ENH: Series.explode to support pyarrow-backed list types

doc/source/whatsnew/v2.1.0.rst

@@ -102,6 +102,7 @@ Other enhancements
 - :meth:`DataFrame.stack` gained the ``sort`` keyword to dictate whether the resulting :class:`MultiIndex` levels are sorted (:issue:`15105`)
 - :meth:`DataFrame.unstack` gained the ``sort`` keyword to dictate whether the resulting :class:`MultiIndex` levels are sorted (:issue:`15105`)
 - :meth:`DataFrameGroupby.agg` and :meth:`DataFrameGroupby.transform` now support grouping by multiple keys when the index is not a :class:`MultiIndex` for ``engine="numba"`` (:issue:`53486`)
+- :meth:`Series.explode` now supports pyarrow-backed list types (:issue:`53602`)
 - :meth:`SeriesGroupby.agg` and :meth:`DataFrameGroupby.agg` now support passing in multiple functions for ``engine="numba"`` (:issue:`53486`)
 - Added ``engine_kwargs`` parameter to :meth:`DataFrame.to_excel` (:issue:`53220`)
 - Added a new parameter ``by_row`` to :meth:`Series.apply`. When set to ``False`` the supplied callables will always operate on the whole Series (:issue:`53400`).

pandas/core/arrays/arrow/array.py

@@ -347,9 +347,9 @@ def _box_pa(
         -------
         pa.Array or pa.ChunkedArray or pa.Scalar
         """
-        if is_list_like(value):
-            return cls._box_pa_array(value, pa_type)
-        return cls._box_pa_scalar(value, pa_type)
+        if isinstance(value, pa.Scalar) or not is_list_like(value):
+            return cls._box_pa_scalar(value, pa_type)
+        return cls._box_pa_array(value, pa_type)
 
     @classmethod
     def _box_pa_scalar(cls, value, pa_type: pa.DataType | None = None) -> pa.Scalar:
@@ -1549,6 +1549,24 @@ def _reduce(self, name: str, *, skipna: bool = True, **kwargs):
 
         return result.as_py()
 
+    def _explode(self):
+        """
+        See Series.explode.__doc__.
+        """
+        values = self
+        counts = pa.compute.list_value_length(values._pa_array)
+        counts = counts.fill_null(1).to_numpy()
+        fill_value = pa.scalar([None], type=self._pa_array.type)
+        mask = counts == 0
+        if mask.any():
+            values = values.copy()
+            values[mask] = fill_value
+            counts = counts.copy()
+            counts[mask] = 1
+        values = values.fillna(fill_value)
+        values = type(self)(pa.compute.list_flatten(values._pa_array))
+        return values, counts
+
     def __setitem__(self, key, value) -> None:
         """Set one or more values inplace.
 
@@ -1591,10 +1609,10 @@ def __setitem__(self, key, value) -> None:
                 raise IndexError(
                     f"index {key} is out of bounds for axis 0 with size {n}"
                 )
-            if is_list_like(value):
-                raise ValueError("Length of indexer and values mismatch")
-            elif isinstance(value, pa.Scalar):
+            if isinstance(value, pa.Scalar):
                 value = value.as_py()
+            elif is_list_like(value):
+                raise ValueError("Length of indexer and values mismatch")
             chunks = [
                 *self._pa_array[:key].chunks,
                 pa.array([value], type=self._pa_array.type, from_pandas=True),

pandas/core/series.py

@@ -72,7 +72,10 @@
     pandas_dtype,
     validate_all_hashable,
 )
-from pandas.core.dtypes.dtypes import ExtensionDtype
+from pandas.core.dtypes.dtypes import (
+    ArrowDtype,
+    ExtensionDtype,
+)
 from pandas.core.dtypes.generic import ABCDataFrame
 from pandas.core.dtypes.inference import is_hashable
 from pandas.core.dtypes.missing import (
@@ -4267,12 +4270,14 @@ def explode(self, ignore_index: bool = False) -> Series:
         3      4
         dtype: object
         """
-        if not len(self) or not is_object_dtype(self.dtype):
+        if isinstance(self.dtype, ArrowDtype) and self.dtype.type == list:
+            values, counts = self._values._explode()
+        elif len(self) and is_object_dtype(self.dtype):
+            values, counts = reshape.explode(np.asarray(self._values))
+        else:
             result = self.copy()
             return result.reset_index(drop=True) if ignore_index else result
 
-        values, counts = reshape.explode(np.asarray(self._values))
-
         if ignore_index:
             index = default_index(len(values))
         else:

pandas/tests/series/methods/test_explode.py

@@ -1,6 +1,8 @@
 import numpy as np
 import pytest
 
+from pandas.compat import pa_version_under7p0
+
 import pandas as pd
 import pandas._testing as tm
 
@@ -141,3 +143,25 @@ def test_explode_scalars_can_ignore_index():
     result = s.explode(ignore_index=True)
     expected = pd.Series([1, 2, 3])
     tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.skipif(pa_version_under7p0, reason="minimum pyarrow not installed")
+def test_explode_pyarrow_list_type():
+    # GH 53602
+    import pyarrow as pa
+
+    data = [
+        [None, None],
+        [1],
+        [],
+        [2, 3],
+        None,
+    ]
+    ser = pd.Series(data, dtype=pd.ArrowDtype(pa.list_(pa.int64())))
+    result = ser.explode()
+    expected = pd.Series(
+        data=[None, None, 1, None, 2, 3, None],
+        index=[0, 0, 1, 2, 3, 3, 4],
+        dtype=pd.ArrowDtype(pa.int64()),
+    )
+    tm.assert_series_equal(result, expected)