def test_unique_nan(df_factory):
x = [np.nan, 0, 1, np.nan, 2, np.nan]
df = df_factory(x=x)
assert set(df.x.unique(dropnan=True)) == {0, 1, 2}
assert dropnan(set(df.x.unique()), expect=1) == {0, 1, 2}
with small_buffer(df, 2):
values, indices = df.unique(df.x, return_inverse=True)
values = np.array(values)
values = values[indices]
mask = np.isnan(values)
assert values[~mask].tolist() == df.x.to_numpy()[~mask].tolist()
def test_set_float(repickle, nan, missing, nmaps):
ar = np.arange(4, dtype='f8')[::-1].copy()
keys_expected = [3, 2, 1, 0]
null_index = 2
if missing:
mask = [0, 0, 1, 0]
keys_expected[null_index] = None
if nan:
ar[1] = np.nan
keys_expected[1] = np.nan
oset = ordered_set_float64(nmaps)
if missing:
ordinals_local, map_index = oset.update(ar, mask, return_values=True)
else:
ordinals_local, map_index = oset.update(ar, return_values=True)
ordinals = np.empty(len(keys_expected), dtype='i8')
ordinals = oset.flatten_values(ordinals_local, map_index, ordinals)
keys = oset.keys()
# if missing:
# ordinals[oset.null_value] = oset.null_value
assert dropnan(np.take(keys, ordinals).tolist()) == dropnan(keys_expected)
# plain object keys
oset.seal()
keys = oset.keys()
expect_nan = 1 if nan else None
assert dropnan(set(keys), expect=expect_nan) == dropnan(set(keys_expected), expect=expect_nan)
assert oset.map_ordinal(keys).dtype.name == 'int8'
# arrays
keys = oset.key_array().tolist()
if missing:
keys[oset.null_value] = None
assert dropnan(set(keys), expect=expect_nan) == dropnan(set(keys_expected), expect=expect_nan)
if nan:
assert np.isnan(keys[oset.nan_value])
ordinals = oset.map_ordinal(keys).tolist()
if missing:
ordinals[oset.null_value] = oset.null_value
assert ordinals == list(range(4))
# tests extraction and constructor
keys = oset.key_array()
set_copy = ordered_set_float64(keys, oset.null_value, oset.nan_count, oset.null_count, '')
keys = set_copy.key_array().tolist()
if missing:
keys[oset.null_value] = None
assert dropnan(set(keys)) == dropnan(set(keys_expected))
if nan:
assert np.isnan(keys[set_copy.nan_value])
ordinals = set_copy.map_ordinal(keys).tolist()
if missing:
ordinals[set_copy.null_value] = set_copy.null_value
assert ordinals == list(range(4))
# test pickle
set_copy = repickle(oset)
keys = set_copy.key_array().tolist()
if missing:
keys[oset.null_value] = None
assert dropnan(set(keys)) == dropnan(set(keys_expected))
if nan:
assert np.isnan(keys[set_copy.nan_value])
ordinals = set_copy.map_ordinal(keys).tolist()
if missing:
ordinals[set_copy.null_value] = set_copy.null_value
assert ordinals == list(range(4))
def test_unique_f4(df_factory):
x = np.array([np.nan, 0, 1, np.nan, 2, np.nan], dtype='f4')
df = df_factory(x=x)
assert dropnan(set(df.x.unique(dropnan=True))) == {0, 1, 2}
assert dropnan(set(df.x.unique()), expect=1) == {0, 1, 2}