TST: skip scipy tests for >= 0.19.0 as needed in interpolation / window / sparse

closes #15668

Author: jreback

pandas/tests/frame/test_missing.py

@@ -548,7 +548,8 @@ def test_interp_nan_idx(self):
             df.interpolate(method='values')
 
     def test_interp_various(self):
-        tm._skip_if_no_scipy()
+        tm.skip_if_no_package('scipy', max_version='0.19.0')
+
         df = DataFrame({'A': [1, 2, np.nan, 4, 5, np.nan, 7],
                         'C': [1, 2, 3, 5, 8, 13, 21]})
         df = df.set_index('C')

pandas/tests/frame/test_rank.py

@@ -193,7 +193,8 @@ def test_rank_axis(self):
         tm.assert_frame_equal(df.rank(axis=1), df.rank(axis='columns'))
 
     def test_rank_methods_frame(self):
-        tm.skip_if_no_package('scipy', '0.13', 'scipy.stats.rankdata')
+        tm.skip_if_no_package('scipy', min_version='0.13',
+                              app='scipy.stats.rankdata')
         import scipy
         from scipy.stats import rankdata
 

pandas/tests/series/test_missing.py

@@ -827,7 +827,8 @@ def test_interp_quad(self):
         assert_series_equal(result, expected)
 
     def test_interp_scipy_basic(self):
-        tm._skip_if_no_scipy()
+        tm.skip_if_no_package('scipy', max_version='0.19.0')
+
         s = Series([1, 3, np.nan, 12, np.nan, 25])
         # slinear
         expected = Series([1., 3., 7.5, 12., 18.5, 25.])
@@ -1027,8 +1028,8 @@ def test_spline(self):
 
     def test_spline_extrapolate(self):
         tm.skip_if_no_package(
-            'scipy', '0.15',
-            'setting ext on scipy.interpolate.UnivariateSpline')
+            'scipy', min_version='0.15',
+            app='setting ext on scipy.interpolate.UnivariateSpline')
         s = Series([1, 2, 3, 4, np.nan, 6, np.nan])
         result3 = s.interpolate(method='spline', order=1, ext=3)
         expected3 = Series([1., 2., 3., 4., 5., 6., 6.])

pandas/tests/series/test_rank.py

@@ -246,7 +246,8 @@ def _check(s, expected, method='average'):
             _check(series, results[method], method=method)
 
     def test_rank_methods_series(self):
-        tm.skip_if_no_package('scipy', '0.13', 'scipy.stats.rankdata')
+        tm.skip_if_no_package('scipy', min_version='0.13',
+                              app='scipy.stats.rankdata')
         import scipy
         from scipy.stats import rankdata
 

pandas/tests/sparse/test_frame.py

@@ -1129,10 +1129,10 @@ def test_isnotnull(self):
 @pytest.mark.parametrize('index', [None, list('ab')])    # noqa: F811
 @pytest.mark.parametrize('columns', [None, list('cd')])
 @pytest.mark.parametrize('fill_value', [None, 0, np.nan])
-@pytest.mark.parametrize('dtype', [object, bool, int, float, np.uint16])
+@pytest.mark.parametrize('dtype', [bool, int, float, np.uint16])
 def test_from_to_scipy(spmatrix, index, columns, fill_value, dtype):
     # GH 4343
-    tm._skip_if_no_scipy()
+    tm.skip_if_no_package('scipy')
 
     # Make one ndarray and from it one sparse matrix, both to be used for
     # constructing frames and comparing results
@@ -1180,6 +1180,51 @@ def test_from_to_scipy(spmatrix, index, columns, fill_value, dtype):
     tm.assert_equal(sdf.to_coo().dtype, np.object_)
 
 
+@pytest.mark.parametrize('fill_value', [None, 0, np.nan])    # noqa: F811
+def test_from_to_scipy_object(spmatrix, fill_value):
+    # GH 4343
+    dtype = object
+    columns = list('cd')
+    index = list('ab')
+    tm.skip_if_no_package('scipy', max_version='0.19.0')
+
+    # Make one ndarray and from it one sparse matrix, both to be used for
+    # constructing frames and comparing results
+    arr = np.eye(2, dtype=dtype)
+    try:
+        spm = spmatrix(arr)
+        assert spm.dtype == arr.dtype
+    except (TypeError, AssertionError):
+        # If conversion to sparse fails for this spmatrix type and arr.dtype,
+        # then the combination is not currently supported in NumPy, so we
+        # can just skip testing it thoroughly
+        return
+
+    sdf = pd.SparseDataFrame(spm, index=index, columns=columns,
+                             default_fill_value=fill_value)
+
+    # Expected result construction is kind of tricky for all
+    # dtype-fill_value combinations; easiest to cast to something generic
+    # and except later on
+    rarr = arr.astype(object)
+    rarr[arr == 0] = np.nan
+    expected = pd.SparseDataFrame(rarr, index=index, columns=columns).fillna(
+        fill_value if fill_value is not None else np.nan)
+
+    # Assert frame is as expected
+    sdf_obj = sdf.astype(object)
+    tm.assert_sp_frame_equal(sdf_obj, expected)
+    tm.assert_frame_equal(sdf_obj.to_dense(), expected.to_dense())
+
+    # Assert spmatrices equal
+    tm.assert_equal(dict(sdf.to_coo().todok()), dict(spm.todok()))
+
+    # Ensure dtype is preserved if possible
+    res_dtype = object
+    tm.assert_contains_all(sdf.dtypes, {np.dtype(res_dtype)})
+    tm.assert_equal(sdf.to_coo().dtype, res_dtype)
+
+
 class TestSparseDataFrameArithmetic(tm.TestCase):
 
     def test_numeric_op_scalar(self):

pandas/tests/test_nanops.py

@@ -5,7 +5,7 @@
 
 import warnings
 import numpy as np
-from pandas import Series, isnull
+from pandas import Series, isnull, _np_version_under1p9
 from pandas.types.common import is_integer_dtype
 import pandas.core.nanops as nanops
 import pandas.util.testing as tm
@@ -338,8 +338,7 @@ def test_nanmean_overflow(self):
         # is now consistent with numpy
 
         # numpy < 1.9.0 is not computing this correctly
-        from distutils.version import LooseVersion
-        if LooseVersion(np.__version__) >= '1.9.0':
+        if not _np_version_under1p9:
             for a in [2 ** 55, -2 ** 55, 20150515061816532]:
                 s = Series(a, index=range(500), dtype=np.int64)
                 result = s.mean()
@@ -388,8 +387,7 @@ def test_nanstd(self):
                              allow_tdelta=True, allow_obj='convert')
 
     def test_nansem(self):
-        tm.skip_if_no_package('scipy.stats')
-        tm._skip_if_scipy_0_17()
+        tm.skip_if_no_package('scipy', min_version='0.17.0')
         from scipy.stats import sem
         self.check_funs_ddof(nanops.nansem, sem, allow_complex=False,
                              allow_str=False, allow_date=False,
@@ -448,16 +446,14 @@ def _skew_kurt_wrap(self, values, axis=None, func=None):
         return result
 
     def test_nanskew(self):
-        tm.skip_if_no_package('scipy.stats')
-        tm._skip_if_scipy_0_17()
+        tm.skip_if_no_package('scipy', min_version='0.17.0')
         from scipy.stats import skew
         func = partial(self._skew_kurt_wrap, func=skew)
         self.check_funs(nanops.nanskew, func, allow_complex=False,
                         allow_str=False, allow_date=False, allow_tdelta=False)
 
     def test_nankurt(self):
-        tm.skip_if_no_package('scipy.stats')
-        tm._skip_if_scipy_0_17()
+        tm.skip_if_no_package('scipy', min_version='0.17.0')
         from scipy.stats import kurtosis
         func1 = partial(kurtosis, fisher=True)
         func = partial(self._skew_kurt_wrap, func=func1)

pandas/tests/test_window.py

@@ -905,7 +905,7 @@ def test_cmov_window_na_min_periods(self):
 
     def test_cmov_window_regular(self):
         # GH 8238
-        tm._skip_if_no_scipy()
+        tm.skip_if_no_package('scipy', max_version='0.19.0')
 
         win_types = ['triang', 'blackman', 'hamming', 'bartlett', 'bohman',
                      'blackmanharris', 'nuttall', 'barthann']
@@ -938,7 +938,7 @@ def test_cmov_window_regular(self):
 
     def test_cmov_window_regular_linear_range(self):
         # GH 8238
-        tm._skip_if_no_scipy()
+        tm.skip_if_no_package('scipy', max_version='0.19.0')
 
         win_types = ['triang', 'blackman', 'hamming', 'bartlett', 'bohman',
                      'blackmanharris', 'nuttall', 'barthann']
@@ -955,7 +955,7 @@ def test_cmov_window_regular_linear_range(self):
 
     def test_cmov_window_regular_missing_data(self):
         # GH 8238
-        tm._skip_if_no_scipy()
+        tm.skip_if_no_package('scipy', max_version='0.19.0')
 
         win_types = ['triang', 'blackman', 'hamming', 'bartlett', 'bohman',
                      'blackmanharris', 'nuttall', 'barthann']
@@ -988,7 +988,7 @@ def test_cmov_window_regular_missing_data(self):
 
     def test_cmov_window_special(self):
         # GH 8238
-        tm._skip_if_no_scipy()
+        tm.skip_if_no_package('scipy', max_version='0.19.0')
 
         win_types = ['kaiser', 'gaussian', 'general_gaussian', 'slepian']
         kwds = [{'beta': 1.}, {'std': 1.}, {'power': 2.,
@@ -1015,7 +1015,7 @@ def test_cmov_window_special(self):
 
     def test_cmov_window_special_linear_range(self):
         # GH 8238
-        tm._skip_if_no_scipy()
+        tm.skip_if_no_package('scipy', max_version='0.19.0')
 
         win_types = ['kaiser', 'gaussian', 'general_gaussian', 'slepian']
         kwds = [{'beta': 1.}, {'std': 1.}, {'power': 2.,

pandas/util/testing.py

@@ -304,14 +304,6 @@ def _skip_if_no_scipy():
         pytest.skip('scipy.sparse missing')
 
 
-def _skip_if_scipy_0_17():
-    import scipy
-    v = scipy.__version__
-    if v >= LooseVersion("0.17.0"):
-        import pytest
-        pytest.skip("scipy 0.17")
-
-
 def _check_if_lzma():
     try:
         return compat.import_lzma()
@@ -2020,15 +2012,18 @@ def __init__(self, *args, **kwargs):
 
 # Dependency checks.  Copied this from Nipy/Nipype (Copyright of
 # respective developers, license: BSD-3)
-def package_check(pkg_name, version=None, app='pandas', checker=LooseVersion):
-    """Check that the minimal version of the required package is installed.
+def package_check(pkg_name, min_version=None, max_version=None, app='pandas',
+                  checker=LooseVersion):
+    """Check that the min/max version of the required package is installed.
 
     Parameters
     ----------
     pkg_name : string
         Name of the required package.
-    version : string, optional
+    min_version : string, optional
         Minimal version number for required package.
+    max_version : string, optional
+        Max version number for required package.
     app : string, optional
         Application that is performing the check.  For instance, the
         name of the tutorial being executed that depends on specific
@@ -2040,7 +2035,6 @@ def package_check(pkg_name, version=None, app='pandas', checker=LooseVersion):
     Examples
     --------
     package_check('numpy', '1.3')
-    package_check('networkx', '1.0', 'tutorial1')
 
     """
 
@@ -2049,8 +2043,10 @@ def package_check(pkg_name, version=None, app='pandas', checker=LooseVersion):
         msg = '%s requires %s' % (app, pkg_name)
     else:
         msg = 'module requires %s' % pkg_name
-    if version:
-        msg += ' with version >= %s' % (version,)
+    if min_version:
+        msg += ' with version >= %s' % (min_version,)
+    if max_version:
+        msg += ' with version < %s' % (max_version,)
     try:
         mod = __import__(pkg_name)
     except ImportError:
@@ -2059,7 +2055,9 @@ def package_check(pkg_name, version=None, app='pandas', checker=LooseVersion):
         have_version = mod.__version__
     except AttributeError:
         pytest.skip('Cannot find version for %s' % pkg_name)
-    if version and checker(have_version) < checker(version):
+    if min_version and checker(have_version) < checker(min_version):
+        pytest.skip(msg)
+    if max_version and checker(have_version) >= checker(max_version):
         pytest.skip(msg)