Skip to content
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

Jump to bottom

fix: raise ValueError if date is out-of-bounds #46

Merged
merged 3 commits into from
Dec 4, 2021
Merged

fix: raise ValueError if date is out-of-bounds #46

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
8399215
fix: raise ValueError if date is out-of-bounds
tswast Dec 3, 2021
f95e396
unify _datetime return type
tswast Dec 3, 2021
2ea4e51
add relevant unit test
tswast Dec 3, 2021
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
50 changes: 30 additions & 20 deletions db_dtypes/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,7 @@

import datetime
import re
from typing import Union
from typing import Optional, Union

import numpy
import packaging.version
Expand Down Expand Up @@ -103,7 +103,7 @@ def _datetime(
r"(?::(?P<seconds>\d+)"
r"(?:\.(?P<fraction>\d*))?)?)?\s*$"
).match,
):
) -> Optional[numpy.datetime64]:
# Convert pyarrow values to datetime.time.
if isinstance(scalar, (pyarrow.Time32Scalar, pyarrow.Time64Scalar)):
scalar = (
Expand All @@ -115,8 +115,16 @@ def _datetime(

if scalar is None:
return None
elif isinstance(scalar, datetime.time):
return datetime.datetime.combine(_EPOCH, scalar)
if isinstance(scalar, datetime.time):
return pandas.Timestamp(
year=1970,
month=1,
day=1,
hour=scalar.hour,
minute=scalar.minute,
second=scalar.second,
microsecond=scalar.microsecond,
).to_datetime64()
elif isinstance(scalar, pandas.Timestamp):
return scalar.to_datetime64()
elif isinstance(scalar, str):
Expand All @@ -125,20 +133,20 @@ def _datetime(
if not parsed:
raise ValueError(f"Bad time string: {repr(scalar)}")

hours = parsed.group("hours")
minutes = parsed.group("minutes")
seconds = parsed.group("seconds")
hour = parsed.group("hours")
minute = parsed.group("minutes")
second = parsed.group("seconds")
fraction = parsed.group("fraction")
microseconds = int(fraction.ljust(6, "0")[:6]) if fraction else 0
return datetime.datetime(
1970,
1,
1,
int(hours),
int(minutes) if minutes else 0,
int(seconds) if seconds else 0,
microseconds,
)
nanosecond = int(fraction.ljust(9, "0")[:9]) if fraction else 0
return pandas.Timestamp(
year=1970,
month=1,
day=1,
hour=int(hour),
minute=int(minute) if minute else 0,
second=int(second) if second else 0,
nanosecond=nanosecond,
).to_datetime64()
Copy link
Collaborator Author

Choose a reason for hiding this comment

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

Towards #19

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

Probably should have been a separate PR, but I noticed _datetime had inconsistent types (sometimes pandas.Timestamp, sometime datetime.datetime, sometime numpy.datetime64). We want a numpy.datetime64 in the end, so I unified it.

else:
raise TypeError("Invalid value type", scalar)

Expand Down Expand Up @@ -225,23 +233,25 @@ class DateArray(core.BaseDatetimeArray):
def _datetime(
scalar,
match_fn=re.compile(r"\s*(?P<year>\d+)-(?P<month>\d+)-(?P<day>\d+)\s*$").match,
):
) -> Optional[numpy.datetime64]:
# Convert pyarrow values to datetime.date.
if isinstance(scalar, (pyarrow.Date32Scalar, pyarrow.Date64Scalar)):
scalar = scalar.as_py()

if scalar is None:
return None
elif isinstance(scalar, datetime.date):
return datetime.datetime(scalar.year, scalar.month, scalar.day)
return pandas.Timestamp(
year=scalar.year, month=scalar.month, day=scalar.day
).to_datetime64()
elif isinstance(scalar, str):
match = match_fn(scalar)
if not match:
raise ValueError(f"Bad date string: {repr(scalar)}")
year = int(match.group("year"))
month = int(match.group("month"))
day = int(match.group("day"))
return datetime.datetime(year, month, day)
return pandas.Timestamp(year=year, month=month, day=day).to_datetime64()
else:
raise TypeError("Invalid value type", scalar)

Expand Down
4 changes: 1 addition & 3 deletions db_dtypes/core.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -127,9 +127,7 @@ def take(
if allow_fill:
fill_value = self._validate_scalar(fill_value)
fill_value = (
numpy.datetime64()
if fill_value is None
else numpy.datetime64(self._datetime(fill_value))
numpy.datetime64() if fill_value is None else self._datetime(fill_value)
)
if (indices < -1).any():
raise ValueError(
Expand Down
25 changes: 20 additions & 5 deletions tests/unit/test_arrow.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -183,13 +183,13 @@ def types_mapper(
type=pyarrow.time64("us"),
),
),
(
# Only microseconds are supported when reading data. See:
# https://github.com/googleapis/python-db-dtypes-pandas/issues/19
# Still, round-trip with pyarrow nanosecond precision scalars
# is supported.
pytest.param(
pandas.Series(
[
# Only microseconds are supported when reading data. See:
# https://github.com/googleapis/python-db-dtypes-pandas/issues/19
# Still, round-trip with pyarrow nanosecond precision scalars
# is supported.
pyarrow.scalar(0, pyarrow.time64("ns")),
pyarrow.scalar(
12 * HOUR_NANOS
Expand All @@ -216,6 +216,21 @@ def types_mapper(
],
type=pyarrow.time64("ns"),
),
id="time-nanoseconds-arrow-round-trip",
Copy link
Collaborator Author

Choose a reason for hiding this comment

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

Leftover from debugging why when I made the "time" _datetime method return a Timestamp too, this broke. Since that change isn't necessary and it broke this test I reverted it. Kept the test id, since I thought it was useful.

Copy link

Choose a reason for hiding this comment

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

Test ID is indeed useful IMO. 👍

),
pytest.param(
pandas.Series(
["0:0:0", "12:30:15.123456789", "23:59:59.999999999"], dtype="dbtime",
),
pyarrow.array(
[
0,
12 * HOUR_NANOS + 30 * MINUTE_NANOS + 15 * SECOND_NANOS + 123_456_789,
23 * HOUR_NANOS + 59 * MINUTE_NANOS + 59 * SECOND_NANOS + 999_999_999,
],
type=pyarrow.time64("ns"),
),
id="time-nanoseconds-arrow-from-string",
),
]

Expand Down
5 changes: 5 additions & 0 deletions tests/unit/test_date.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,6 +55,11 @@ def test_date_parsing(value, expected):
("2021-2-99", "day is out of range for month"),
("2021-99-1", "month must be in 1[.][.]12"),
("10000-1-1", "year 10000 is out of range"),
# Outside of min/max values pandas.Timestamp.
("0001-01-01", "Out of bounds"),
("9999-12-31", "Out of bounds"),
("1677-09-21", "Out of bounds"),
("2262-04-12", "Out of bounds"),
],
)
def test_date_parsing_errors(value, error):
Expand Down