From cbf5299a065e20a5b129ad5eed6953262ce54f37 Mon Sep 17 00:00:00 2001 From: Elliott Sales de Andrade Date: Wed, 21 Feb 2024 06:55:19 -0500 Subject: [PATCH 6/6] Fix accidental loss-of-precision for to_datetime(str, unit=...) In Pandas 1.5.3, the `float(val)` cast was inline to the `cast_from_unit` call in `array_with_unit_to_datetime`. This caused the intermediate (unnamed) value to be a Python float. Since #50301, a temporary variable was added to avoid multiple casts, but with explicit type `cdef float`, which defines a _Cython_ float. This type is 32-bit, and causes a loss of precision, and a regression in parsing from 1.5.3. So widen the explicit type of the temporary `fval` variable to (64-bit) `double`, which will not lose precision. Fixes #57051 Signed-off-by: Elliott Sales de Andrade --- pandas/_libs/tslib.pyx | 2 +- pandas/tests/tools/test_to_datetime.py | 8 ++++++++ 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/pandas/_libs/tslib.pyx b/pandas/_libs/tslib.pyx index 017fdc4bc8..dd23c2f27c 100644 --- a/pandas/_libs/tslib.pyx +++ b/pandas/_libs/tslib.pyx @@ -277,7 +277,7 @@ def array_with_unit_to_datetime( bint is_raise = errors == "raise" ndarray[int64_t] iresult tzinfo tz = None - float fval + double fval assert is_ignore or is_coerce or is_raise diff --git a/pandas/tests/tools/test_to_datetime.py b/pandas/tests/tools/test_to_datetime.py index 6791ac0340..a4194dcff2 100644 --- a/pandas/tests/tools/test_to_datetime.py +++ b/pandas/tests/tools/test_to_datetime.py @@ -1912,6 +1912,14 @@ class TestToDatetimeUnit: with pytest.raises(ValueError, match=msg): to_datetime([1], unit="D", format="%Y%m%d", cache=cache) + def test_unit_str(self, cache): + # GH 57051 + # Test that strs aren't dropping precision to 32-bit accidentally. + with tm.assert_produces_warning(FutureWarning): + res = pd.to_datetime(["1704660000"], unit="s", origin="unix") + expected = pd.to_datetime([1704660000], unit="s", origin="unix") + tm.assert_index_equal(res, expected) + def test_unit_array_mixed_nans(self, cache): values = [11111111111111111, 1, 1.0, iNaT, NaT, np.nan, "NaT", ""] result = to_datetime(values, unit="D", errors="ignore", cache=cache) -- 2.43.0