def test_to_series_comparison(self):
kidx1 = ks.Index([1, 2, 3, 4, 5])
kidx2 = ks.Index([1, 2, 3, 4, 5])
self.assert_eq((kidx1.to_series() == kidx2.to_series()).all(), True)
kidx1.name = "koalas"
kidx2.name = "koalas"
self.assert_eq((kidx1.to_series() == kidx2.to_series()).all(), True)
def test_sort_values(self):
pidx = pd.Index([-10, -100, 200, 100])
kidx = ks.Index([-10, -100, 200, 100])
self.assert_eq(pidx.sort_values(), kidx.sort_values())
self.assert_eq(pidx.sort_values(ascending=False),
kidx.sort_values(ascending=False))
pidx.name = "koalas"
kidx.name = "koalas"
self.assert_eq(pidx.sort_values(), kidx.sort_values())
self.assert_eq(pidx.sort_values(ascending=False),
kidx.sort_values(ascending=False))
pidx = pd.MultiIndex.from_tuples([("a", "x", 1), ("b", "y", 2),
("c", "z", 3)])
kidx = ks.MultiIndex.from_tuples([("a", "x", 1), ("b", "y", 2),
("c", "z", 3)])
pidx.names = ["hello", "koalas", "goodbye"]
kidx.names = ["hello", "koalas", "goodbye"]
self.assert_eq(pidx.sort_values(), kidx.sort_values())
self.assert_eq(pidx.sort_values(ascending=False),
kidx.sort_values(ascending=False))
def test_sort_values(self):
pidx = pd.Index([-10, -100, 200, 100])
kidx = ks.Index([-10, -100, 200, 100])
self.assert_eq(pidx.sort_values(), kidx.sort_values())
self.assert_eq(pidx.sort_values(ascending=False),
kidx.sort_values(ascending=False))
pidx.name = 'koalas'
kidx.name = 'koalas'
self.assert_eq(pidx.sort_values(), kidx.sort_values())
self.assert_eq(pidx.sort_values(ascending=False),
kidx.sort_values(ascending=False))
pidx = pd.MultiIndex.from_tuples([('a', 'x', 1), ('b', 'y', 2),
('c', 'z', 3)])
kidx = ks.MultiIndex.from_tuples([('a', 'x', 1), ('b', 'y', 2),
('c', 'z', 3)])
pidx.names = ['hello', 'koalas', 'goodbye']
kidx.names = ['hello', 'koalas', 'goodbye']
self.assert_eq(pidx.sort_values(), kidx.sort_values())
self.assert_eq(pidx.sort_values(ascending=False),
kidx.sort_values(ascending=False))
def test_series_iloc_setitem(self):
pser = pd.Series([1, 2, 3], index=["cobra", "viper", "sidewinder"])
kser = ks.from_pandas(pser)
piloc = pser.iloc
kiloc = kser.iloc
pser1 = pser + 1
kser1 = kser + 1
for key, value in [
([1, 2], 10),
(1, 50),
(slice(None), 10),
(slice(None, 1), 20),
(slice(1, None), 30),
]:
with self.subTest(key=key, value=value):
pser.iloc[key] = value
kser.iloc[key] = value
self.assert_eq(kser, pser)
piloc[key] = -value
kiloc[key] = -value
self.assert_eq(kser, pser)
pser1.iloc[key] = value
kser1.iloc[key] = value
self.assert_eq(kser1, pser1)
with self.assertRaises(ValueError):
kser.iloc[1] = -kser
pser = pd.Index([1, 2, 3]).to_series()
kser = ks.Index([1, 2, 3]).to_series()
pser1 = pser + 1
kser1 = kser + 1
pser.iloc[0] = 10
kser.iloc[0] = 10
self.assert_eq(kser, pser)
pser1.iloc[0] = 20
kser1.iloc[0] = 20
self.assert_eq(kser1, pser1)
pdf = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
kdf = ks.from_pandas(pdf)
pser = pdf.a
kser = kdf.a
pser.iloc[[1]] = -pdf.b
kser.iloc[[1]] = -kdf.b
self.assert_eq(kser, pser)
with self.assertRaisesRegex(ValueError,
"Incompatible indexer with DataFrame"):
kser.iloc[1] = kdf[["b"]]
def test_index_drop_duplicates(self):
pidx = pd.Index([1, 1, 2])
kidx = ks.Index([1, 1, 2])
self.assert_eq(pidx.drop_duplicates(), kidx.drop_duplicates())
pidx = pd.MultiIndex.from_tuples([(1, 1), (1, 1), (2, 2)], names=['level1', 'level2'])
kidx = ks.MultiIndex.from_tuples([(1, 1), (1, 1), (2, 2)], names=['level1', 'level2'])
self.assert_eq(pidx.drop_duplicates(), kidx.drop_duplicates())
def test_len(self):
pidx = pd.Index(range(10000))
kidx = ks.Index(range(10000))
self.assert_eq(len(pidx), len(kidx))
pidx = pd.MultiIndex.from_tuples([('a', 'x', 1), ('b', 'y', 2), ('c', 'z', 3)])
kidx = ks.MultiIndex.from_tuples([('a', 'x', 1), ('b', 'y', 2), ('c', 'z', 3)])
self.assert_eq(len(pidx), len(kidx))
def test_len(self):
pidx = pd.Index(range(10000))
kidx = ks.Index(range(10000))
self.assert_eq(len(pidx), len(kidx))
pidx = pd.MultiIndex.from_tuples([("a", "x", 1), ("b", "y", 2), ("c", "z", 3)])
kidx = ks.MultiIndex.from_tuples([("a", "x", 1), ("b", "y", 2), ("c", "z", 3)])
self.assert_eq(len(pidx), len(kidx))
def test_multi_index_symmetric_difference(self):
idx = ks.Index(['a', 'b', 'c'])
midx = ks.MultiIndex.from_tuples([('a', 'x'), ('b', 'y'), ('c', 'z')])
midx_ = ks.MultiIndex.from_tuples([('a', 'x'), ('b', 'y'), ('c', 'z')])
self.assert_eq(
midx.symmetric_difference(midx_),
midx.to_pandas().symmetric_difference(midx_.to_pandas()))
with self.assertRaisesRegexp(NotImplementedError, "Doesn't support*"):
midx.symmetric_difference(idx)
def indexer_between_time(
self,
start_time: Union[datetime.time, str],
end_time: Union[datetime.time, str],
include_start: bool = True,
include_end: bool = True,
) -> Index:
"""
Return index locations of values between particular times of day
(e.g., 9:00-9:30AM).
Parameters
----------
start_time, end_time : datetime.time, str
Time passed either as object (datetime.time) or as string in
appropriate format ("%H:%M", "%H%M", "%I:%M%p", "%I%M%p",
"%H:%M:%S", "%H%M%S", "%I:%M:%S%p","%I%M%S%p").
include_start : bool, default True
include_end : bool, default True
Returns
-------
values_between_time : Index of integers
Examples
--------
>>> kidx = ks.date_range("2000-01-01", periods=3, freq="T")
>>> kidx # doctest: +NORMALIZE_WHITESPACE
DatetimeIndex(['2000-01-01 00:00:00', '2000-01-01 00:01:00',
'2000-01-01 00:02:00'],
dtype='datetime64[ns]', freq=None)
>>> kidx.indexer_between_time("00:01", "00:02").sort_values()
Int64Index([1, 2], dtype='int64')
>>> kidx.indexer_between_time("00:01", "00:02", include_end=False)
Int64Index([1], dtype='int64')
>>> kidx.indexer_between_time("00:01", "00:02", include_start=False)
Int64Index([2], dtype='int64')
"""
def pandas_between_time(pdf) -> ks.DataFrame[int]:
return pdf.between_time(start_time, end_time, include_start,
include_end)
kdf = self.to_frame()[[]]
id_column_name = verify_temp_column_name(kdf, "__id_column__")
kdf = kdf.koalas.attach_id_column("distributed-sequence",
id_column_name)
with ks.option_context("compute.default_index_type", "distributed"):
# The attached index in the statement below will be dropped soon,
# so we enforce “distributed” default index type
kdf = kdf.koalas.apply_batch(pandas_between_time)
return ks.Index(first_series(kdf).rename(self.name))
def test_argmax(self):
pidx = pd.Index([100, 50, 10, 20, 30, 60, 0, 50, 0, 100, 100, 100, 20, 0, 0])
kidx = ks.Index([100, 50, 10, 20, 30, 60, 0, 50, 0, 100, 100, 100, 20, 0, 0])
self.assert_eq(pidx.argmax(), kidx.argmax())
# MultiIndex
kidx = ks.MultiIndex.from_tuples([("a", "x", 1), ("b", "y", 2), ("c", "z", 3)])
with self.assertRaisesRegex(
TypeError, "reduction operation 'argmax' not allowed for this dtype"
):
kidx.argmax()
def test_index_sort(self):
idx = ks.Index([1, 2, 3, 4, 5])
midx = ks.MultiIndex.from_tuples([('a', 'x', 1), ('b', 'y', 2)])
with self.assertRaisesRegex(
TypeError,
"cannot sort an Index object in-place, use sort_values instead"):
idx.sort()
with self.assertRaisesRegex(
TypeError,
"cannot sort an Index object in-place, use sort_values instead"):
midx.sort()
def test_multi_index_symmetric_difference(self):
idx = ks.Index(["a", "b", "c"])
midx = ks.MultiIndex.from_tuples([("a", "x"), ("b", "y"), ("c", "z")])
midx_ = ks.MultiIndex.from_tuples([("a", "x"), ("b", "y"), ("c", "z")])
self.assert_eq(
midx.symmetric_difference(midx_),
midx.to_pandas().symmetric_difference(midx_.to_pandas()),
)
with self.assertRaisesRegexp(NotImplementedError, "Doesn't support*"):
midx.symmetric_difference(idx)
def test_argmin(self):
pidx = pd.Index(
[100, 50, 10, 20, 30, 60, 0, 50, 0, 100, 100, 100, 20, 0, 0])
kidx = ks.Index(
[100, 50, 10, 20, 30, 60, 0, 50, 0, 100, 100, 100, 20, 0, 0])
self.assert_eq(pidx.argmin(), kidx.argmin())
# MultiIndex
kidx = ks.MultiIndex.from_tuples([('a', 'x', 1), ('b', 'y', 2),
('c', 'z', 3)])
with self.assertRaisesRegex(
TypeError,
"reduction operation 'argmin' not allowed for this dtype"):
kidx.argmin()
def test_categorical_index(self):
pidx = pd.CategoricalIndex([1, 2, 3])
kidx = ks.CategoricalIndex([1, 2, 3])
self.assert_eq(kidx, pidx)
self.assert_eq(kidx.categories, pidx.categories)
self.assert_eq(kidx.codes, pd.Index(pidx.codes))
self.assert_eq(kidx.ordered, pidx.ordered)
pidx = pd.Index([1, 2, 3], dtype="category")
kidx = ks.Index([1, 2, 3], dtype="category")
self.assert_eq(kidx, pidx)
self.assert_eq(kidx.categories, pidx.categories)
self.assert_eq(kidx.codes, pd.Index(pidx.codes))
self.assert_eq(kidx.ordered, pidx.ordered)
pdf = pd.DataFrame(
{
"a":
pd.Categorical([1, 2, 3, 1, 2, 3]),
"b":
pd.Categorical(["a", "b", "c", "a", "b", "c"],
categories=["c", "b", "a"]),
},
index=pd.Categorical([10, 20, 30, 20, 30, 10],
categories=[30, 10, 20],
ordered=True),
)
kdf = ks.from_pandas(pdf)
pidx = pdf.set_index("b").index
kidx = kdf.set_index("b").index
self.assert_eq(kidx, pidx)
self.assert_eq(kidx.categories, pidx.categories)
self.assert_eq(kidx.codes, pd.Index(pidx.codes))
self.assert_eq(kidx.ordered, pidx.ordered)
pidx = pdf.set_index(["a", "b"]).index.get_level_values(0)
kidx = kdf.set_index(["a", "b"]).index.get_level_values(0)
self.assert_eq(kidx, pidx)
self.assert_eq(kidx.categories, pidx.categories)
self.assert_eq(kidx.codes, pd.Index(pidx.codes))
self.assert_eq(kidx.ordered, pidx.ordered)
def test_arithmetic_op_exceptions(self):
kser = self.ks_start_date
py_datetime = self.pd_start_date.dt.to_pydatetime()
datetime_index = ks.Index(self.pd_start_date)
for other in [1, 0.1, kser, datetime_index, py_datetime]:
expected_err_msg = "addition can not be applied to date times."
self.assertRaisesRegex(TypeError, expected_err_msg,
lambda: kser + other)
self.assertRaisesRegex(TypeError, expected_err_msg,
lambda: other + kser)
expected_err_msg = "multiplication can not be applied to date times."
self.assertRaisesRegex(TypeError, expected_err_msg,
lambda: kser * other)
self.assertRaisesRegex(TypeError, expected_err_msg,
lambda: other * kser)
expected_err_msg = "division can not be applied to date times."
self.assertRaisesRegex(TypeError, expected_err_msg,
lambda: kser / other)
self.assertRaisesRegex(TypeError, expected_err_msg,
lambda: other / kser)
self.assertRaisesRegex(TypeError, expected_err_msg,
lambda: kser // other)
self.assertRaisesRegex(TypeError, expected_err_msg,
lambda: other // kser)
expected_err_msg = "modulo can not be applied to date times."
self.assertRaisesRegex(TypeError, expected_err_msg,
lambda: kser % other)
self.assertRaisesRegex(TypeError, expected_err_msg,
lambda: other % kser)
expected_err_msg = "datetime subtraction can only be applied to datetime series."
for other in [1, 0.1]:
self.assertRaisesRegex(TypeError, expected_err_msg,
lambda: kser - other)
self.assertRaisesRegex(TypeError, expected_err_msg,
lambda: other - kser)
self.assertRaisesRegex(TypeError, expected_err_msg,
lambda: kser - other)
self.assertRaises(NotImplementedError, lambda: py_datetime - kser)
def indexer_at_time(self, time: Union[datetime.time, str], asof: bool = False) -> Index:
"""
Return index locations of values at particular time of day
(e.g. 9:30AM).
Parameters
----------
time : datetime.time or str
Time passed in either as object (datetime.time) or as string in
appropriate format ("%H:%M", "%H%M", "%I:%M%p", "%I%M%p",
"%H:%M:%S", "%H%M%S", "%I:%M:%S%p", "%I%M%S%p").
Returns
-------
values_at_time : Index of integers
Examples
--------
>>> kidx = ks.date_range("2000-01-01", periods=3, freq="T")
>>> kidx # doctest: +NORMALIZE_WHITESPACE
DatetimeIndex(['2000-01-01 00:00:00', '2000-01-01 00:01:00',
'2000-01-01 00:02:00'],
dtype='datetime64[ns]', freq=None)
>>> kidx.indexer_at_time("00:00")
Int64Index([0], dtype='int64')
>>> kidx.indexer_at_time("00:01")
Int64Index([1], dtype='int64')
"""
if asof:
raise NotImplementedError("'asof' argument is not supported")
def pandas_at_time(pdf) -> ks.DataFrame[int]:
return pdf.at_time(time, asof)
kdf = self.to_frame()[[]]
id_column_name = verify_temp_column_name(kdf, "__id_column__")
kdf = kdf.koalas.attach_id_column("distributed-sequence", id_column_name)
with ks.option_context("compute.default_index_type", "distributed"):
# The attached index in the statement below will be dropped soon,
# so we enforce “distributed” default index type
kdf = kdf.koalas.apply_batch(pandas_at_time)
return ks.Index(first_series(kdf).rename(self.name))
def test_series_iloc_setitem(self):
pser = pd.Series([1, 2, 3], index=["cobra", "viper", "sidewinder"])
kser = ks.from_pandas(pser)
pser1 = pser + 1
kser1 = kser + 1
for key, value in [
([1, 2], 10),
(1, 50),
(slice(None), 10),
(slice(None, 1), 20),
(slice(1, None), 30),
]:
with self.subTest(key=key, value=value):
pser.iloc[key] = value
kser.iloc[key] = value
self.assert_eq(kser, pser)
pser1.iloc[key] = value
kser1.iloc[key] = value
self.assert_eq(kser1, pser1)
with self.assertRaises(ValueError):
kser.iloc[1] = -kser
pser = pd.Index([1, 2, 3]).to_series()
kser = ks.Index([1, 2, 3]).to_series()
pser1 = pser + 1
kser1 = kser + 1
pser.iloc[0] = 10
kser.iloc[0] = 10
self.assert_eq(kser, pser)
pser1.iloc[0] = 20
kser1.iloc[0] = 20
self.assert_eq(kser1, pser1)
def test_index_symmetric_difference(self):
pidx1 = pd.Index([1, 2, 3, 4])
pidx2 = pd.Index([2, 3, 4, 5])
kidx1 = ks.from_pandas(pidx1)
kidx2 = ks.from_pandas(pidx2)
self.assert_eq(
kidx1.symmetric_difference(kidx2).sort_values(),
pidx1.symmetric_difference(pidx2).sort_values(),
)
self.assert_eq(
(kidx1 + 1).symmetric_difference(kidx2).sort_values(),
(pidx1 + 1).symmetric_difference(pidx2).sort_values(),
)
pmidx1 = pd.MultiIndex(
[["lama", "cow", "falcon"], ["speed", "weight", "length"]],
[[0, 0, 0, 1, 1, 1, 2, 2, 2], [0, 0, 0, 0, 1, 2, 0, 1, 2]],
)
pmidx2 = pd.MultiIndex(
[["koalas", "cow", "falcon"], ["speed", "weight", "length"]],
[[0, 0, 0, 1, 1, 1, 2, 2, 2], [0, 0, 0, 0, 1, 2, 0, 1, 2]],
)
kmidx1 = ks.from_pandas(pmidx1)
kmidx2 = ks.from_pandas(pmidx2)
self.assert_eq(
kmidx1.symmetric_difference(kmidx2).sort_values(),
pmidx1.symmetric_difference(pmidx2).sort_values(),
)
idx = ks.Index(["a", "b", "c"])
midx = ks.MultiIndex.from_tuples([("a", "x"), ("b", "y"), ("c", "z")])
with self.assertRaisesRegexp(NotImplementedError, "Doesn't support*"):
idx.symmetric_difference(midx)
def test_index_nunique(self):
pidx = pd.Index([1, 1, 2, None])
kidx = ks.Index([1, 1, 2, None])
self.assert_eq(pidx.nunique(), kidx.nunique())
self.assert_eq(pidx.nunique(dropna=True), kidx.nunique(dropna=True))
def test_index_symmetric_difference(self):
idx = ks.Index(["a", "b", "c"])
midx = ks.MultiIndex.from_tuples([("a", "x"), ("b", "y"), ("c", "z")])
with self.assertRaisesRegexp(NotImplementedError, "Doesn't support*"):
idx.symmetric_difference(midx)
def test_difference(self):
# Index
kidx1 = ks.Index([1, 2, 3, 4], name="koalas")
kidx2 = ks.Index([3, 4, 5, 6], name="koalas")
pidx1 = kidx1.to_pandas()
pidx2 = kidx2.to_pandas()
self.assert_eq(
kidx1.difference(kidx2).sort_values(),
pidx1.difference(pidx2).sort_values())
self.assert_eq(
kidx1.difference([3, 4, 5, 6]).sort_values(),
pidx1.difference([3, 4, 5, 6]).sort_values(),
)
self.assert_eq(
kidx1.difference((3, 4, 5, 6)).sort_values(),
pidx1.difference((3, 4, 5, 6)).sort_values(),
)
self.assert_eq(
kidx1.difference({3, 4, 5, 6}).sort_values(),
pidx1.difference({3, 4, 5, 6}).sort_values(),
)
self.assert_eq(
kidx1.difference({
3: 1,
4: 2,
5: 3,
6: 4
}).sort_values(),
pidx1.difference({
3: 1,
4: 2,
5: 3,
6: 4
}).sort_values(),
)
# Exceptions for Index
with self.assertRaisesRegex(TypeError,
"Input must be Index or array-like"):
kidx1.difference("1234")
with self.assertRaisesRegex(TypeError,
"Input must be Index or array-like"):
kidx1.difference(1234)
with self.assertRaisesRegex(TypeError,
"Input must be Index or array-like"):
kidx1.difference(12.34)
with self.assertRaisesRegex(TypeError,
"Input must be Index or array-like"):
kidx1.difference(None)
with self.assertRaisesRegex(TypeError,
"Input must be Index or array-like"):
kidx1.difference(np.nan)
with self.assertRaisesRegex(
ValueError,
"The 'sort' keyword only takes the values of None or True; 1 was passed."
):
kidx1.difference(kidx2, sort=1)
# MultiIndex
kidx1 = ks.MultiIndex.from_tuples([("a", "x", 1), ("b", "y", 2),
("c", "z", 3)],
names=["hello", "koalas", "world"])
kidx2 = ks.MultiIndex.from_tuples([("a", "x", 1), ("b", "z", 2),
("k", "z", 3)],
names=["hello", "koalas", "world"])
pidx1 = kidx1.to_pandas()
pidx2 = kidx2.to_pandas()
self.assert_eq(
kidx1.difference(kidx2).sort_values(),
pidx1.difference(pidx2).sort_values())
self.assert_eq(
kidx1.difference({("a", "x", 1)}).sort_values(),
pidx1.difference({("a", "x", 1)}).sort_values(),
)
self.assert_eq(
kidx1.difference({
("a", "x", 1): [1, 2, 3]
}).sort_values(),
pidx1.difference({
("a", "x", 1): [1, 2, 3]
}).sort_values(),
)
# Exceptions for MultiIndex
with self.assertRaisesRegex(
TypeError, "other must be a MultiIndex or a list of tuples"):
kidx1.difference(["b", "z", "2"])
def test_append(self):
# Index
pidx = pd.Index(range(10000))
kidx = ks.Index(range(10000))
self.assert_eq(pidx.append(pidx), kidx.append(kidx))
# Index with name
pidx1 = pd.Index(range(10000), name="a")
pidx2 = pd.Index(range(10000), name="b")
kidx1 = ks.Index(range(10000), name="a")
kidx2 = ks.Index(range(10000), name="b")
self.assert_eq(pidx1.append(pidx2), kidx1.append(kidx2))
self.assert_eq(pidx2.append(pidx1), kidx2.append(kidx1))
# Index from DataFrame
pdf1 = pd.DataFrame({
"a": [1, 2, 3],
"b": [4, 5, 6]
},
index=["a", "b", "c"])
pdf2 = pd.DataFrame({
"a": [7, 8, 9],
"d": [10, 11, 12]
},
index=["x", "y", "z"])
kdf1 = ks.from_pandas(pdf1)
kdf2 = ks.from_pandas(pdf2)
pidx1 = pdf1.set_index("a").index
pidx2 = pdf2.set_index("d").index
kidx1 = kdf1.set_index("a").index
kidx2 = kdf2.set_index("d").index
self.assert_eq(pidx1.append(pidx2), kidx1.append(kidx2))
self.assert_eq(pidx2.append(pidx1), kidx2.append(kidx1))
# Index from DataFrame with MultiIndex columns
pdf1 = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
pdf2 = pd.DataFrame({"a": [7, 8, 9], "d": [10, 11, 12]})
pdf1.columns = pd.MultiIndex.from_tuples([("a", "x"), ("b", "y")])
pdf2.columns = pd.MultiIndex.from_tuples([("a", "x"), ("d", "y")])
kdf1 = ks.from_pandas(pdf1)
kdf2 = ks.from_pandas(pdf2)
pidx1 = pdf1.set_index(("a", "x")).index
pidx2 = pdf2.set_index(("d", "y")).index
kidx1 = kdf1.set_index(("a", "x")).index
kidx2 = kdf2.set_index(("d", "y")).index
self.assert_eq(pidx1.append(pidx2), kidx1.append(kidx2))
self.assert_eq(pidx2.append(pidx1), kidx2.append(kidx1))
# MultiIndex
pmidx = pd.MultiIndex.from_tuples([("a", "x", 1), ("b", "y", 2),
("c", "z", 3)])
kmidx = ks.MultiIndex.from_tuples([("a", "x", 1), ("b", "y", 2),
("c", "z", 3)])
self.assert_eq(pmidx.append(pmidx), kmidx.append(kmidx))
# MultiIndex with names
pmidx1 = pd.MultiIndex.from_tuples([("a", "x", 1), ("b", "y", 2),
("c", "z", 3)],
names=["x", "y", "z"])
pmidx2 = pd.MultiIndex.from_tuples([("a", "x", 1), ("b", "y", 2),
("c", "z", 3)],
names=["p", "q", "r"])
kmidx1 = ks.MultiIndex.from_tuples([("a", "x", 1), ("b", "y", 2),
("c", "z", 3)],
names=["x", "y", "z"])
kmidx2 = ks.MultiIndex.from_tuples([("a", "x", 1), ("b", "y", 2),
("c", "z", 3)],
names=["p", "q", "r"])
self.assert_eq(pmidx1.append(pmidx2), kmidx1.append(kmidx2))
self.assert_eq(pmidx2.append(pmidx1), kmidx2.append(kmidx1))
self.assert_eq(
pmidx1.append(pmidx2).names,
kmidx1.append(kmidx2).names)
self.assert_eq(
pmidx1.append(pmidx2).names,
kmidx1.append(kmidx2).names)
# Index & MultiIndex currently is not supported
expected_error_message = r"append\(\) between Index & MultiIndex currently is not supported"
with self.assertRaisesRegex(NotImplementedError,
expected_error_message):
kidx.append(kmidx)
with self.assertRaisesRegex(NotImplementedError,
expected_error_message):
kmidx.append(kidx)
def test_append(self):
# Index
pidx = pd.Index(range(10000))
kidx = ks.Index(range(10000))
self.assert_eq(pidx.append(pidx), kidx.append(kidx))
# Index with name
pidx1 = pd.Index(range(10000), name='a')
pidx2 = pd.Index(range(10000), name='b')
kidx1 = ks.Index(range(10000), name='a')
kidx2 = ks.Index(range(10000), name='b')
self.assert_eq(pidx1.append(pidx2), kidx1.append(kidx2))
self.assert_eq(pidx2.append(pidx1), kidx2.append(kidx1))
# Index from DataFrame
pdf1 = pd.DataFrame({
'a': [1, 2, 3],
'b': [4, 5, 6]
},
index=['a', 'b', 'c'])
pdf2 = pd.DataFrame({
'a': [7, 8, 9],
'd': [10, 11, 12]
},
index=['x', 'y', 'z'])
kdf1 = ks.from_pandas(pdf1)
kdf2 = ks.from_pandas(pdf2)
pidx1 = pdf1.set_index('a').index
pidx2 = pdf2.set_index('d').index
kidx1 = kdf1.set_index('a').index
kidx2 = kdf2.set_index('d').index
self.assert_eq(pidx1.append(pidx2), kidx1.append(kidx2))
self.assert_eq(pidx2.append(pidx1), kidx2.append(kidx1))
# Index from DataFrame with MultiIndex columns
pdf1 = pd.DataFrame({'a': [1, 2, 3], 'b': [4, 5, 6]})
pdf2 = pd.DataFrame({'a': [7, 8, 9], 'd': [10, 11, 12]})
pdf1.columns = pd.MultiIndex.from_tuples([('a', 'x'), ('b', 'y')])
pdf2.columns = pd.MultiIndex.from_tuples([('a', 'x'), ('d', 'y')])
kdf1 = ks.from_pandas(pdf1)
kdf2 = ks.from_pandas(pdf2)
pidx1 = pdf1.set_index(('a', 'x')).index
pidx2 = pdf2.set_index(('d', 'y')).index
kidx1 = kdf1.set_index(('a', 'x')).index
kidx2 = kdf2.set_index(('d', 'y')).index
self.assert_eq(pidx1.append(pidx2), kidx1.append(kidx2))
self.assert_eq(pidx2.append(pidx1), kidx2.append(kidx1))
# MultiIndex
pmidx = pd.MultiIndex.from_tuples([('a', 'x', 1), ('b', 'y', 2),
('c', 'z', 3)])
kmidx = ks.MultiIndex.from_tuples([('a', 'x', 1), ('b', 'y', 2),
('c', 'z', 3)])
self.assert_eq(pmidx.append(pmidx), kmidx.append(kmidx))
# MultiIndex with names
pmidx1 = pd.MultiIndex.from_tuples([('a', 'x', 1), ('b', 'y', 2),
('c', 'z', 3)],
names=['x', 'y', 'z'])
pmidx2 = pd.MultiIndex.from_tuples([('a', 'x', 1), ('b', 'y', 2),
('c', 'z', 3)],
names=['p', 'q', 'r'])
kmidx1 = ks.MultiIndex.from_tuples([('a', 'x', 1), ('b', 'y', 2),
('c', 'z', 3)],
names=['x', 'y', 'z'])
kmidx2 = ks.MultiIndex.from_tuples([('a', 'x', 1), ('b', 'y', 2),
('c', 'z', 3)],
names=['p', 'q', 'r'])
self.assert_eq(pmidx1.append(pmidx2), kmidx1.append(kmidx2))
self.assert_eq(pmidx2.append(pmidx1), kmidx2.append(kmidx1))
self.assert_eq(
pmidx1.append(pmidx2).names,
kmidx1.append(kmidx2).names)
self.assert_eq(
pmidx1.append(pmidx2).names,
kmidx1.append(kmidx2).names)
# Index & MultiIndex currently is not supported
expected_error_message = r"append\(\) between Index & MultiIndex currently is not supported"
with self.assertRaisesRegex(NotImplementedError,
expected_error_message):
kidx.append(kmidx)
with self.assertRaisesRegex(NotImplementedError,
expected_error_message):
kmidx.append(kidx)