Bug#893618: please suggest ffmpeg for video reader

[Lumin]

Package: python3-imageio
Version: 2.2.0-1
Severity: minor

When ffmpeg is not installed, running

   reader = imageio.get_reader('imageio:cockatoo.mp4', 'ffmpeg')

will cause the following error:

   NeedDownloadError: Need ffmpeg exe. You can obtain it with either:
  - install using conda: conda install ffmpeg -c conda-forge
  - download by calling: imageio.plugins.ffmpeg.download()

which could be solved by installing ffmpeg.

-- 
Best,

-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers

Bug#889539: pandas FTBFS: test failures

[Lumin]

Yaroslav has just uploaded pandas 0.22.0, let's see if this problem
still exists.

On 7 February 2018 at 14:57, Andreas Tille <andr...@an3as.eu> wrote:
> On Tue, Feb 06, 2018 at 03:51:27PM +0000, Lumin wrote:
>> Apart from that, the pandas packaging needs a patch [2] to reduce
>> autopkgtest failures.
>
> As always:  Please push your patches. :-)
> You and I are in the same position: We are team members of the maintainer
> team.  I even have the strong impression, that you are the more competent
> team member than me in terms of pandas.  So please do not be shy. ;-)
>
> Thanks a lot and feel to keep on pushing promising patches
>
> Andreas.
>
>> [2]
>> diff --git a/debian/tests/control b/debian/tests/control
>> index 38521c8..ab54101 100644
>> --- a/debian/tests/control
>> +++ b/debian/tests/control
>> @@ -15,6 +15,7 @@ Depends: python-all,
>>   python-xlwt,
>>   python-bs4,
>>   python-html5lib,
>> + python-pytest,
>>   xauth,
>>   xvfb
>>
>> @@ -29,5 +30,6 @@ Depends: python3-all,
>>   python3-tk,
>>   python3-tz,
>>   python3-bs4,
>> + python3-pytest,
>>   xauth,
>>   xvfb
>
> --
> http://fam-tille.de



-- 
Best,

-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers

Bug#889539: pandas FTBFS: test failures

[Lumin]

Hi Andreas,

On 6 February 2018 at 11:10, Andreas Tille  wrote:

> you did a really good job on latest pandas issues.  Do you think you can
> have a look at this problem as well?

I found no related upsteam issue about the failed tests. However there are
some links that might be useful:

https://stackoverflow.com/questions/40659212/futurewarning-elementwise-comparison-failed-returning-scalar-but-in-the-futur
https://stackoverflow.com/questions/15777951/how-to-suppress-pandas-future-warning

I noticed these lines [1] in the test log. The failures seem to be
numpy issues again
according to the stackoverflow answers.

Apart from that, the pandas packaging needs a patch [2] to reduce
autopkgtest failures.


> BTW, to upgrade pandas to the latest upstream version python-moto is
> needed.  I prepared the packaging (#777089) and may be there is only a
> minor issue to finalise this (see [1]).  It might be a sensible strategy
> To upgrade to latest pandas upstream to not spent to much time on the
> old code - however, I have no idea about compatibility issues of all of
> its rdepends.

I checked and tried to build that package. Feedbacks were sent in
another mail. I think the 0.20.3-12 upload is not necessary as long as
we manage to build the new pandas before the removal, because
there is no obvious bug in pandas itself -- the patch for -12 will be
workarounds to those external problems.


[1] 
debian/tmp/usr/lib/python2.7/dist-packages/pandas/tests/dtypes/test_missing.py::test_array_equivalent_compat
  /usr/lib/python2.7/dist-packages/numpy/core/numeric.py:2604:
FutureWarning: elementwise == comparison failed and returning scalar
instead; this will raise an error or perform elementwise comparison in
the future.
return bool(asarray(a1 == a2).all())

debian/tmp/usr/lib/python2.7/dist-packages/pandas/tests/frame/test_analytics.py::TestDataFrameAnalytics::()::test_corr_int_and_boolean
  /usr/lib/python2.7/dist-packages/scipy/stats/stats.py:3577:
RuntimeWarning: invalid value encountered in double_scalars
size * (size - 1) * (size - 2))

debian/tmp/usr/lib/python2.7/dist-packages/pandas/tests/plotting/test_boxplot_method.py::TestDataFramePlots::()::test_boxplot_legacy
  /usr/lib/python2.7/dist-packages/numpy/core/fromnumeric.py:57:
FutureWarning: reshape is deprecated and will raise in a subsequent
release. Please use .values.reshape(...) instead
return getattr(obj, method)(*args, **kwds)
  /usr/lib/python2.7/dist-packages/numpy/core/fromnumeric.py:57:
FutureWarning: reshape is deprecated and will raise in a subsequent
release. Please use .values.reshape(...) instead
return getattr(obj, method)(*args, **kwds)
  /usr/lib/python2.7/dist-packages/numpy/core/fromnumeric.py:57:
FutureWarning: reshape is deprecated and will raise in a subsequent
release. Please use .values.reshape(...) instead
return getattr(obj, method)(*args, **kwds)
  /usr/lib/python2.7/dist-packages/numpy/core/fromnumeric.py:57:
FutureWarning: reshape is deprecated and will raise in a subsequent
release. Please use .values.reshape(...) instead
return getattr(obj, method)(*args, **kwds)
  /usr/lib/python2.7/dist-packages/numpy/core/fromnumeric.py:57:
FutureWarning: reshape is deprecated and will raise in a subsequent
release. Please use .values.reshape(...) instead
return getattr(obj, method)(*args, **kwds)
  /usr/lib/python2.7/dist-packages/numpy/core/fromnumeric.py:57:
FutureWarning: reshape is deprecated and will raise in a subsequent
release. Please use .values.reshape(...) instead
return getattr(obj, method)(*args, **kwds)
  /usr/lib/python2.7/dist-packages/numpy/core/fromnumeric.py:57:
FutureWarning: reshape is deprecated and will raise in a subsequent
release. Please use .values.reshape(...) instead
return getattr(obj, method)(*args, **kwds)
  /usr/lib/python2.7/dist-packages/numpy/core/fromnumeric.py:57:
FutureWarning: reshape is deprecated and will raise in a subsequent
release. Please use .values.reshape(...) instead
return getattr(obj, method)(*args, **kwds)

debian/tmp/usr/lib/python2.7/dist-packages/pandas/tests/plotting/test_boxplot_method.py::TestDataFramePlots::()::test_boxplot_axis_limits
  /usr/lib/python2.7/dist-packages/numpy/core/fromnumeric.py:57:
FutureWarning: reshape is deprecated and will raise in a subsequent
release. Please use .values.reshape(...) instead
return getattr(obj, method)(*args, **kwds)

debian/tmp/usr/lib/python2.7/dist-packages/pandas/tests/plotting/test_datetimelike.py::TestTSPlot::()::test_irreg_hf
  /usr/lib/python2.7/dist-packages/matplotlib/cbook/deprecation.py:106:
MatplotlibDeprecationWarning: Adding an axes using the same arguments
as a previous axes currently reuses the earlier instance.  In a future
version, a new instance will always be created and returned.
Meanwhile, this warning can be suppressed, and the future behavior
ensured, by passing a unique label to each axes instance.
warnings.warn(message, mplDeprecation, stacklevel=1)
  

Bug#884294: Patch for pandas RC 884294, someone to sponsor?

[Lumin]

> I'd volunteer to upload once it is pushed.  Please ping me after pushing
> since there is not commit mailing list and I'm not sure whether the
> tracker solution is implemented yet.

Done.
https://salsa.debian.org/science-team/pandas/commits/debian

-- 
Best,

-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers

Bug#884294: Patch for pandas RC 884294, someone to sponsor?

[Lumin]

Control: tags -1 +patch

Pandas FTBFS on amd64 due to test failure:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=884294
I reproduced this FTBFS on my machine (sid/amd64).

This test failure was due to a numpy problem. Upstream workaround
(bypass) is available:
* 
https://github.com/pandas-dev/pandas/commit/5f2b96bb637f6ddeec169c5ef8ad20013a03c853

Another test failure arose (not related to #884924). The new failure
could be solved by referencing:
* https://github.com/pandas-dev/pandas/issues/17046
* https://github.com/pandas-dev/pandas/pull/17051

For details about more test failures arose when I'm writing this email please
check the patches.

debomatic-amd64 build was successful:
http://debomatic-amd64.debian.net/distribution#unstable/pandas/0.20.3-11/buildlog

git-formatted patches are attached. I can push the commits to salsa if
the fix is fine.
Can someone review and sponsor the Team upload?
Thanks.

pandas (0.20.3-11) unstable; urgency=medium

  * Team upload.
  * Cherry-pick upstream commit 5f2b96bb637f6ddeec169c5ef8ad20013a03c853
to workaround a numpy bug. (Closes: #884294)
+ patches/up_tst_dont_assert_that_a_bug_exists_in_numpy
  * Cherry-pick upstream commits to fix test failure caused by test_argsort().
+ patches/up_tst_np_argsort_comparison2
  * Workaround test failure of test_basic_indexing() in file
pandas/tests/series/test_indexing.py .
+ patches/deb_fix_test_failure_test_basic_indexing



-- 
Best,
From 23745eb17f921b42e1b3ee574c2aef0f05d4252f Mon Sep 17 00:00:00 2001
From: Mo Zhou 
Date: Sat, 20 Jan 2018 09:02:43 +
Subject: [PATCH 1/7] Cherry-pick upstream commit
 5f2b96bb637f6ddeec169c5ef8ad20013a03c853 to workaround a numpy bug. (Closes:
 #884294)

---
 debian/patches/series  |  1 +
 .../up_tst_dont_assert_that_a_bug_exists_in_numpy  | 33 ++
 2 files changed, 34 insertions(+)
 create mode 100644 debian/patches/up_tst_dont_assert_that_a_bug_exists_in_numpy

diff --git a/debian/patches/series b/debian/patches/series
index b0c042d..209965f 100644
--- a/debian/patches/series
+++ b/debian/patches/series
@@ -9,6 +9,7 @@ deb_skip_sequencelike_on_armel
 # 858260.patch
 up_print_versions
 up_network_seaborn
+up_tst_dont_assert_that_a_bug_exists_in_numpy
 dateutil-2.6.1-fixed-ambiguous-tz-dst-be.patch
 mark_tests_working_on_intel.patch
 mark_tests_working_on_intel_s390x.patch
diff --git a/debian/patches/up_tst_dont_assert_that_a_bug_exists_in_numpy b/debian/patches/up_tst_dont_assert_that_a_bug_exists_in_numpy
new file mode 100644
index 000..4242da7
--- /dev/null
+++ b/debian/patches/up_tst_dont_assert_that_a_bug_exists_in_numpy
@@ -0,0 +1,33 @@
+From 5f2b96bb637f6ddeec169c5ef8ad20013a03c853 Mon Sep 17 00:00:00 2001
+From: Eric Wieser 
+Date: Sat, 15 Jul 2017 13:30:03 +0100
+Subject: [PATCH] TST: Don't assert that a bug exists in numpy (#16940)
+
+Better to ignore the warning from the bug, rather than assert the bug is still there
+
+After this change, numpy/numpy#9412 _could_ be backported to fix the bug
+---
+ pandas/tests/test_algos.py | 3 ++-
+ 1 file changed, 2 insertions(+), 1 deletion(-)
+
+diff --git a/pandas/tests/test_algos.py b/pandas/tests/test_algos.py
+index 9504d2a9426..993dcc4f527 100644
+--- a/pandas/tests/test_algos.py
 b/pandas/tests/test_algos.py
+@@ -2,6 +2,7 @@
+ 
+ import numpy as np
+ import pytest
++import warnings
+ 
+ from numpy.random import RandomState
+ from numpy import nan
+@@ -127,7 +128,7 @@ def test_unsortable(self):
+ arr = np.array([1, 2, datetime.now(), 0, 3], dtype=object)
+ if compat.PY2 and not pd._np_version_under1p10:
+ # RuntimeWarning: tp_compare didn't return -1 or -2 for exception
+-with tm.assert_produces_warning(RuntimeWarning):
++with warnings.catch_warnings():
+ pytest.raises(TypeError, algos.safe_sort, arr)
+ else:
+ pytest.raises(TypeError, algos.safe_sort, arr)
-- 
2.15.1

From b92f86abb26073326ec80dffaa017c1e3f77eab5 Mon Sep 17 00:00:00 2001
From: Mo Zhou 
Date: Sat, 20 Jan 2018 09:04:32 +
Subject: [PATCH 2/7] dch: finalize, upload to unstable

---
 debian/changelog | 8 
 1 file changed, 8 insertions(+)

diff --git a/debian/changelog b/debian/changelog
index bed979e..53b0deb 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,11 @@
+pandas (0.20.3-11) unstable; urgency=medium
+
+  * Team upload.
+  * Cherry-pick upstream commit 5f2b96bb637f6ddeec169c5ef8ad20013a03c853
+to workaround a numpy bug. (Closes: #884294)
+
+ -- Mo Zhou   Sat, 20 Jan 2018 09:00:31 +
+
 pandas (0.20.3-10) unstable; urgency=medium
 
   * Team upload.
-- 
2.15.1

From c37cee1b4a9c44522c6bf3d74bceb8f0b20364cd Mon Sep 17 00:00:00 2001
From: Mo Zhou 
Date: Sat, 20 Jan 2018 10:12:15 +
Subject: [PATCH 3/7] Cherry-pick upstream commit to fix test failure caused by

Bug#873586: lua-torch-torch7: please add arm64

[Lumin]

Hi,

This package was enabled on arm64 in the past. However it fails
to build on it recently so I disabled it. I have no arm64 device
so I cannot test that. Could you please test this package on
arm64?

Reference:

https://buildd.debian.org/status/logs.php?pkg=lua-torch-torch7

https://github.com/torch/torch7/issues/1035

-- 
Best,

-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers

Bug#873405: Please update to llvm-toolchain 4.0 or, better, 5.0

[Lumin]

Control: clone -1 -2
Control: reassign -2 nvidia-cuda-toolkit 8.0.44-4
Control: retitle -2 cuda: moving away from clang-3.8 to gcc-6 ?
Control: severity -2 important

Hi,

The CUDA toolkit will not be ready for Clang-4.0 . As claimed by Nvidia,
the highest version that CUDA 9.0 (will be released soon) supports
is Clang-3.9 . Since the Clang maintainer would like to remove
Clang-3.8 and Clang-3.9, the cuda package cannot depend on
them anymore.

The current version (CUDA 8.0.44) supports GCC (<= 5) and
Clang (<= 3.8). GCC-5 is still available in the archive.
We can unblock the ROM of clang-3.8 by temporarily by
switching back to GCC-5 but that will block the ROM of
GCC-5 in the future.

CUDA 9.0.103 (release-candidate) supports GCC-6
and Clang-3.9 . Clang-3.9 is now suffering from the GCC-7
headers as shown at bottom. GCC-6 works with CUDA 9 compiling
the examples provided by upstream, although some CUDA
applications seems not ready for CUDA 9.0 .

So if doko is about to also keep GCC-6 in the archive for a long time,
we should switch the compiler to GCC-6 for the CUDA 9.0 upload.

-

❯❯❯ nvcc -ccbin clang++-3.9 matrixMul.cu -I../../common/inc
   Mon 28 Aug 2017 04:36:32 AM UTC
/usr/bin/../lib/gcc/x86_64-linux-gnu/7.2.0/../../../../include/c++/7.2.0/bits/std_abs.h(101):
error: identifier "__float128" is undefined

/usr/bin/../lib/gcc/x86_64-linux-gnu/7.2.0/../../../../include/c++/7.2.0/bits/std_abs.h(102):
error: identifier "__float128" is undefined

2 errors detected in the compilation of
"/tmp/tmpxft_2c27_-8_matrixMul.cpp1.ii".

❯❯❯ nvcc -ccbin clang++-4.0 matrixMul.cu -I../../common/inc
   439ms  Mon 28 Aug 2017 04:36:34 AM UTC
nvcc fatal   : The version ('40001') of the host compiler ('clang') is
not supported

❯❯❯ nvcc --version
   Mon 28 Aug 2017 04:47:22 AM UTC
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2017 NVIDIA Corporation
Built on Mon_Jun_26_16:13:28_CDT_2017
Cuda compilation tools, release 9.0, V9.0.102

-- 
Best,
Lumin

-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers

Bug#870029: Missing new dependency "moses"

[Lumin]

Package: lua-torch-nn
Version: 0~20170726-gf613412+dfsg-1
Severity: serious

Recently the upstream added a new package "moses"
as its runtime dependency, which is not packaged
for Debian.

The missing of 'moses' package renders 'lua-torch-nn'
unusable.

-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers

Bug#864365: A simple program ends up with SIGSEGV [amd64]

[Lumin]

Control: tag -1 +patch

The fix is attached. It fails on my side simply
because the python3-llvmlite package is too
old.

The default version in sid (0.16.0) does not
work with numba.

Listing... Done
python3-llvmlite/experimental,now 0.18.0-1 amd64 [installed]
python3-llvmlite/unstable 0.16.0-1 amd64


diff --git a/debian/control b/debian/control
index 66d7146..81f0e06 100644
--- a/debian/control
+++ b/debian/control
@@ -31,7 +31,7 @@ Depends:
  ${misc:Depends},
  ${python:Depends},
  ${shlibs:Depends},
- python-llvmlite
+ python-llvmlite (>= 0.18.0~),
 Recommends:
  numba-doc
 Suggests:
@@ -55,7 +55,7 @@ Depends:
  ${misc:Depends},
  ${python3:Depends},
  ${shlibs:Depends},
- python3-llvmlite
+ python3-llvmlite (>= 0.18.0~),
 Recommends:
  numba-doc
 Suggests:

On 2 July 2017 at 13:18, Lumin <cdlumin...@gmail.com> wrote:
> The numba package installed with pip3 seems working.
> It is able to run numba.runtests.main()
>
> However the Debian package will segfault immediately
> if you do so.
>
> On 7 June 2017 at 16:23, Daniel Stender <deb...@danielstender.com> wrote:
>> On 07.06.2017 18:13, Lumin wrote:
>>> Oops, that's a sorrowful story ...
>>
>> I've put it on RFA now (#864367). If you would like to pick it up, welcome!
>>
>> Dan
>>
>> --
>> 4096R/DF5182C8
>> Debian Developer (sten...@debian.org)
>> http://www.danielstender.com/
>
>
>
> --
> Best,
> Lumin



-- 
Best,
Lumin

-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers

Bug#864365: A simple program ends up with SIGSEGV [amd64]

[Lumin]

The numba package installed with pip3 seems working.
It is able to run numba.runtests.main()

However the Debian package will segfault immediately
if you do so.

On 7 June 2017 at 16:23, Daniel Stender <deb...@danielstender.com> wrote:
> On 07.06.2017 18:13, Lumin wrote:
>> Oops, that's a sorrowful story ...
>
> I've put it on RFA now (#864367). If you would like to pick it up, welcome!
>
> Dan
>
> --
> 4096R/DF5182C8
> Debian Developer (sten...@debian.org)
> http://www.danielstender.com/



-- 
Best,
Lumin

-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers

Bug#864365: A simple program ends up with SIGSEGV [amd64]

[Lumin]

Oops, that's a sorrowful story ...

On 7 June 2017 at 16:01, Daniel Stender <deb...@danielstender.com> wrote:
> On 07.06.2017 17:52, Lumin wrote:
>> Package: python3-numba
>> Version: 0.33.0-1
>> Severity: severe
>>
>> Hi,
>>
>> It fails to run even the simplest code:
>>
>> ```
>> #!/usr/bin/python3
>> from numba import jit
>> from numpy import arange
>> import time
>>
>> @jit
>> def sum2d(arr):
>> M, N = arr.shape
>> result = 0.0
>> for i in range(M):
>> for j in range(N):
>> result += arr[i,j]
>> return result
>>
>> a = arange(65536).reshape(256,256)
>> t1 = time.time()
>> print(sum2d(a))
>> t2 = time.time()
>> ```
>>
>> terminated by signal SIGSEGV (Address boundary error)
>
> Yup. That's why it's still in experimental ... :-)
>
> DS
>
> --
> 4096R/DF5182C8
> Debian Developer (sten...@debian.org)
> http://www.danielstender.com/



-- 
Best,
Lumin

-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers

Bug#864365: A simple program ends up with SIGSEGV [amd64]

[Lumin]

Package: python3-numba
Version: 0.33.0-1
Severity: severe

Hi,

It fails to run even the simplest code:

```
#!/usr/bin/python3
from numba import jit
from numpy import arange
import time

@jit
def sum2d(arr):
M, N = arr.shape
result = 0.0
for i in range(M):
for j in range(N):
result += arr[i,j]
return result

a = arange(65536).reshape(256,256)
t1 = time.time()
print(sum2d(a))
t2 = time.time()
```

terminated by signal SIGSEGV (Address boundary error)

-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers

Bug#862528: libcaffe-cuda-dev: libproto.a missing in libcaffe-cuda-dev

[lumin]

Hi Erik,

Thank you for this bug report.

I think I will change the name of this static library
since its name `libproto.a` is too ambiguous -- maybe
libcaffeproto.a or something else.

-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers

Bug#851578: add nmu patch for this bug, however encountered FTBFS due to testsuite failure

[lumin]

Control: tags -1 +patch

Here is an NMU patch for this package. I intended to file an
RFS but the debomatic build failure stopped me to do so.

The build failure was caused by testsuite failure:
http://debomatic-amd64.debian.net/distribution#experimental/numba/0.30.0-2.1/buildlog

diff -Nru numba-0.30.0/debian/changelog numba-0.30.0/debian/changelog
--- numba-0.30.0/debian/changelog	2016-12-29 18:50:40.0 +
+++ numba-0.30.0/debian/changelog	2017-02-17 12:27:37.0 +
@@ -1,3 +1,12 @@
+numba (0.30.0-2.1) experimental; urgency=medium
+
+  * Non-maintainer / Team upload.
+  * Remove unwanted file "annotation_usecases.py" for the python2 package.
+(Closes: #851578)
+  * Add basic autopkgtest support.
+
+ -- Zhou Mo   Fri, 17 Feb 2017 12:27:37 +
+
 numba (0.30.0-2) experimental; urgency=medium
 
   * deb/control:
diff -Nru numba-0.30.0/debian/rules numba-0.30.0/debian/rules
--- numba-0.30.0/debian/rules	2016-12-29 18:16:39.0 +
+++ numba-0.30.0/debian/rules	2017-02-17 12:22:23.0 +
@@ -7,6 +7,11 @@
 %:
 	dh $@ --with python2,python3,sphinxdoc --buildsystem=pybuild
 
+override_dh_auto_install:
+	dh_auto_install
+	# remove the unwanted file for python2 package
+	find debian/python-numba -type f -name annotation_usecases.py -delete
+
 override_dh_auto_test:
 	PYBUILD_SYSTEM=custom PYBUILD_TEST_ARGS="cd {build_dir} && {interpreter} -Wd -m pytest numba/tests -v -rs" dh_auto_test
 
diff -Nru numba-0.30.0/debian/tests/control numba-0.30.0/debian/tests/control
--- numba-0.30.0/debian/tests/control	1970-01-01 00:00:00.0 +
+++ numba-0.30.0/debian/tests/control	2017-02-17 12:24:02.0 +
@@ -0,0 +1,3 @@
+Tests: simpletest.py2.sh
+
+Tests: simpletest.py3.sh
diff -Nru numba-0.30.0/debian/tests/simpletest.py2.sh numba-0.30.0/debian/tests/simpletest.py2.sh
--- numba-0.30.0/debian/tests/simpletest.py2.sh	1970-01-01 00:00:00.0 +
+++ numba-0.30.0/debian/tests/simpletest.py2.sh	2017-02-17 12:25:58.0 +
@@ -0,0 +1,22 @@
+#!/bin/sh
+set -e
+
+# example taken from http://numba.pydata.org/
+python2 -c "
+from numba import jit
+from numpy import arange
+
+# jit decorator tells Numba to compile this function.
+# The argument types will be inferred by Numba when function is called.
+@jit
+def sum2d(arr):
+M, N = arr.shape
+result = 0.0
+for i in range(M):
+for j in range(N):
+result += arr[i,j]
+return result
+
+a = arange(9).reshape(3,3)
+print(sum2d(a))
+"
diff -Nru numba-0.30.0/debian/tests/simpletest.py3.sh numba-0.30.0/debian/tests/simpletest.py3.sh
--- numba-0.30.0/debian/tests/simpletest.py3.sh	1970-01-01 00:00:00.0 +
+++ numba-0.30.0/debian/tests/simpletest.py3.sh	2017-02-17 12:26:07.0 +
@@ -0,0 +1,22 @@
+#!/bin/sh
+set -e
+
+# example taken from http://numba.pydata.org/
+python3 -c "
+from numba import jit
+from numpy import arange
+
+# jit decorator tells Numba to compile this function.
+# The argument types will be inferred by Numba when function is called.
+@jit
+def sum2d(arr):
+M, N = arr.shape
+result = 0.0
+for i in range(M):
+for j in range(N):
+result += arr[i,j]
+return result
+
+a = arange(9).reshape(3,3)
+print(sum2d(a))
+"
-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers

Bug#844509: [openblas/custom build] please consider appending -march&-mtune parameters when building custom package

[lumin]

Package: libopenblas-base
Version: 0.2.19-1
Severity: minor

Hi,

BLAS is performance sensitive software, would you mind adding

  -march=native -mtune==native

to CFLAGS when building custom package with the debian source?

Thanks.

-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers

Re: lua-torch-trepl_0~20160613-g06128f9-1_amd64.changes REJECTED

[lumin]

On Tue, 2016-09-13 at 22:00 +, Thorsten Alteholz wrote:
> Hi,
> 
> unfortunately I have to reject your package. Please mention readline.c 
> in your debian/coypright.

Thank you for pointing this out. That file is published under
the MIT/Expat licence, where the copyright holder is Mike Pall.

I found a tool named `licensecheck` which automates the license
check ...

> Thanks!
>  Thorsten> 

-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers

Re: lua-torch-xlua_0~20160617-g0dd5f4c-1_amd64.changes REJECTED

[Lumin]

Hi Thorsten Alteholz,

Thank you for pointing it out.
The two files mentioned are released under the MIT license instead of
BSD-3-Clause.

I'll fix this.

On 17 August 2016 at 20:00, Thorsten Alteholz
<ftpmas...@ftp-master.debian.org> wrote:
>
> Hi,
>
> unfortunately I have to reject your package.
> The license of OptionParser.lua and init.lua doesn't seem to be BSD3,
> so please mention them in your debian/coypright.
>
> Thanks!
>  Thorsten
>
> ===
>
> Please feel free to respond to this email if you don't understand why
> your files were rejected, or if you upload new files which address our
> concerns.
>



-- 
Best,
Lumin

-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers

Bug#799262: request +1 for python3-opencv

[lumin]

request +1 for python3-opencv

-- 
debian-science-maintainers mailing list
debian-science-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-science-maintainers