pandas-dev/pandas · error · ValueError

does not have a resolution.

Error message

{dtype=} does not have a resolution.

What it means

Raised by dtype_to_unit when given an ArrowDtype whose kind is not 'm' (timedelta) or 'M' (datetime). The function exists to extract a time resolution string; for arrow types that have no temporal resolution (e.g. pyarrow strings, ints, bools) the concept is undefined, so it refuses rather than returning a nonsense unit.

Solutions

1. Guard the call: only invoke dtype_to_unit when dtype.kind in 'mM'.
2. Filter to temporal columns before resolving units: [c for c in df.columns if df[c].dtype.kind in 'mM'].
3. If you intended a datetime, construct the right dtype: pd.ArrowDtype(pa.timestamp('us')).

Example fix

// before
unit = dtype_to_unit(pd.ArrowDtype(pa.string()))  # ValueError: does not have a resolution

// after
if dtype.kind in 'mM':
    unit = dtype_to_unit(dtype)
else:
    unit = None

def safe_dtype_to_unit(dtype):
    if isinstance(dtype, pd.ArrowDtype) and dtype.kind not in "mM":
        return None  # or raise a descriptive error
    return dtype_to_unit(dtype)

def has_resolution(dtype) -> bool:
    if isinstance(dtype, pd.ArrowDtype):
        return dtype.kind in "mM"
    return True

try:
    unit = dtype_to_unit(dtype)
except ValueError as e:
    if "does not have a resolution" in str(e):
        unit = None
    else:
        raise

When it happens

Trigger: Calling internal dtype_to_unit(dtype) with an ArrowExtensionArray dtype that is not a pyarrow timestamp/duration; indirectly through code that converts a mixed list of dtypes to units assuming all are temporal.

Common situations: Interoperability code that iterates DataFrame dtypes and assumes every Arrow-backed column has a time resolution; passing an ArrowDtype accidentally where a DatetimeTZDtype/np.dtype was expected.

Related errors

• dtype must be PeriodDtype
• Expected array of type, got instead
• Supported timedelta64 resolutions are 's', 'ms', 'us', 'ns'
• Supported units are 's', 'ms', 'us', 'ns'
• 'unit' must be one of 's', 'ms', 'us', 'ns'

AI-assisted analysis of pandas-dev/pandas@3b7651241d (2026-08-11). Data as JSON: /api/errors/df26de8b04595a7b. Report an issue: GitHub.

Appendix: source

Thrown at pandas/core/arrays/datetimelike.py:2543

def dtype_to_unit(dtype: DatetimeTZDtype | np.dtype | ArrowDtype) -> str:
    """
    Return the unit str corresponding to the dtype's resolution.

    Parameters
    ----------
    dtype : DatetimeTZDtype or np.dtype
        If np.dtype, we assume it is a datetime64 dtype.

    Returns
    -------
    str
    """
    if isinstance(dtype, DatetimeTZDtype):
        return dtype.unit
    elif isinstance(dtype, ArrowDtype):
        if dtype.kind not in "mM":
            raise ValueError(f"{dtype=} does not have a resolution.")
        return dtype.pyarrow_dtype.unit
    return np.datetime_data(dtype)[0]

View on GitHub (pinned to 3b7651241d)