pandas.git
5 years agoDefault to openpyxl not xlrd for read_excel
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Default to openpyxl not xlrd for read_excel

xlrd 1.2 fails if defusedxml (needed for odf) is installed

Bug: https://github.com/pandas-dev/pandas/pull/35029
Bug-Debian: https://bugs.debian.org/976620
Origin: upstream b3a3932af6aafaa2fd41f17e9b7995643e5f92eb
Author: Robert de Vries, Rebecca N. Palmer <rebecca_palmer@zoho.com>
Forwarded: not-needed

Gbp-Pq: Name xlrd_976620.patch

5 years agoFix pytables expressions with Python 3.9
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Fix pytables expressions with Python 3.9

ast.Index has been replaced by a bare value, so we need to do the
conversion from Constant to int

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Bug-Debian: https://bugs.debian.org/972015
Forwarded: https://github.com/pandas-dev/pandas/issues/37217

Gbp-Pq: Name python39_compat.patch

5 years agoSkip test that fails from outside the source tree
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Skip test that fails from outside the source tree

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Forwarded: not-needed

Gbp-Pq: Name skip_test_missing_required_dependency.patch

5 years agoClean up so test_ts_plot_with_tz doesn't fail
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Clean up so test_ts_plot_with_tz doesn't fail

Author: Steffen Rehberg, Rebecca N. Palmer <rebecca_palmer@zoho.com>
Bug: https://github.com/pandas-dev/pandas/issues/35080
Forwarded: no

Gbp-Pq: Name 35080_plot_tests.patch

5 years agoSkip rather than fail tests if fsspec is too old
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Skip rather than fail tests if fsspec is too old

(which it currently is in Debian)

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Forwarded: no

Gbp-Pq: Name fsspec_version.patch

5 years agoDon't fail test for rounding difference on i386
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Don't fail test for rounding difference on i386

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Bug-Debian: https://bugs.debian.org/968208
Forwarded: no

Gbp-Pq: Name 968208_rounding.patch

5 years agoMatplotlib 3.3 compatibility fixup
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Matplotlib 3.3 compatibility fixup

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Bug-Debian: https://bugs.debian.org/966393
Forwarded: no

Gbp-Pq: Name matplotlib33_compat.patch

5 years agoAvoid test failures on Hurd
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Avoid test failures on Hurd

Allow multiprocessing to be unavailable
Accept any errno not just 2 for (intentionally) nonexistent files
(Hurd appears to use 2**30+2)

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Forwarded: no

Gbp-Pq: Name hurd_compat.patch

5 years agoAllow some numba errors on non-amd64, warn on non-x86
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Allow some numba errors on non-amd64, warn on non-x86

Specifying the exception type allows only explicit errors,
not silently wrong answers

Numba has been observed to give wrong answers on mipsel,
and crash on armel (LLVM ERROR) and s390x (segfault).

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Forwarded: no

Gbp-Pq: Name numba_fail_32bit.patch

5 years agoRemove code from Stack Overflow
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Remove code from Stack Overflow

Stack Overflow content is CC-BY-SA licensed,
which this package is not supposed to be.  These snippets may be
too small to be copyrightable, but removing them to be safe.

https://lists.debian.org/debian-legal/2020/04/threads.html#00018

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Forwarded: no - deletes some tests/examples without replacement

Gbp-Pq: Name remove_ccbysa_snippets.patch

5 years agoUse test URLs that are less likely to disappear
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Use test URLs that are less likely to disappear

Avoid 404 errors in stable when upstream reorganize the test data
(happened to two of these in 0.25 -> 1.0).

It is _not_ necessary to update the tag version on every package release,
only if these tests fail because they expect moved/changed data.

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Forwarded: not-needed

Gbp-Pq: Name stable_test_urls.patch

5 years agoAllow tests to use the data files in the source tree
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Allow tests to use the data files in the source tree

We don't ship these in the package,
but do want to run the tests that use them

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Forwarded: not-needed

Gbp-Pq: Name find_test_data.patch

5 years agoLink to rather than embed Google calendar
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Link to rather than embed Google calendar

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Forwarded: not-needed

Gbp-Pq: Name privacy.patch

5 years agoDon't require the pandas theme
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Don't require the pandas theme

(TODO - package it - here as a multi-upstream-tarball package?)

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Forwarded: not-needed

Gbp-Pq: Name sphinx_no_pandas_theme.patch

5 years agoXfail a test that doesn't work in the C locale
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Xfail a test that doesn't work in the C locale

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Forwarded: no

Gbp-Pq: Name xfail_c_locale.patch

5 years agoUse fixed seeds for reproducible pseudorandomness
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Use fixed seeds for reproducible pseudorandomness

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Forwarded: no

Gbp-Pq: Name fix_random_seeds.patch

5 years agoFix or skip tests that assume the wrong endianness
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Fix or skip tests that assume the wrong endianness

This is a bug in the tests not pandas itself -
the expected values explicitly specify little-endian

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Forwarded: no

Gbp-Pq: Name tests_dont_assume_endian.patch

5 years agoDon't try to read a contributor list from the git log
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Don't try to read a contributor list from the git log

Debian packages are built from tarballs, so there isn't a git log.

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Forwarded: not-needed

Gbp-Pq: Name contributor_list_not_in_tarball.patch

5 years agoUse packaged intersphinx indexes
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Use packaged intersphinx indexes

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Bug-Debian: https://bugs.debian.org/876417
Forwarded: not-needed

Gbp-Pq: Name use_system_intersphinx.patch

5 years agoDon't test datetime in locales with no encoding
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Don't test datetime in locales with no encoding

Some datetime tests run the test in every available locale.
If this set includes locales without an encoding (currently dsb_DE
and sah_RU), it fails due to Python bug
https://bugs.python.org/issue20088

Failure log
https://tests.reproducible-builds.org/debian/rbuild/buster/amd64/pandas_0.23.3+dfsg-3.rbuild.log.gz

Author: Rebecca N. Palmer <rebecca_palmer@zoho.com>
Bug: https://github.com/pandas-dev/pandas/issues/20957
Forwarded: no

Gbp-Pq: Name skip_noencoding_locales.patch

5 years agoXfail NaN <-> NaT tests on non-x86 and warn on cast
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Xfail NaN <-> NaT tests on non-x86 and warn on cast

pd.Series([np.nan]).astype('datetime64[ns]')[0] = pd.NaT on x86
but 1970-01-01 on arm* because float NaN -> int is undefined:
https://github.com/numpy/numpy/issues/8325
https://github.com/pandas-dev/pandas/issues/17792
https://github.com/pandas-dev/pandas/issues/26964

On s390x it's the maximum _positive_ value (2**63-1 ns = year 2262)

Author: Andreas Tille <tille@debian.org>, Graham Inggs <ginggs@debian.org>, Rebecca N. Palmer <rebecca_palmer@zoho.com>
Bug-Debian: https://bugs.debian.org/877754
Forwarded: no

Gbp-Pq: Name xfail_tests_nonintel_nannat.patch

5 years agoUse Debian packaged mathjax
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Use Debian packaged mathjax

Author: Andreas Tille <tille@debian.org>
Forwarded: not-needed

Gbp-Pq: Name mathjax-path.patch

5 years agoDon't fail tests because we don't ship test data
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Don't fail tests because we don't ship test data

This applies to users running them; our build/autopkgtest re-enable
this and use the data from the source tree

Author: Yaroslav Halchenko
Forwarded: no

Gbp-Pq: Name deb_no_strict_data.patch

5 years agoAvoid privacy breach by Google Analytics
Yaroslav Halchenko [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Avoid privacy breach by Google Analytics

Last-Update: Mon, 23 Oct 2017 08:55:28 +0200
Forwarded: not-needed

Gbp-Pq: Name deb_disable_googleanalytics.patch

5 years agoHDF5 and Stata I/O are broken on some architectures
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
HDF5 and Stata I/O are broken on some architectures

Fix some issues, warn on use and xfail tests for the remainder

armhf TestHDF5Store::test*encoding only sometimes crashes
(1.1.3+dfsg-1 passed on build but failed autopkgtest)

HDF5 and Stata are known to fail on big-endian architectures
Stata also fails on qemu-ppc64el, but not real ppc64el

Author: Andreas Tille <tille@debian.org>, Graham Inggs <ginggs@debian.org>, Yaroslav Halchenko <debian@onerussian.com>, Rebecca N. Palmer <rebecca_palmer@zoho.com>
Bug-Debian: https://bugs.debian.org/877419
Forwarded: no

Gbp-Pq: Name xfail_tests_nonintel_io.patch

5 years agoDon't try to import from the source directory
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Don't try to import from the source directory

Needed as we build the extension modules elsewhere

Author: Yaroslav Halchenko
Forwarded: not-needed

Gbp-Pq: Name deb_doc_donotoverride_PYTHONPATH.patch

5 years agoDon't try to use git to find the version number
Debian Science Team [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Don't try to use git to find the version number

Needed as Debian buildds use tarballs

Author: Yaroslav Halchenko
Forwarded: not-needed

Gbp-Pq: Name deb_nonversioneer_version.patch

5 years agopandas (1.1.5+dfsg-1) unstable; urgency=medium
Rebecca N. Palmer [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
pandas (1.1.5+dfsg-1) unstable; urgency=medium

  * Upstream bugfix release.  Refresh patches, contributors_list.
  * Default to openpyxl not xlrd in read_excel, and skip xlrd tests,
    as xlrd fails if defusedxml is installed (#976620).
  * Skip numba tests, and not other rolling.apply tests, on s390x.
    (LP: #1901860)
  * Tests: on 32 bit systems, avoid time input that overflows.
  * Print uname etc during build (test for #973854).
  * Bump Standards-Version to 4.5.1 (no changes needed).

[dgit import unpatched pandas 1.1.5+dfsg-1]

5 years agoImport pandas_1.1.5+dfsg.orig.tar.xz
Rebecca N. Palmer [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Import pandas_1.1.5+dfsg.orig.tar.xz

[dgit import orig pandas_1.1.5+dfsg.orig.tar.xz]

5 years agoImport pandas_1.1.5+dfsg-1.debian.tar.xz
Rebecca N. Palmer [Mon, 7 Dec 2020 23:06:28 +0000 (23:06 +0000)]
Import pandas_1.1.5+dfsg-1.debian.tar.xz

[dgit import tarball pandas 1.1.5+dfsg-1 pandas_1.1.5+dfsg-1.debian.tar.xz]