This essentially unbreaks deploying new Hetzner machines with NixOps,
because the Hetzner robot has changed its way of handling admin
accounts.
It also now provides a more helpful error message (instead of an
AssertionError) if admin account creation has failed.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Reported-by: Graham Christensen <graham@grahamc.com>
Issue: https://github.com/NixOS/nixops/issues/563
While not explicitly checked by setup.py or by the "chkdeps" command
from the project I have added pyinsane2 and pyocr to the list of
dependencies as well, because they're referenced in the source.
Tested by building against Python 3.3, 3.4, 3.5 and 3.6.
The build against Python 3.6 failed because pycairo doesn't build, so
it's a non-issue at least for paperwork-backend.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
First of all: This is NOT the same package as "pillowfight".
I'm not sure why people want to choose this particular name, but well,
so be it.
I haven't investigated why test_ace and test_all_2 fail, but I've
disabled these tests by now and reported the failures upstream at
jflesch/libpillowfight#2.
Tested by building against Python 3.3, 3.4, 3.5 and 3.6.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This package is a bit more involved because it assumes a lot of paths
being there in a FHS compliant way, so we need to patch the data and
binary directories for Tesseract and Cuneiform.
I've also tried to get the tests working, but they produce different
results comparing input/output. This is probably related to the
following issue:
https://github.com/jflesch/pyocr/issues/52
So I've disabled certain tests that fail but don't generally impede the
functionality of pyocr.
Tested by building against Python 3.3, 3.4, 3.5 and 3.6.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
The tests require a scanner to be physically attached.
Quote from the upstream README:
> Tests require at least one scanner with a flatbed and an ADF
> (Automatic Document Feeder).
>
> If possible, they should be run with at least 2 scanners connected.
> The first that appear in "scanimage -L" must be the one with the ADF.
>
> For reference, my current setup is:
>
> - HP Officejet 4620 (Flatbed + ADF)
> - HP Deskjet 2050 J510 series (Flatbed)
So we disable the tests even though it might be theoretically possible
to use qemu and an emulated scanner. Instead of the upstream tests we
just do a quick check whether initialization of the library succeeds.
Other than that the library uses ctypes.cdll to dlopen() the libsane
shared library, so we need to patch in the right store path.
Tested by building against Python 2.7, 3.3, 3.4, 3.5 and 3.6.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
The upstream tag actually says 1.5.7 but the commit actually bumps the
version to 1.5.8:
https://github.com/hickeroar/simplebayes/commit/b8da72c50d20b6f8c0d
We needed to patch the setup.py because the upstream project's setup.py
reads in the README.rst for the longDescription. That very README.rst
contains non-ASCII characters which in turn throws a decoding error with
Python 3 on Nix because I think this has to do with our setup.py wrapper
that doesn't seem to recognize the right encoding when using compile().
Tested by building against Python 2.7, 3.3, 3.4, 3.5 and 3.6.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Upstream changelog without issue numbers:
* Fixes several issues with C++ template deduction.
* Fixes a issue with bound method type inference.
* Fixes a bug with cascaded tuple assignment.
* Fixed or silenced many Clang warnings.
* Fixes bug with powers of pure real complex numbers.
The full changelog with issue numbers can be found here:
https://github.com/cython/cython/blob/0.25.2/CHANGES.rst
My main reason for updating is because there were test failures on
i686-linux, although version 0.25.2 still has one test that fails.
So if we're on i686-linux and on Python 2 we just fix that one little
doctest.
The test failure has already been reported upstream at:
https://github.com/cython/cython/issues/1548
All of the failing tests (including the latter) had to do with integer
representations in that long integers are suffixed by an L while the
test cases weren't expecting this.
Built successfully on i686-linux and x86_64-linux against Python 2.7 and
Python 3.5.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Without this we get the following Python exception when trying to fetch
a graph in the graphite web app:
File "/nix/store/nj62jqk2xmp5c3h93pfnlqn66qj1kkvs-python-2.7.12-env/lib/python2.7/site-packages/opt/graphite/webapp/graphite/storage.py", line 335, in fetch
return whisper.fetch(self.fs_path, startTime, endTime, now)
TypeError: fetch() takes at most 3 arguments (4 given)
Fixes#21032.
Use a fixed-point combinator for the Python package set to allow easier overriding of its contents.
Earlier implementations were proposed in #16784 and #17428. This commit is by comparison much smaller
and changes only what is needed.
There's no reason to disable ALL tests just because only one particular
test module is failing.
Tested on i686-linux and x86_64-linux against these Python versions:
Python 2.6: The interpreter itself doesn't build
Python 2.7: Successful for both architectures
Python 3.3: Successful for both architectures
Python 3.4: Successful for both architectures
Python 3.5: Successful for both architectures
Python 3.6: One of the dependencies of pillow doesn't build (pytest)
Tests for PyPy still fail, which is why the doCheck attribute is only
set to false if we're building for PyPy.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Cc: @desiderius, @goibhniu, @prikhi
pytest plugins should not propagate pytest. Instead, packages depending
on pytest and plugins, should explicitly depend on both the plugin(s)
and pytest. This change will more easily allow packages to depend on
another version of pytest when needed.
- rename to PyLTI to follow upstream
- remove overriding because it is not necessary; newer mock can be used,
and modules python packages should be fixed to not propagate pytest (see
separate commit).
cc maintainer @layus
We have two derivations, one that supports Cuda, and one that does not.
The names, TheanoWithCuda and TheanoWithoutCuda, now reflect that.
Furthermore, a boolean passthru.cudaSupport was added.
In the future the two derivations should be merged in one, with a
parameter `cudaSupport`.
