Solved the TC Issue raised

Author: raunakpratapkushwaha-ai

pandas/core/arrays/arrow/array.py

@@ -1810,7 +1810,7 @@ def searchsorted(
             Value(s) to insert into `self`.
         side : {'left', 'right'}, optional
             If 'left', the index of the first suitable location found is given.
-            If 'right', return the last such index.  If there is no suitable
+            If 'right', return the last such index. If there is no suitable
             index, return either 0 or N (where N is the length of `self`).
         sorter : 1-D array-like, optional
             Optional array of integer indices that sort array a into ascending
@@ -1837,19 +1837,95 @@ def searchsorted(
                 "searchsorted requires array to be sorted, which is impossible "
                 "with NAs present."
             )
-        if isinstance(value, ExtensionArray):
-            value = value.astype(object)
-        # Base class searchsorted would cast to object, which is *much* slower.
+
+        if sorter is not None:
+            dtype = None
+
+            if isinstance(self.dtype, ArrowDtype):
+                pa_dtype = self.dtype.pyarrow_dtype
+
+                if (
+                    pa.types.is_timestamp(pa_dtype) or pa.types.is_duration(pa_dtype)
+                ) and pa_dtype.unit == "ns":
+                    dtype = object
+
+            return self.to_numpy(dtype=dtype).searchsorted(
+                value,
+                side=side,
+                sorter=sorter,
+            )
+
+        arr = self._pa_array.combine_chunks()
+        pa_dtype = arr.type
+
+        # Fast Arrow-native path for strings
+        if pa.types.is_string(pa_dtype) or pa.types.is_large_string(pa_dtype):
+            offsets_buf = arr.buffers()[1]
+            data_buf = arr.buffers()[2]
+
+            offset_dtype = np.int64 if pa.types.is_large_string(pa_dtype) else np.int32
+
+            offsets = np.frombuffer(offsets_buf, dtype=offset_dtype)
+            data = memoryview(data_buf)
+
+            def get_string(i: int) -> bytes:
+                start = offsets[i]
+                end = offsets[i + 1]
+                return data[start:end].tobytes()
+
+            def binary_search(target, side_local):
+                if isinstance(target, str):
+                    target = target.encode()
+                elif not isinstance(target, bytes):
+                    target = str(target).encode()
+
+                left = 0
+                right = len(arr)
+
+                while left < right:
+                    mid = (left + right) // 2
+                    mid_value = get_string(mid)
+
+                    if side_local == "left":
+                        if mid_value < target:
+                            left = mid + 1
+                        else:
+                            right = mid
+                    elif mid_value <= target:
+                        left = mid + 1
+                    else:
+                        right = mid
+
+                return left
+
+            # scalar input
+            if is_scalar(value):
+                return np.intp(binary_search(value, side))
+
+            # vector input
+            result = np.empty(len(value), dtype=np.intp)
+
+            for i, val in enumerate(value):
+                result[i] = binary_search(val, side)
+
+            return result
+
+        # Fallback for non-string dtypes
         dtype = None
+
         if isinstance(self.dtype, ArrowDtype):
             pa_dtype = self.dtype.pyarrow_dtype
+
             if (
                 pa.types.is_timestamp(pa_dtype) or pa.types.is_duration(pa_dtype)
             ) and pa_dtype.unit == "ns":
-                # np.array[datetime/timedelta].searchsorted(datetime/timedelta)
-                # erroneously fails when numpy type resolution is nanoseconds
                 dtype = object
-        return self.to_numpy(dtype=dtype).searchsorted(value, side=side, sorter=sorter)
+
+        return self.to_numpy(dtype=dtype).searchsorted(
+            value,
+            side=side,
+            sorter=sorter,
+        )
 
     def take(
         self,