Skip to content

Fix wrong save of datetime64[s] in HDFStore #59018

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 9 commits into from
Jun 18, 2024
+27 −7
Merged

Fix wrong save of datetime64[s] in HDFStore #59018

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
52a45be
Fix wrong save of datetime64[s] in HDFStore
chaoyihu Jun 14, 2024
d235341
generic datetime unit parsing
chaoyihu Jun 14, 2024
135f2de
use tmp_path
chaoyihu Jun 14, 2024
b842245
Adding entry to whatsnew
chaoyihu Jun 14, 2024
afd6526
datetime64 dtype parsing using numpy api
chaoyihu Jun 15, 2024
bcf6572
move whatsnew entry
chaoyihu Jun 15, 2024
b8ee8f1
update test comment
chaoyihu Jun 15, 2024
bf3ee62
update hdfstore dtypes test case
chaoyihu Jun 15, 2024
b794f6d
Merge branch 'main' into hdfstore-datetime64-wrong-unit
mroeschke Jun 18, 2024
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
12 changes: 9 additions & 3 deletions pandas/io/pytables.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3036,7 +3036,10 @@ def read_array(self, key: str, start: int | None = None, stop: int | None = None
if dtype and dtype.startswith("datetime64"):
# reconstruct a timezone if indicated
tz = getattr(attrs, "tz", None)
ret = _set_tz(ret, tz)
if dtype == "datetime64[s]":
ret = _set_tz(ret, tz, unit="s")
else:
ret = _set_tz(ret, tz) # default unit is ns

elif dtype == "timedelta64":
ret = np.asarray(ret, dtype="m8[ns]")
Expand Down Expand Up @@ -4964,19 +4967,22 @@ def _get_tz(tz: tzinfo) -> str | tzinfo:
return zone


def _set_tz(values: npt.NDArray[np.int64], tz: str | tzinfo | None) -> DatetimeArray:
def _set_tz(
values: npt.NDArray[np.int64], tz: str | tzinfo | None, unit: str = "ns"
) -> DatetimeArray:
"""
Coerce the values to a DatetimeArray with appropriate tz.

Parameters
----------
values : ndarray[int64]
tz : str, tzinfo, or None
unit : str. The default unit is ns. Needs to be specified otherwise.
"""
assert values.dtype == "i8", values.dtype
# Argument "tz" to "tz_to_dtype" has incompatible type "str | tzinfo | None";
# expected "tzinfo"
dtype = tz_to_dtype(tz=tz, unit="ns") # type: ignore[arg-type]
dtype = tz_to_dtype(tz=tz, unit=unit) # type: ignore[arg-type]
dta = DatetimeArray._from_sequence(values, dtype=dtype)
return dta

Expand Down
11 changes: 11 additions & 0 deletions pandas/tests/io/pytables/test_read.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -317,3 +317,14 @@ def test_read_infer_string(tmp_path, setup_path):
columns=Index(["a"], dtype="string[pyarrow_numpy]"),
)
tm.assert_frame_equal(result, expected)


def test_hdfstore_read_datetime64_unit_s():
# Fix issue 59004: HDFStore doesn't save datetime64[s] right
df_s = DataFrame(["2001-01-01", "2002-02-02"], dtype="datetime64[s]")
with HDFStore("deleteme.h5", mode="w") as store:
store.put("df_s", df_s)
with HDFStore("deleteme.h5", mode="r") as store:
df_fromstore = store.get("df_s")
tm.assert_frame_equal(df_s, df_fromstore)
Path("deleteme.h5").unlink() # Delete created file
Loading