BUG: Implement elementwise IntervalArray.overlaps (#62004)

Author: Aniketsy

doc/source/whatsnew/v2.3.2.rst

@@ -1423,9 +1423,20 @@ def mid(self) -> Index:
     )
     def overlaps(self, other):
         if isinstance(other, (IntervalArray, ABCIntervalIndex)):
-            raise NotImplementedError
+            if not isinstance(other, IntervalArray):
+                other = IntervalArray(other)
+            if len(self) != len(other):
+                raise ValueError("Both IntervalArrays must have the same length.")
+            if self.closed != other.closed:
+                raise ValueError(
+                    "Both IntervalArrays must have the same 'closed' property."
+                )
+
+            op1 = le if (self.closed_left and other.closed_right) else lt
+            op2 = le if (other.closed_left and self.closed_right) else lt
+            return op1(self.left, other.right) & op2(other.left, self.right)
         if not isinstance(other, Interval):
-            msg = f"`other` must be Interval-like, got {type(other).__name__}"
+            msg = f" `other` must be Interval-like, got {type(other).__name__}"
             raise TypeError(msg)
 
         # equality is okay if both endpoints are closed (overlap at a point)

pandas/core/arrays/interval.py

@@ -346,8 +346,6 @@ def _str_contains(
     ):
         if flags:
             return super()._str_contains(pat, case, flags, na, regex)
-        if isinstance(pat, re.Pattern):
-            pat = pat.pattern
 
         return ArrowStringArrayMixin._str_contains(self, pat, case, flags, na, regex)
 

pandas/core/arrays/string_arrow.py

@@ -229,3 +229,34 @@ def test_min_max(self, left_right_dtypes, index_or_series_or_array):
             res = arr_na.max(**kws)
             assert res == MAX
             assert type(res) == type(MAX)
+
+    def test_intervalarray_overlaps_all_cases(self):
+        # Basic self-overlap
+        data = [(0, 1), (1, 3), (2, 4)]
+        arr = IntervalArray.from_tuples(data)
+        result = arr.overlaps(arr)
+        expected = np.array([True, True, True])
+        tm.testing.assert_array_equal(result, expected)
+
+        # Overlap with different intervals
+        arr2 = IntervalArray.from_tuples([(0.5, 1.5), (2, 2.5), (3, 5)])
+        result2 = arr.overlaps(arr2)
+        expected2 = np.array([True, False, True])
+        tm.testing.assert_array_equal(result2, expected2)
+
+        # Length mismatch
+        arr_short = IntervalArray.from_tuples([(0, 1)])
+        with pytest.raises(ValueError, match="same length"):
+            arr.overlaps(arr_short)
+
+        # Closed property mismatch
+        arr_left = IntervalArray.from_tuples(data, closed="left")
+        arr_right = IntervalArray.from_tuples(data, closed="right")
+        with pytest.raises(ValueError, match="same 'closed' property"):
+            arr_left.overlaps(arr_right)
+
+        # Overlap with scalar interval
+        interval = Interval(1, 2)
+        result3 = arr.overlaps(interval)
+        expected3 = np.array([False, True, False])
+        tm.testing.assert_array_equal(result3, expected3)

pandas/tests/arrays/interval/test_interval.py

@@ -281,21 +281,6 @@ def test_contains_nan(any_string_dtype):
     tm.assert_series_equal(result, expected)
 
 
-def test_str_contains_compiled_regex_arrow_dtype(any_string_dtype):
-    # GH#61942
-    ser = Series(["foo", "bar", "baz"], dtype=any_string_dtype)
-    pat = re.compile("ba.")
-    result = ser.str.contains(pat)
-    # Determine expected dtype and values
-    expected_dtype = {
-        "string[pyarrow]": "bool[pyarrow]",
-        "string": "boolean",
-        "str": bool,
-    }.get(any_string_dtype, object)
-    expected = Series([False, True, True], dtype=expected_dtype)
-    tm.assert_series_equal(result, expected)
-
-
 # --------------------------------------------------------------------------------------
 # str.startswith
 # --------------------------------------------------------------------------------------