BUG: Fix dt64[non_nano] + some_offsets incorrectly rounding #62383
Conversation
pandas/core/arrays/datetimes.py
Outdated
|
|
||
| else: | ||
| result = type(self)._simple_new(res_values, dtype=res_values.dtype) | ||
| units = ["ns", "us", "ms", "s", "m", "h", "D"] |
There was a problem hiding this comment.
No, I'm wrong I will update this .
pandas/core/arrays/datetimes.py
Outdated
| idx_self = units.index(self.unit) | ||
| idx_offset = units.index(offset_unit) | ||
| res_unit = units[min(idx_self, idx_offset)] | ||
| dtype = tz_to_dtype(self.tz, unit=res_unit) |
There was a problem hiding this comment.
wont you need to cast res_values before calling simple_new?
There was a problem hiding this comment.
yes, thanks! . Should we wrap this in try/except block?
There was a problem hiding this comment.
i suspect that The Right Way to do this would be to find the correct dtype before calling _from_sequence on L817 and pass it there
pandas/core/arrays/datetimes.py
Outdated
| result = type(self)._simple_new(res_values, dtype=dtype) | ||
| offset_td = Timedelta(offset.offset) | ||
| offset_unit = offset_td.unit | ||
| if self.unit in units and offset_unit in units: |
There was a problem hiding this comment.
Then the check is not necessary
pandas/core/arrays/datetimes.py
Outdated
| self.tz is not None | ||
| and getattr(result.dtype, "tz", None) is None | ||
| and res_unit == "ns" | ||
| ): |
There was a problem hiding this comment.
why does this codnition need tom change?
There was a problem hiding this comment.
it prevents errors by making sure we only localize when it's valid, please correct me
There was a problem hiding this comment.
Well result.dtype.tz is always None, so that part of the check is unnecessary. And I suspect the u it part is just wrong.
pandas/core/arrays/datetimes.py
Outdated
| idx_self = units.index(self.unit) | ||
| idx_offset = units.index(offset_unit) | ||
| res_unit = units[min(idx_self, idx_offset)] | ||
| dtype_naive = np.dtype(f"datetime64[{res_unit}]") |
There was a problem hiding this comment.
instead of defining dtype_naive and doing astype, could you do .as_unit(res_unit) after simple_new?
2 participants
This PR updates logic selects the finest unit between the array and the offset, preventing rounding errors. A regression test is included to confirm correct behavior for affected offsets and units.
Please let me know if my approach or fix needs any improvements . I’m open to feedback and happy to make changes based on suggestions.
Thankyou !