craftbeerpi4-pione/venv/lib/python3.8/site-packages/pandas/tests/frame/test_api.py

348 lines
11 KiB
Python

from copy import deepcopy
import inspect
import pydoc
import warnings
import numpy as np
import pytest
import pandas.util._test_decorators as td
from pandas.util._test_decorators import async_mark, skip_if_no
import pandas as pd
from pandas import DataFrame, Series, date_range, timedelta_range
import pandas._testing as tm
class TestDataFrameMisc:
def test_getitem_pop_assign_name(self, float_frame):
s = float_frame["A"]
assert s.name == "A"
s = float_frame.pop("A")
assert s.name == "A"
s = float_frame.loc[:, "B"]
assert s.name == "B"
s2 = s.loc[:]
assert s2.name == "B"
def test_get_axis(self, float_frame):
f = float_frame
assert f._get_axis_number(0) == 0
assert f._get_axis_number(1) == 1
assert f._get_axis_number("index") == 0
assert f._get_axis_number("rows") == 0
assert f._get_axis_number("columns") == 1
assert f._get_axis_name(0) == "index"
assert f._get_axis_name(1) == "columns"
assert f._get_axis_name("index") == "index"
assert f._get_axis_name("rows") == "index"
assert f._get_axis_name("columns") == "columns"
assert f._get_axis(0) is f.index
assert f._get_axis(1) is f.columns
with pytest.raises(ValueError, match="No axis named"):
f._get_axis_number(2)
with pytest.raises(ValueError, match="No axis.*foo"):
f._get_axis_name("foo")
with pytest.raises(ValueError, match="No axis.*None"):
f._get_axis_name(None)
with pytest.raises(ValueError, match="No axis named"):
f._get_axis_number(None)
def test_column_contains_raises(self, float_frame):
with pytest.raises(TypeError, match="unhashable type: 'Index'"):
float_frame.columns in float_frame
def test_tab_completion(self):
# DataFrame whose columns are identifiers shall have them in __dir__.
df = DataFrame([list("abcd"), list("efgh")], columns=list("ABCD"))
for key in list("ABCD"):
assert key in dir(df)
assert isinstance(df.__getitem__("A"), pd.Series)
# DataFrame whose first-level columns are identifiers shall have
# them in __dir__.
df = DataFrame(
[list("abcd"), list("efgh")],
columns=pd.MultiIndex.from_tuples(list(zip("ABCD", "EFGH"))),
)
for key in list("ABCD"):
assert key in dir(df)
for key in list("EFGH"):
assert key not in dir(df)
assert isinstance(df.__getitem__("A"), pd.DataFrame)
def test_not_hashable(self):
empty_frame = DataFrame()
df = DataFrame([1])
msg = "'DataFrame' objects are mutable, thus they cannot be hashed"
with pytest.raises(TypeError, match=msg):
hash(df)
with pytest.raises(TypeError, match=msg):
hash(empty_frame)
def test_column_name_contains_unicode_surrogate(self):
# GH 25509
colname = "\ud83d"
df = DataFrame({colname: []})
# this should not crash
assert colname not in dir(df)
assert df.columns[0] == colname
def test_new_empty_index(self):
df1 = DataFrame(np.random.randn(0, 3))
df2 = DataFrame(np.random.randn(0, 3))
df1.index.name = "foo"
assert df2.index.name is None
def test_get_agg_axis(self, float_frame):
cols = float_frame._get_agg_axis(0)
assert cols is float_frame.columns
idx = float_frame._get_agg_axis(1)
assert idx is float_frame.index
msg = r"Axis must be 0 or 1 \(got 2\)"
with pytest.raises(ValueError, match=msg):
float_frame._get_agg_axis(2)
def test_empty(self, float_frame, float_string_frame):
empty_frame = DataFrame()
assert empty_frame.empty
assert not float_frame.empty
assert not float_string_frame.empty
# corner case
df = DataFrame({"A": [1.0, 2.0, 3.0], "B": ["a", "b", "c"]}, index=np.arange(3))
del df["A"]
assert not df.empty
def test_len(self, float_frame):
assert len(float_frame) == len(float_frame.index)
# single block corner case
arr = float_frame[["A", "B"]].values
expected = float_frame.reindex(columns=["A", "B"]).values
tm.assert_almost_equal(arr, expected)
def test_axis_aliases(self, float_frame):
f = float_frame
# reg name
expected = f.sum(axis=0)
result = f.sum(axis="index")
tm.assert_series_equal(result, expected)
expected = f.sum(axis=1)
result = f.sum(axis="columns")
tm.assert_series_equal(result, expected)
def test_class_axis(self):
# GH 18147
# no exception and no empty docstring
assert pydoc.getdoc(DataFrame.index)
assert pydoc.getdoc(DataFrame.columns)
def test_series_put_names(self, float_string_frame):
series = float_string_frame._series
for k, v in series.items():
assert v.name == k
def test_empty_nonzero(self):
df = DataFrame([1, 2, 3])
assert not df.empty
df = DataFrame(index=[1], columns=[1])
assert not df.empty
df = DataFrame(index=["a", "b"], columns=["c", "d"]).dropna()
assert df.empty
assert df.T.empty
empty_frames = [
DataFrame(),
DataFrame(index=[1]),
DataFrame(columns=[1]),
DataFrame({1: []}),
]
for df in empty_frames:
assert df.empty
assert df.T.empty
def test_with_datetimelikes(self):
df = DataFrame(
{
"A": date_range("20130101", periods=10),
"B": timedelta_range("1 day", periods=10),
}
)
t = df.T
result = t.dtypes.value_counts()
expected = Series({np.dtype("object"): 10})
tm.assert_series_equal(result, expected)
def test_deepcopy(self, float_frame):
cp = deepcopy(float_frame)
series = cp["A"]
series[:] = 10
for idx, value in series.items():
assert float_frame["A"][idx] != value
def test_inplace_return_self(self):
# GH 1893
data = DataFrame(
{"a": ["foo", "bar", "baz", "qux"], "b": [0, 0, 1, 1], "c": [1, 2, 3, 4]}
)
def _check_f(base, f):
result = f(base)
assert result is None
# -----DataFrame-----
# set_index
f = lambda x: x.set_index("a", inplace=True)
_check_f(data.copy(), f)
# reset_index
f = lambda x: x.reset_index(inplace=True)
_check_f(data.set_index("a"), f)
# drop_duplicates
f = lambda x: x.drop_duplicates(inplace=True)
_check_f(data.copy(), f)
# sort
f = lambda x: x.sort_values("b", inplace=True)
_check_f(data.copy(), f)
# sort_index
f = lambda x: x.sort_index(inplace=True)
_check_f(data.copy(), f)
# fillna
f = lambda x: x.fillna(0, inplace=True)
_check_f(data.copy(), f)
# replace
f = lambda x: x.replace(1, 0, inplace=True)
_check_f(data.copy(), f)
# rename
f = lambda x: x.rename({1: "foo"}, inplace=True)
_check_f(data.copy(), f)
# -----Series-----
d = data.copy()["c"]
# reset_index
f = lambda x: x.reset_index(inplace=True, drop=True)
_check_f(data.set_index("a")["c"], f)
# fillna
f = lambda x: x.fillna(0, inplace=True)
_check_f(d.copy(), f)
# replace
f = lambda x: x.replace(1, 0, inplace=True)
_check_f(d.copy(), f)
# rename
f = lambda x: x.rename({1: "foo"}, inplace=True)
_check_f(d.copy(), f)
@async_mark()
@td.check_file_leaks
async def test_tab_complete_warning(self, ip, frame_or_series):
# GH 16409
pytest.importorskip("IPython", minversion="6.0.0")
from IPython.core.completer import provisionalcompleter
if frame_or_series is DataFrame:
code = "from pandas import DataFrame; obj = DataFrame()"
else:
code = "from pandas import Series; obj = Series(dtype=object)"
await ip.run_code(code)
# TODO: remove it when Ipython updates
# GH 33567, jedi version raises Deprecation warning in Ipython
import jedi
if jedi.__version__ < "0.17.0":
warning = tm.assert_produces_warning(None)
else:
warning = tm.assert_produces_warning(
DeprecationWarning, check_stacklevel=False
)
with warning:
with provisionalcompleter("ignore"):
list(ip.Completer.completions("obj.", 1))
def test_attrs(self):
df = DataFrame({"A": [2, 3]})
assert df.attrs == {}
df.attrs["version"] = 1
result = df.rename(columns=str)
assert result.attrs == {"version": 1}
@pytest.mark.parametrize("allows_duplicate_labels", [True, False, None])
def test_set_flags(self, allows_duplicate_labels, frame_or_series):
obj = DataFrame({"A": [1, 2]})
key = (0, 0)
if frame_or_series is Series:
obj = obj["A"]
key = 0
result = obj.set_flags(allows_duplicate_labels=allows_duplicate_labels)
if allows_duplicate_labels is None:
# We don't update when it's not provided
assert result.flags.allows_duplicate_labels is True
else:
assert result.flags.allows_duplicate_labels is allows_duplicate_labels
# We made a copy
assert obj is not result
# We didn't mutate obj
assert obj.flags.allows_duplicate_labels is True
# But we didn't copy data
result.iloc[key] = 0
assert obj.iloc[key] == 0
# Now we do copy.
result = obj.set_flags(
copy=True, allows_duplicate_labels=allows_duplicate_labels
)
result.iloc[key] = 10
assert obj.iloc[key] == 0
@skip_if_no("jinja2")
def test_constructor_expanddim_lookup(self):
# GH#33628 accessing _constructor_expanddim should not
# raise NotImplementedError
df = DataFrame()
with warnings.catch_warnings(record=True) as wrn:
# _AXIS_NUMBERS, _AXIS_NAMES lookups
inspect.getmembers(df)
# some versions give FutureWarning, others DeprecationWarning
assert len(wrn)
assert any(x.category in [FutureWarning, DeprecationWarning] for x in wrn)
with pytest.raises(NotImplementedError, match="Not supported for DataFrames!"):
df._constructor_expanddim(np.arange(27).reshape(3, 3, 3))