Skip to content

15819 rolling window on empty df #16431

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 8 commits into from
May 25, 2017
Merged

15819 rolling window on empty df #16431

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
9cbaef6
BUG Support empty DataFrame for datetime rolling window
chernrick May 22, 2017
d31b983
BUG update release note for issue 15819
chernrick May 22, 2017
5fb1170
Address code review comments
chernrick May 22, 2017
670a33a
Add additional tests from code review and move release note to rollin…
chernrick May 23, 2017
e913664
Separate TestRolling.test_empty_df_datetime_index_rolling_sum() into …
chernrick May 23, 2017
985522f
Combine tests into parameterized tests; update grammar in release note
chernrick May 24, 2017
6007654
Fix typo in comment
chernrick May 24, 2017
4a81a5e
Address code review comments
chernrick May 25, 2017
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions doc/source/whatsnew/v0.20.2.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,6 +37,7 @@ Bug Fixes
~~~~~~~~~

- Bug in using ``pathlib.Path`` or ``py.path.local`` objects with io functions (:issue:`16291`)
- Bug creating datetime rolling window on empty DataFrame (:issue:`15819`)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

can you move to the rolling bug fix section


Conversion
^^^^^^^^^^
Expand Down
2 changes: 1 addition & 1 deletion pandas/core/window.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1074,7 +1074,7 @@ def validate(self):
super(Rolling, self).validate()

# we allow rolling on a datetimelike index
if (self.is_datetimelike and
if ((self.obj.empty or self.is_datetimelike) and
isinstance(self.window, (compat.string_types, DateOffset,
timedelta))):

Expand Down
14 changes: 14 additions & 0 deletions pandas/tests/test_window.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -441,6 +441,20 @@ def test_closed(self):
with pytest.raises(ValueError):
df.rolling(window=3, closed='neither')

def test_empty_df_datetime_rolling_sum(self):
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

rather than making all separate tests, either use parametrize, or you can consolidate these into 2 tests for rolling (and 2 for expanding)

e.g.

@pytest.mark.parametrize('roller', ['1s', 1]):
def tests_empty_df_rolling(self, roller):
    expected = DataFrame()
    result = DataFrame().rolling(roller).sum()
    tm.assert_frame_equal(result, expected)

    expected = DataFrame(index=pd.DatetimeIndex([]))
    result = DataFrame(index=pd.DatetimeIndex([])).rolling(roller).sum()
    tm.assert_frame_equal(result, expected)

# Verifies that datetime rolling window can be applied to empty DataFrame
# GH 15819
result = DataFrame().rolling('1s').sum()
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Cna you leave a comment here with the github issue number?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ok, I have updated with a comment

expected = DataFrame()
tm.assert_frame_equal(result, expected)

def test_empty_df_integer_rolling_sum(self):
# Verifies that integer rolling window can be applied to empty DataFrame
# GH 15819
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

can you add the same tests for expanding (not that the expanding will raise for the time-aware, a separate issue, so xfail that one, put a ref to #16425 there).

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

can you add these as well (which are known working)

DataFrame(index=pd.DatetimeIndex([])).rolling(1).sum()

DataFrame(index=pd.DatetimeIndex([])).rolling('1s').sum()

result = DataFrame().rolling(1).sum()
expected = DataFrame()
tm.assert_frame_equal(result, expected)


class TestExpanding(Base):

Expand Down