My personal project and infrastructure archive
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
nomicon/pkgs/development/python-modules/pandas/default.nix

138 lines
3.8 KiB

{ lib, stdenv
, buildPythonPackage
, fetchPypi
, python
, isPy38
, beautifulsoup4
, bottleneck
, cython
, dateutil
, html5lib
, lxml
, numexpr
, openpyxl
, pytz
, sqlalchemy
, scipy
, tables
, xlrd
, xlwt
# Test Inputs
, glibcLocales
, hypothesis
, pytestCheckHook
# Darwin inputs
, runtimeShell
, libcxx ? null
}:
buildPythonPackage rec {
pname = "pandas";
version = "1.2.2";
src = fetchPypi {
inherit pname version;
sha256 = "14ed84b463e9b84c8ff9308a79b04bf591ae3122a376ee0f62c68a1bd917a773";
};
# See https://github.com/scipy/scipy/issues/13585 and https://github.com/pandas-dev/pandas/pull/40020
patches = [ ./fix-tests.patch ];
nativeBuildInputs = [ cython ];
buildInputs = lib.optional stdenv.isDarwin libcxx;
propagatedBuildInputs = [
beautifulsoup4
bottleneck
dateutil
html5lib
numexpr
lxml
openpyxl
pytz
scipy
sqlalchemy
tables
xlrd
xlwt
];
checkInputs = [ pytestCheckHook glibcLocales hypothesis ];
# doesn't work with -Werror,-Wunused-command-line-argument
# https://github.com/NixOS/nixpkgs/issues/39687
hardeningDisable = lib.optional stdenv.cc.isClang "strictoverflow";
# Parallel Cythonization is broken in Python 3.8 on Darwin. Fixed in the next
# release. https://github.com/pandas-dev/pandas/pull/30862
setupPyBuildFlags = lib.optionals (!(isPy38 && stdenv.isDarwin)) [
# As suggested by
# https://pandas.pydata.org/pandas-docs/stable/development/contributing.html#creating-a-python-environment
"--parallel=$NIX_BUILD_CORES"
];
doCheck = !stdenv.isAarch64; # upstream doesn't test this architecture
pytestFlagsArray = [
"--skip-slow"
"--skip-network"
];
disabledTests = [
# since dateutil 0.6.0 the following fails: test_fallback_plural, test_ambiguous_flags, test_ambiguous_compat
# was supposed to be solved by https://github.com/dateutil/dateutil/issues/321, but is not the case
"test_fallback_plural"
"test_ambiguous_flags"
"test_ambiguous_compat"
# Locale-related
"test_names"
"test_dt_accessor_datetime_name_accessors"
"test_datetime_name_accessors"
# Can't import from test folder
"test_oo_optimizable"
# Disable IO related tests because IO data is no longer distributed
"io"
# KeyError Timestamp
"test_to_excel"
# ordering logic has changed
"numpy_ufuncs_other"
"order_without_freq"
# tries to import from pandas.tests post install
"util_in_top_level"
# Fails with 1.0.5
"test_constructor_list_frames"
"test_constructor_with_embedded_frames"
# tries to import compiled C extension locally
"test_missing_required_dependency"
] ++ lib.optionals stdenv.isDarwin [
"test_locale"
"test_clipboard"
];
# tests have relative paths, and need to reference compiled C extensions
# so change directory where `import .test` is able to be resolved
preCheck = ''
cd $out/${python.sitePackages}/pandas
export LC_ALL="en_US.UTF-8"
PYTHONPATH=$out/${python.sitePackages}:$PYTHONPATH
''
# TODO: Get locale and clipboard support working on darwin.
# Until then we disable the tests.
+ lib.optionalString stdenv.isDarwin ''
# Fake the impure dependencies pbpaste and pbcopy
echo "#!${runtimeShell}" > pbcopy
echo "#!${runtimeShell}" > pbpaste
chmod a+x pbcopy pbpaste
export PATH=$(pwd):$PATH
'';
meta = with lib; {
# https://github.com/pandas-dev/pandas/issues/14866
# pandas devs are no longer testing i686 so safer to assume it's broken
broken = stdenv.isi686;
homepage = "https://pandas.pydata.org/";
changelog = "https://pandas.pydata.org/docs/whatsnew/index.html";
description = "Python Data Analysis Library";
license = licenses.bsd3;
maintainers = with maintainers; [ raskin fridh knedlsepp ];
platforms = platforms.unix;
};
}