
(charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

Versions of packages debootstrap depends on:
ii  wget  1.20.1-1.1

Versions of packages debootstrap recommends:
ii  arch-test   0.15-2
ii  debian-archive-keyring  2019.1
ii  gnupg   2.2.19-2

Versions of packages debootstrap suggests:
pn  squid-deb-proxy-client   
ii  ubuntu-keyring [ubuntu-archive-keyring]  2018.09.18.1-5

-- no debconf information



Bug#939303: matplotlib2: FTBFS unsatisfied b-d on python-pyshp

2019-09-02 Thread Johannes 'josch' Schauer
Source: matplotlib2
Version: 2.2.3-6
Severity: serious
Justification: fails to build from source

Hi,

src:matplotlib2 currently FTBFS because its build dependencies cannot be
satisfied in Debian unstable. It follows some evidence:

$ dose-builddebcheck --explain --failures --deb-native-arch amd64 
--checkonly=matplotlib2 Packages_main_amd64 Sources
output-version: 1.2
native-architecture: amd64
report:
 -
  package: matplotlib2
  version: 2.2.3-6
  architecture: any,all
  type: src
  status: broken
  reasons:
   -
missing:
 pkg:
  package: python-mpltoolkits.basemap
  version: 1.2.0+dfsg-1
  architecture: amd64
  unsat-dependency: python-pyshp:amd64
 depchains:
  -
   depchain:
-
 package: matplotlib2
 version: 2.2.3-6
 architecture: any,all
 type: src
 depends: python-mpltoolkits.basemap:amd64
 
binary-packages: 90294
source-packages: 30649
broken-packages: 1

Or the debcheck results:

https://qa.debian.org/dose/debcheck/src_unstable_main/1567400407/packages/matplotlib2.html

This failure is also important, because matplotlib2 is necessary to
compile build-essential. See the dependency graph here:

http://paste.debian.net/1098490/

Thanks!

cheers, josch



Bug#930212: libmupdf-dev: undefined reference when linking

2019-06-08 Thread Johannes 'josch' Schauer
Package: libmupdf-dev
Version: 1.14.0+ds1-4
Severity: grave
Justification: renders package unusable

Hi,

it does not seem to be possible to use the static library from
libmupdf-dev at all.

Consider the following minimal test case:

#include "mupdf/fitz.h"
#include "mupdf/pdf.h"

#include 
#include 

int main(int argc, char **argv)
{
fz_context *ctx;

ctx = fz_new_context(NULL, NULL, FZ_STORE_UNLIMITED);
if (!ctx)
{
fprintf(stderr, "cannot initialise context\n");
exit(1);
}

return 0;
}

Lets compile it:

$ gcc $(pkg-config --cflags --libs --static mupdf) test.c -o out
/usr/bin/ld: cannot find -lopenjpeg
collect2: error: ld returned 1 exit status

So one bug that should be fixed is that the supplied pkg-config file
isn't working. This should be fixed but is not the RC bug I'm reporting
here. Lets try to pass the options manually:

$ gcc -I/usr/include/freetype2 -I/usr/include/libpng16 -lmupdf -ljbig2dec 
-ljpeg -lz -lm -lfreetype -lpng16 -lm -lz -lm -lz test.c
/usr/bin/ld: /tmp/ccCPwEIx.o: in function `main':
test.c:(.text+0x26): undefined reference to `fz_new_context_imp'
collect2: error: ld returned 1 exit status

So it seems the shipped static library is useless? I hope I'm wrong and
made a silly mistake.

Thanks!

cheers, josch



Bug#928986: CloudCompare: error while loading shared libraries: libQCC_IO_LIB.so:

2019-05-14 Thread Johannes 'josch' Schauer
Package: cloudcompare
Version: 2.10.1-1
Severity: grave
Justification: renders package unusable

Hi,

after installing the package, when attempting to start CloudCompare I
get the following error message:

CloudCompare: error while loading shared libraries: libQCC_IO_LIB.so:
cannot open shared object file: No such file or directory

Indeed that file seems to be contain in no binary package in the
archive?

Thanks!

cheers, josch


-- System Information:
Debian Release: buster/sid
  APT prefers testing
  APT policy: (990, 'testing'), (500, 'unstable-debug'), (500, 'unstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386, armhf

Kernel: Linux 4.19.0-2-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_US.utf8, LC_CTYPE=en_US.utf8 (charmap=UTF-8), LANGUAGE=en_US:en 
(charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

Versions of packages cloudcompare depends on:
ii  libboost-atomic1.67.0 1.67.0-11
ii  libboost-chrono1.67.0 1.67.0-11
ii  libboost-date-time1.67.0  1.67.0-11
ii  libboost-system1.67.0 1.67.0-10
ii  libboost-thread1.67.0 1.67.0-11
ii  libc6 2.28-2
ii  libcgal13 4.12-1
ii  libgcc1   1:8.2.0-9
ii  libgl11.1.0-1
ii  libglu1-mesa [libglu1]9.0.0-2.1
ii  libgmp10  2:6.1.2+dfsg-4
ii  libgmpxx4ldbl 2:6.1.2+dfsg-4
ii  libgomp1  8.2.0-9
ii  libmpfr6  4.0.1-1
ii  libpdal-base7 1.8.0+ds-1+b2
ii  libpdal-util7 1.8.0+ds-1+b2
ii  libqt5concurrent5 5.11.3+dfsg-2
ii  libqt5core5a  5.11.3+dfsg-2
ii  libqt5gui55.11.3+dfsg-2
ii  libqt5opengl5 5.11.3+dfsg-2
ii  libqt5printsupport5   5.11.3+dfsg-2
ii  libqt5widgets55.11.3+dfsg-2
ii  libstdc++68.2.0-9

cloudcompare recommends no packages.

cloudcompare suggests no packages.

-- debconf-show failed



Bug#921924: fakechroot: mv(1) cannot be used inside fakechroot

2019-02-10 Thread Johannes 'josch' Schauer
Package: fakechroot
Version: 2.19-3.2
Severity: grave
Justification: renders package unusable
Control: block -1 by 915559

Hi,

currently, it is not possible to use the mv(1) tool from coreutils
within fakechroot. Since mv is also used in the dash and apt postinst
scripts, this means that:

 - fakechroot is unfit for a debootstrap-like scenario
 - shell scripts using mv will not work inside fakechroot

Since one of the main uses of fakechroot is to be used together with
debootstrap and since the mv tool is such an essential utility, I'm
marking this bug with RC severity.

Steps to reproduce:

$ fakechroot fakeroot -s fakeroot.state debootstrap --variant=fakechroot 
unstable debian-unstable
[...]
W: Failure while configuring required packages.
W: See /home/josch/debian-unstable/debootstrap/debootstrap.log for details 
(possibly the package dash is at fault)

And from the log:

Setting up dash (0.5.10.2-5) ...
No diversion 'diversion of /bin/sh by bash', none removed.
Adding 'diversion of /bin/sh to /bin/sh.distrib by dash'
mv: cannot move '/bin/sh.tmp' to '/bin/sh': No such file or directory
dpkg: error processing package dash (--configure):
 installed dash package post-installation script subprocess returned error 
exit status 1

To reproduce it manually:

fakechroot fakeroot -s fakeroot.state chroot debian-unstable mv /bin/sh.tmp 
/bin/sh
mv: cannot move '/bin/sh.tmp' to '/bin/sh': No such file or directory

The problem is, that coreutils directly invokes the renameat2 systemcall
instead of its glibc wrapper function. fakechroot cannot intercept
system calls, so the bare renameat2 call will be executed. From wrapping
the above in strace:

[pid  8332] renameat2(AT_FDCWD, "/bin/sh.tmp", AT_FDCWD, "/bin/sh", 
RENAME_NOREPLACE) = -1 ENOENT (No such file or directory)

As one can see, the paths did not get adjusted by fakechroot.

There is a bug against coreutils which backports a patch from coreutils
upstream that fixes this problem. The issue is tracked in #915559.

Thanks!

cheers, josch



Bug#920227: Bug #920227 in sbuild marked as pending

2019-01-29 Thread Johannes 'josch' Schauer
Control: tag -1 pending

Hello,

Bug #920227 in sbuild reported by you has been fixed in the
Git repository and is awaiting an upload. You can see the commit
message below and you can check the diff of the fix at:

https://salsa.debian.org/debian/sbuild/commit/97030b80674f723d227f46510dab7d5ef36d2735


fix syntax of generated Sources.gz files (Closes: #920227)

The rewrite of dpkg-scan* performed to fix #909847 introduced a
problem in 18f423619c176471d2adaafb7742cb204951a10c: Sources.gz
entries are not correctly separated by a newline. Furthermore, they
have Source: entries instead of Package:

This confuses older version of APT (previous to jessie) which have
extra sanity checks on the contents of those files, which breaks
building in older chroots.


(this message was generated automatically)
-- 
Greetings

https://bugs.debian.org/920227



Bug#916913: libpcl-dev: PCLConfig.cmake fails to find eigen

2018-12-20 Thread Johannes 'josch' Schauer
Package: libpcl-dev
Version: 1.9.1+dfsg1-2
Severity: grave
Justification: renders package unusable

Hi,

a CMakeLists.txt with "find_package(PCL REQUIRED)" will currently result in:

CMake Warning at /usr/lib/x86_64-linux-gnu/cmake/pcl/PCLConfig.cmake:159 
(find_package):
  By not providing "FindEigen.cmake" in CMAKE_MODULE_PATH this project has
  asked CMake to find a package configuration file provided by "Eigen", but
  CMake did not find one.

  Could not find a package configuration file provided by "Eigen" with any of
  the following names:

EigenConfig.cmake
eigen-config.cmake

  Add the installation prefix of "Eigen" to CMAKE_PREFIX_PATH or set
  "Eigen_DIR" to a directory containing one of the above files.  If "Eigen"
  provides a separate development package or SDK, be sure it has been
  installed.
Call Stack (most recent call first):
  /usr/lib/x86_64-linux-gnu/cmake/pcl/PCLConfig.cmake:313 (find_eigen)
  /usr/lib/x86_64-linux-gnu/cmake/pcl/PCLConfig.cmake:544 
(find_external_library)
  src/ros/CMakeLists.txt:22 (find_package)


CMake Error at /usr/lib/x86_64-linux-gnu/cmake/pcl/PCLConfig.cmake:59 (message):
  common is required but eigen was not found
Call Stack (most recent call first):
  /usr/lib/x86_64-linux-gnu/cmake/pcl/PCLConfig.cmake:362 (pcl_report_not_found)
  /usr/lib/x86_64-linux-gnu/cmake/pcl/PCLConfig.cmake:544 
(find_external_library)
  src/ros/CMakeLists.txt:22 (find_package)


-- Configuring incomplete, errors occurred!



But libeigen3-dev is installed. For a full failing build log see:

https://api.travis-ci.org/v3/job/470351936/log.txt

Thanks!

cheers, josch



Bug#915934: libeigen3-dev: cannot build with Eigen anymore after upgrade of gcc-8 to 8.2.0-11

2018-12-07 Thread Johannes 'josch' Schauer
Package: libeigen3-dev
Version: 3.3.5-2
Severity: grave
Justification: renders package unusable

Hi,

steps to reproduce:

--%<---
#include "eigen3/Eigen/Eigen"
int main(int argc, char* argv[]) { return 0; }
-->%---

$ gcc main.cc
In file included from /usr/include/eigen3/Eigen/SparseCore:50,
 from /usr/include/eigen3/Eigen/Sparse:26,
 from /usr/include/eigen3/Eigen/Eigen:2,
 from main.cc:1:
/usr/include/eigen3/Eigen/src/SparseCore/SparseBlock.h: In member function 
'Eigen::internal::sparse_matrix_block_impl::BlockType& 
Eigen::internal::sparse_matrix_block_impl::operator=(const BlockType&)':
/usr/include/eigen3/Eigen/src/SparseCore/SparseBlock.h:216:33: error: expected 
primary-expression before '>' token
   return operator=(other);
 ^

This did not happen with snapshot 20181207T090635Z where src:gcc-8
packages were still in version 8.2.0-10. It started happening one
dinstall later in snapshot 20181207T153749Z where src:gcc-8 version
8.2.0-11 got uploaded.

Thanks!

cheers, josch



Bug#914378: with Python 3.7: RuntimeError: generator raised StopIteration

2018-11-22 Thread Johannes 'josch' Schauer
Source: python-pygraphviz
Version: 1.4~rc1-1+b2
Severity: grave
Tags: patch
Justification: renders package unusable

Hi,

steps to reproduce:

$ python3 --version
Python 3.7.1
$ python3 -c 'import pygraphviz; A=pygraphviz.AGraph(); A.graph_attr.keys()'
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/pygraphviz/agraph.py", line 1743, in 
iteritems
ah = gv.agnxtattr(self.handle, self.type, ah)
StopIteration: agnxtattr

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "", line 1, in 
  File "/usr/lib/python3/dist-packages/pygraphviz/agraph.py", line 1733, in keys
return list(self.__iter__())
  File "/usr/lib/python3/dist-packages/pygraphviz/agraph.py", line 1736, in 
__iter__
for (k, v) in self.iteritems():
RuntimeError: generator raised StopIteration

This problem does not happen with snapshot 20181121T102052Z but does
happen with snapshot 20181121T232318Z. There is only one change between
chroots made from these two snapshots, the following packages have been
upgraded from version 3.6.7-1 to 3.7.1-1:

libpython3-stdlib, libpython3.6-minimal, libpython3.6-stdlib, python3,
python3-minimal, python3.6, python3.6-minimal

Thus I conclude that this problem was introduced because of the upgrade
of Python from 3.6 to 3.7.

Upstream has a fix here:

https://github.com/pygraphviz/pygraphviz/commit/b5df022700669ae496f65d20dd9cd387d6af948e

I backported that commit on top of the version of python-pygraphviz from
Debian unstable. Please find the patch attached.

I see that this package did not see an upload since January 2017. If you
are okay with me NMU-ing the package for this fix, then please tell me.

Alternatively, this bug can also be fixed by packaging the latest
upstream version 1.5 of pygraphviz which includes the above commit.

Thanks!

cheers, josch
diff -Nru python-pygraphviz-1.4~rc1/debian/changelog 
python-pygraphviz-1.4~rc1/debian/changelog
--- python-pygraphviz-1.4~rc1/debian/changelog  2017-01-08 21:03:20.0 
+0100
+++ python-pygraphviz-1.4~rc1/debian/changelog  2018-11-22 19:31:03.0 
+0100
@@ -1,3 +1,10 @@
+python-pygraphviz (1.4~rc1-1.1) UNRELEASED; urgency=medium
+
+  * Non-maintainer upload.
+  * Fix StopIteration with Python 3.7
+
+ -- Johannes 'josch' Schauer   Thu, 22 Nov 2018 19:31:03 
+0100
+
 python-pygraphviz (1.4~rc1-1) unstable; urgency=medium
 
   * New upstream release candidate
diff -Nru 
python-pygraphviz-1.4~rc1/debian/patches/catch_stopiterations_created_by_c_code 
python-pygraphviz-1.4~rc1/debian/patches/catch_stopiterations_created_by_c_code
--- 
python-pygraphviz-1.4~rc1/debian/patches/catch_stopiterations_created_by_c_code 
1970-01-01 01:00:00.0 +0100
+++ 
python-pygraphviz-1.4~rc1/debian/patches/catch_stopiterations_created_by_c_code 
2018-11-22 19:30:55.0 +0100
@@ -0,0 +1,174 @@
+From b5df022700669ae496f65d20dd9cd387d6af948e Mon Sep 17 00:00:00 2001
+From: Dan Schult 
+Date: Thu, 2 Aug 2018 21:32:47 -0400
+Subject: [PATCH] catch StopIterations created by C code
+
+--- a/pygraphviz/agraph.py
 b/pygraphviz/agraph.py
+@@ -374,8 +374,10 @@ class AGraph(object):
+ nh = gv.agfstnode(self.handle)
+ while nh is not None:
+ yield Node(self, nh=nh)
+-nh = gv.agnxtnode(self.handle, nh)
+-raise StopIteration
++try:
++nh = gv.agnxtnode(self.handle, nh)
++except StopIteration:
++return
+ 
+ iternodes = nodes_iter
+ 
+@@ -597,8 +599,10 @@ class AGraph(object):
+ yield Node(self, t)
+ else:
+ yield Node(self, s)
+-eh = gv.agnxtedge(self.handle, eh, nh)
+-raise StopIteration
++try:
++eh = gv.agnxtedge(self.handle, eh, nh)
++except StopIteration:
++return
+ 
+ def neighbors(self, n):
+ """Return a list of the nodes attached to n."""
+@@ -627,8 +631,14 @@ class AGraph(object):
+ yield (e[0], e[1], e.name)
+ else:
+ yield e
+-eh = gv.agnxtout(self.handle, eh)
+-nh = gv.agnxtnode(self.handle, nh)
++try:
++eh = gv.agnxtout(self.handle, eh)
++except StopIteration:
++break
++try:
++nh = gv.agnxtnode(self.handle, nh)
++except StopIteration:
++return
+ elif nbunch in self: # if nbunch is a single node
+ n = Node(self, nbunch)
+ nh = n.handle
+@@ -639,7 +649,10 @@ class AGraph(object):
+ yield (e[0], e[1], e.name)
+ else:
+ yield e
+-eh = gv.agnxtout(self.handle, eh)
++ 

Bug#913803: python3-pykdl: PyKDL fails to import

2018-11-15 Thread Johannes 'josch' Schauer
Package: python3-pykdl
Version: 1.4.0-6
Severity: grave
Justification: renders package unusable

Steps to reproduce:

$ python3
Python 3.6.6 (default, Jun 27 2018, 14:44:17)
[GCC 8.1.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from PyKDL import *
Traceback (most recent call last):
  File "", line 1, in 
ModuleNotFoundError: No module named 'sip'


Probably just a missing dependency?


-- System Information:
Debian Release: buster/sid
  APT prefers testing
  APT policy: (990, 'testing'), (500, 'unstable-debug'), (500, 'unstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386, armhf

Kernel: Linux 4.17.0-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_US.utf8, LC_CTYPE=en_US.utf8 (charmap=UTF-8), LANGUAGE=en_US:en 
(charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

Versions of packages python3-pykdl depends on:
ii  libc6 2.27-6
ii  libgcc1   1:8.2.0-8
ii  liborocos-kdl1.4  1.4.0-5
ii  libpython3.6  3.6.6-1
ii  libstdc++68.2.0-8
ii  python3   3.6.6-1

python3-pykdl recommends no packages.

python3-pykdl suggests no packages.

-- no debconf information



Bug#909637: mmdebstrap: cannot install apt under fakechroot or proot

2018-09-25 Thread Johannes 'josch' Schauer
Package: mmdebstrap
Severity: grave
Justification: renders package unusable
Control: block -1 by 909612 909581

Apt uses 'mv' in its post inst. 'mv' from coreutils uses the renameat2
systemcall. The renameat2 systemcall is not supported by fakechroot or
proot.