def test_join_on_fails_with_different_left_index(self):
df = DataFrame(
{"a": np.random.choice(["m", "f"], size=3), "b": np.random.randn(3)},
index=tm.makeCustomIndex(3, 2),
)
df2 = DataFrame(
{"a": np.random.choice(["m", "f"], size=10), "b": np.random.randn(10)}
)
msg = r'len\(right_on\) must equal the number of levels in the index of "left"'
with pytest.raises(ValueError, match=msg):
merge(df, df2, right_on="b", left_index=True)
def test_join_on_fails_with_different_column_counts(self):
df = DataFrame(
{"a": np.random.choice(["m", "f"], size=3), "b": np.random.randn(3)}
)
df2 = DataFrame(
{"a": np.random.choice(["m", "f"], size=10), "b": np.random.randn(10)},
index=tm.makeCustomIndex(10, 2),
)
msg = r"len\(right_on\) must equal len\(left_on\)"
with pytest.raises(ValueError, match=msg):
merge(df, df2, right_on="a", left_on=["a", "b"])
def metadata_column_headers(request: FixtureRequest) -> List[str]:
"""Make a list of metadata column headers.
Returns:
list: A metadata column header list whose length is between 0
and `MAX_METADATA_COLS`.
"""
if request.param == 0:
return list()
# pandas bug (?) in makeCustomIndex when nentries = 1
elif request.param == 1:
return ["M_l0_g0"]
else:
return list(makeCustomIndex(request.param, 1, prefix="M"))
def f(df):
return df["close"] / df["open"]
# it works!
result = grouped.apply(f)
tm.assert_index_equal(result.index, df.index)
@pytest.mark.parametrize(
"name, func",
[
("Int64Index", tm.makeIntIndex),
("Index", tm.makeUnicodeIndex),
("Float64Index", tm.makeFloatIndex),
("MultiIndex", lambda m: tm.makeCustomIndex(m, 2)),
],
)
def test_fails_on_no_datetime_index(name, func):
n = 2
index = func(n)
df = DataFrame({"a": np.random.randn(n)}, index=index)
msg = (
"Only valid with DatetimeIndex, TimedeltaIndex "
f"or PeriodIndex, but got an instance of '{name}'"
)
with pytest.raises(TypeError, match=msg):
df.groupby(Grouper(freq="D"))
class TestResolvePath(object):
archive = ".\\"
headers: Index = makeCustomIndex(7, 1, prefix="C")
multi_schema = SortedDict({"0": headers[1], "1": headers[3]})
multi_schema2 = SortedDict({
"0": headers[0],
"1": headers[2],
"2": headers[3]
})
multi_schema3 = SortedDict({"0": headers[1], "1": headers[6]})
single_schema = SortedDict({"0": headers[0]})
single_schema2 = SortedDict({"0": headers[4]})
@staticmethod
def data_gen_invalid(row: int, col: int) -> Union[float, str]:
# Invalid data has:
# 1) multiple data values per column (excluding NaNs and empty strings).
# fmt: off
valmap_invalid: List[List[Union[float, str]]] = [
["val", "val", "xxx", "val"],
["val", "", "val", "val"], # noqa: E241
["", "val", "val", nan], # noqa: E201, E241
[nan, nan, "val", "xxx"], # noqa: E201, E241
["xxx", nan, "", nan], # noqa: E241
]
# fmt: on
if row < len(valmap_invalid):
if col < len(valmap_invalid[row]):
return valmap_invalid[row][col]
# Use pandas' value generation function if the given (row, column) falls
# outside our pre-defined invalid value map.
return make_dataframe_value(row, col)
@staticmethod
def data_gen(row: int, col: int) -> Union[float, str]:
# Valid data has:
# 1) the same data value for each column (excluding NaNs and empty strings).
# fmt: off
valmap: List[List[Union[float, str]]] = [
["val", "val", "val", "val"],
["val", "", "val", "val"], # noqa: E241
["", "val", "val", nan], # noqa: E201, E241
[nan, nan, "val", "val"], # noqa: E201, E241
["val", nan, "", nan], # noqa: E241
]
# fmt: on
if row < len(valmap):
if col < len(valmap[row]):
return valmap[row][col]
# Fall back on pandas' value generator.
return make_dataframe_value(row, col)
@staticmethod
def data_gen_normalizable(row: int, col: int) -> Union[float, str]:
# A valid data map containing data values that will be normalized by
# `syphon.schema.resolvepath._normalize`.
# fmt: off
valmap: List[List[Union[float, str]]] = [
["Value 1.", "Value 1.", "Value 1.", "Value 1."],
["Value 1.", nan, "Value 1.", "Value 1."], # noqa: E241
["Value 1.", nan, "Value 1.", nan], # noqa: E241
[nan, nan, "Value 1.", nan], # noqa: E201, E241
[nan, nan, "Value 1.", nan], # noqa: E201, E241
]
# fmt: on
if row < len(valmap):
if col < len(valmap[row]):
return valmap[row][col]
# Fall back on pandas' value generator.
return make_dataframe_value(row, col)
@pytest.mark.parametrize(
"schema, expected",
[
(single_schema, join(archive, "val")),
(multi_schema, join(archive, "val", "val")),
(multi_schema2, join(archive, "val", "val", "val")),
],
)
def test_resolve_path(self, schema: SortedDict, expected: str):
data: DataFrame = make_dataframe(5,
4,
data_gen_f=TestResolvePath.data_gen)
actual: str = resolve_path(self.archive, schema, data)
assert actual == expected
@pytest.mark.parametrize(
"schema, expected",
[
(single_schema, join(archive, "value_1")),
(multi_schema, join(archive, "value_1", "value_1")),
(multi_schema2, join(archive, "value_1", "value_1", "value_1")),
],
)
def test_resolve_path_normalized(self, schema: SortedDict, expected: str):
data: DataFrame = make_dataframe(
5, 4, data_gen_f=TestResolvePath.data_gen_normalizable)
actual: str = resolve_path(self.archive, schema, data)
assert actual == expected
@pytest.mark.parametrize("schema", [single_schema2, multi_schema3])
def test_resolve_path_indexerror(self, schema: SortedDict):
data: DataFrame = make_dataframe(5,
4,
data_gen_f=TestResolvePath.data_gen)
with pytest.raises(IndexError):
resolve_path(self.archive, schema, data)
@pytest.mark.parametrize("schema",
[single_schema, multi_schema, multi_schema2])
def test_resolve_path_valueerror(self, schema: SortedDict):
data: DataFrame = make_dataframe(
5, 4, data_gen_f=TestResolvePath.data_gen_invalid)
with pytest.raises(ValueError):
resolve_path(self.archive, schema, data)