pandas-dev/pandas · error · ValueError

Unsupported type ' ' for ArrowExtensionArray

Error message

Unsupported type '{type(values)}' for ArrowExtensionArray

What it means

Raised by ArrowExtensionArray.__init__ when the `values` argument is neither a pyarrow Array nor a pyarrow ChunkedArray. The constructor only accepts those two concrete pyarrow types so it can normalize the internal _pa_array field; any other object (numpy array, list, pandas Series, pandas Index, Python scalar) is rejected even if it is 'array-like'. Use the _from_sequence classmethod or pd.array(..., dtype='...[pyarrow]') to build from non-pyarrow inputs.

Solutions

If you have a list/numpy array/Series, construct via the public path pd.array(data, dtype='<type>[pyarrow]') or cls._from_sequence(data, dtype=dtype) instead of cls(data).
If you have a pyarrow object, ensure it is pa.Array or pa.ChunkedArray: wrap with pa.array(...) for an Array, or pa.chunked_array([...]) for a ChunkedArray.
If you have a pa.RecordBatch or pa.Table, extract a column first with table.column(name) before constructing the extension array.
Verify the type at runtime: isinstance(values, (pa.Array, pa.ChunkedArray)) before calling the constructor.

Example fix

# before
import pyarrow as pa
from pandas.core.arrays.arrow import ArrowExtensionArray
arr = ArrowExtensionArray([1, 2, 3])  # raises ValueError

# after
arr = ArrowExtensionArray(pa.array([1, 2, 3]))
# or the public path:
arr = pd.array([1, 2, 3], dtype="int64[pyarrow]")

Defensive patterns

Strategy: type-guard

Validate before calling

import pyarrow as pa

def is_valid_arrow_input(values) -> bool:
    return isinstance(values, (pa.Array, pa.ChunkedArray))

# usage
if not is_valid_arrow_input(values):
    values = pa.array(values)

Type guard

import pyarrow as pa
from typing import Any

def is_pa_array_or_chunked(v: Any) -> bool:
    return isinstance(v, (pa.Array, pa.ChunkedArray))

Prevention

Construct ArrowExtensionArray via pd.array(data, dtype='...[pyarrow]') or cls._from_sequence rather than the raw constructor.
Keep a runtime isinstance check before passing values into the constructor.
Document the accepted pyarrow types at every internal call site.

When it happens

Trigger: Calling ArrowExtensionArray(values) or ArrowStringArray(values) directly with a list, numpy.ndarray, pandas Series/Index, a pyarrow Scalar/Table/RecordBatch, or any object that is not an instance of pa.Array / pa.ChunkedArray. Also triggered indirectly when a subclass or internal helper passes an unconverted Python iterable into the constructor.

Common situations: Users constructing the extension array by hand instead of going through pd.array()/Series.astype('...[pyarrow]'); deserialization code that holds a numpy buffer; passing pa.array(...) is fine but passing the raw pa.scalar(...) or a tensor type is not; migration from numpy-backed extension arrays where the constructor accepted ndarrays.

Related errors

AI-assisted analysis of pandas-dev/pandas@3b7651241d (2026-08-11). Data as JSON: /api/errors/d7e5227ab8c94dd5. Report an issue: GitHub.

Appendix: source

Thrown at pandas/core/arrays/arrow/array.py:411

    _pa_array: pa.ChunkedArray
    _dtype: ArrowDtype
    # results from calls to methods decorated with cache_readonly get added here
    _cache: dict[str, pa.ChunkedArray]

    def __init__(self, values: pa.Array | pa.ChunkedArray) -> None:
        if not HAS_PYARROW:
            msg = (
                f"pyarrow>={PYARROW_MIN_VERSION} is required for PyArrow "
                "backed ArrowExtensionArray."
            )
            raise ImportError(msg)
        if isinstance(values, pa.Array):
            self._pa_array = pa.chunked_array([values])
        elif isinstance(values, pa.ChunkedArray):
            self._pa_array = values
        else:
            raise ValueError(
                f"Unsupported type '{type(values)}' for ArrowExtensionArray"
            )
        self._dtype = ArrowDtype(self._pa_array.type)
        self._cache = {}

    @classmethod
    def _from_sequence(
        cls, scalars, *, dtype: Dtype | None = None, copy: bool = False
    ) -> Self:
        """
        Construct a new ExtensionArray from a sequence of scalars.
        """
        pa_type = to_pyarrow_type(dtype)
        pa_array = cls._box_pa_array(scalars, pa_type=pa_type, copy=copy)
        arr = cls(pa_array)
        return arr

    @classmethod

View on GitHub (pinned to 3b7651241d)