Backport PR #56766 on branch 2.2.x (BUG: IntervalIndex.from_tuples raising with masked subtype) (#56785)

Backport PR #56766: BUG: IntervalIndex.from_tuples raising with masked subtype

Co-authored-by: Patrick Hoefler <[email protected]>

Author: meeseeksmachine

doc/source/whatsnew/v2.2.0.rst

@@ -817,6 +817,7 @@ Interval
 - Bug in :class:`Interval` ``__repr__`` not displaying UTC offsets for :class:`Timestamp` bounds. Additionally the hour, minute and second components will now be shown (:issue:`55015`)
 - Bug in :meth:`IntervalIndex.factorize` and :meth:`Series.factorize` with :class:`IntervalDtype` with datetime64 or timedelta64 intervals not preserving non-nanosecond units (:issue:`56099`)
 - Bug in :meth:`IntervalIndex.from_arrays` when passed ``datetime64`` or ``timedelta64`` arrays with mismatched resolutions constructing an invalid ``IntervalArray`` object (:issue:`55714`)
+- Bug in :meth:`IntervalIndex.from_tuples` raising if subtype is a nullable extension dtype (:issue:`56765`)
 - Bug in :meth:`IntervalIndex.get_indexer` with datetime or timedelta intervals incorrectly matching on integer targets (:issue:`47772`)
 - Bug in :meth:`IntervalIndex.get_indexer` with timezone-aware datetime intervals incorrectly matching on a sequence of timezone-naive targets (:issue:`47772`)
 - Bug in setting values on a :class:`Series` with an :class:`IntervalIndex` using a slice incorrectly raising (:issue:`54722`)

pandas/core/arrays/interval.py

@@ -79,6 +79,7 @@
     unique,
     value_counts_internal as value_counts,
 )
+from pandas.core.arrays import ArrowExtensionArray
 from pandas.core.arrays.base import (
     ExtensionArray,
     _extension_array_shared_docs,
@@ -370,11 +371,18 @@ def _ensure_simple_new_inputs(
         right = ensure_wrapped_if_datetimelike(right)
         right = extract_array(right, extract_numpy=True)
 
-        lbase = getattr(left, "_ndarray", left).base
-        rbase = getattr(right, "_ndarray", right).base
-        if lbase is not None and lbase is rbase:
-            # If these share data, then setitem could corrupt our IA
-            right = right.copy()
+        if isinstance(left, ArrowExtensionArray) or isinstance(
+            right, ArrowExtensionArray
+        ):
+            pass
+        else:
+            lbase = getattr(left, "_ndarray", left)
+            lbase = getattr(lbase, "_data", lbase).base
+            rbase = getattr(right, "_ndarray", right)
+            rbase = getattr(rbase, "_data", rbase).base
+            if lbase is not None and lbase is rbase:
+                # If these share data, then setitem could corrupt our IA
+                right = right.copy()
 
         dtype = IntervalDtype(left.dtype, closed=closed)
 

pandas/tests/indexes/interval/test_constructors.py

@@ -3,6 +3,8 @@
 import numpy as np
 import pytest
 
+import pandas.util._test_decorators as td
+
 from pandas.core.dtypes.common import is_unsigned_integer_dtype
 from pandas.core.dtypes.dtypes import IntervalDtype
 
@@ -517,3 +519,17 @@ def test_dtype_closed_mismatch():
 
     with pytest.raises(ValueError, match=msg):
         IntervalArray([], dtype=dtype, closed="neither")
+
+
+@pytest.mark.parametrize(
+    "dtype",
+    ["Float64", pytest.param("float64[pyarrow]", marks=td.skip_if_no("pyarrow"))],
+)
+def test_ea_dtype(dtype):
+    # GH#56765
+    bins = [(0.0, 0.4), (0.4, 0.6)]
+    interval_dtype = IntervalDtype(subtype=dtype, closed="left")
+    result = IntervalIndex.from_tuples(bins, closed="left", dtype=interval_dtype)
+    assert result.dtype == interval_dtype
+    expected = IntervalIndex.from_tuples(bins, closed="left").astype(interval_dtype)
+    tm.assert_index_equal(result, expected)