BUG: fix empty suffix and prefix handling in pyarrow string methods (pandas-dev#63395)

Co-authored-by: Joris Van den Bossche <[email protected]>

Author: vkverma9534

pandas/core/arrays/_arrow_string_mixins.py

@@ -203,12 +203,16 @@ def _str_swapcase(self) -> Self:
         return self._from_pyarrow_array(pc.utf8_swapcase(self._pa_array))
 
     def _str_removeprefix(self, prefix: str):
+        if prefix == "":
+            return self._from_pyarrow_array(self._pa_array)
         starts_with = pc.starts_with(self._pa_array, pattern=prefix)
         removed = pc.utf8_slice_codeunits(self._pa_array, len(prefix))
         result = pc.if_else(starts_with, removed, self._pa_array)
         return self._from_pyarrow_array(result)
 
     def _str_removesuffix(self, suffix: str):
+        if suffix == "":
+            return self._from_pyarrow_array(self._pa_array)
         ends_with = pc.ends_with(self._pa_array, pattern=suffix)
         removed = pc.utf8_slice_codeunits(self._pa_array, 0, stop=-len(suffix))
         result = pc.if_else(ends_with, removed, self._pa_array)

pandas/tests/strings/test_strings.py

@@ -544,7 +544,12 @@ def test_strip_lstrip_rstrip_args(any_string_dtype, method, exp):
 
 
 @pytest.mark.parametrize(
-    "prefix, expected", [("a", ["b", " b c", "bc"]), ("ab", ["", "a b c", "bc"])]
+    "prefix, expected",
+    [
+        ("a", ["b", " b c", "bc"]),
+        ("ab", ["", "a b c", "bc"]),
+        ("", ["ab", "a b c", "bc"]),
+    ],
 )
 def test_removeprefix(any_string_dtype, prefix, expected):
     ser = Series(["ab", "a b c", "bc"], dtype=any_string_dtype)
@@ -554,7 +559,12 @@ def test_removeprefix(any_string_dtype, prefix, expected):
 
 
 @pytest.mark.parametrize(
-    "suffix, expected", [("c", ["ab", "a b ", "b"]), ("bc", ["ab", "a b c", ""])]
+    "suffix, expected",
+    [
+        ("c", ["ab", "a b ", "b"]),
+        ("bc", ["ab", "a b c", ""]),
+        ("", ["ab", "a b c", "bc"]),
+    ],
 )
 def test_removesuffix(any_string_dtype, suffix, expected):
     ser = Series(["ab", "a b c", "bc"], dtype=any_string_dtype)