Skip to content

BUG: guess_datetime_format doesn't guess 2018-01-01T00:00:00.000 #49047

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.

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 2 commits into from
Oct 11, 2022
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
16 changes: 14 additions & 2 deletions pandas/_libs/tslibs/parsing.pyx
Original file line number Diff line number Diff line change
Expand Up @@ -976,7 +976,6 @@ def guess_datetime_format(dt_str: str, bint dayfirst=False) -> str | None:
(('hour',), '%H', 2),
(('minute',), '%M', 2),
(('second',), '%S', 2),
(('microsecond',), '%f', 6),
Copy link
Member Author

Choose a reason for hiding this comment

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

this doesn't seem to make a difference - at least, I don't think it's covered by existing tests

(('second', 'microsecond'), '%S.%f', 0),
(('tzinfo',), '%z', 0),
(('tzinfo',), '%Z', 0),
Expand Down Expand Up @@ -1048,7 +1047,7 @@ def guess_datetime_format(dt_str: str, bint dayfirst=False) -> str | None:

parsed_formatted = parsed_datetime.strftime(attr_format)
for i, token_format in enumerate(format_guess):
token_filled = tokens[i].zfill(padding)
token_filled = _fill_token(tokens[i], padding)
if token_format is None and token_filled == parsed_formatted:
format_guess[i] = attr_format
tokens[i] = token_filled
Expand Down Expand Up @@ -1090,6 +1089,19 @@ def guess_datetime_format(dt_str: str, bint dayfirst=False) -> str | None:
else:
return None

cdef str _fill_token(token: str, padding: int):
cdef str token_filled
if '.' not in token:
token_filled = token.zfill(padding)
else:
seconds, nanoseconds = token.split('.')
seconds = f'{int(seconds):02d}'
# right-pad so we get nanoseconds, then only take
# first 6 digits (microseconds) as stdlib datetime
# doesn't support nanoseconds
nanoseconds = nanoseconds.ljust(9, '0')[:6]
token_filled = f'{seconds}.{nanoseconds}'
return token_filled

@cython.wraparound(False)
@cython.boundscheck(False)
Expand Down
2 changes: 1 addition & 1 deletion pandas/core/tools/datetimes.py
Original file line number Diff line number Diff line change
Expand Up @@ -820,7 +820,7 @@ def to_datetime(
to the day starting at noon on January 1, 4713 BC.
- If Timestamp convertible (Timestamp, dt.datetime, np.datetimt64 or date
string), origin is set to Timestamp identified by origin.
- If a float or integer, origin is the mullisecond difference
- If a float or integer, origin is the millisecond difference
relative to 1970-01-01.
cache : bool, default True
If :const:`True`, use a cache of unique, converted dates to apply the
Expand Down
15 changes: 15 additions & 0 deletions pandas/tests/tslibs/test_parsing.py
Original file line number Diff line number Diff line change
Expand Up @@ -295,3 +295,18 @@ def test_is_iso_format(fmt, expected):
# see gh-41047
result = parsing.format_is_iso(fmt)
assert result == expected


@pytest.mark.parametrize(
"input",
[
"2018-01-01T00:00:00.123456789",
"2018-01-01T00:00:00.123456",
"2018-01-01T00:00:00.123",
],
)
def test_guess_datetime_format_f(input):
# https://github.com/pandas-dev/pandas/issues/49043
result = parsing.guess_datetime_format(input)
expected = "%Y-%m-%dT%H:%M:%S.%f"
assert result == expected