fix empty string

Author: Rohan Jain

pandas/core/arrays/arrow/array.py

@@ -2331,20 +2331,32 @@ def _str_find(self, sub: str, start: int = 0, end: int | None = None) -> Self:
         if (start == 0 or start is None) and end is None:
             result = pc.find_substring(self._pa_array, sub)
         else:
+            length = pc.utf8_length(self._pa_array)
             if start is None:
                 start_offset = 0
                 start = 0
             elif start < 0:
-                start_offset = pc.add(start, pc.utf8_length(self._pa_array))
+                start_offset = pc.add(start, length)
                 start_offset = pc.if_else(pc.less(start_offset, 0), 0, start_offset)
             else:
                 start_offset = start
             slices = pc.utf8_slice_codeunits(self._pa_array, start, stop=end)
             result = pc.find_substring(slices, sub)
-            not_found = pc.equal(result, pa.scalar(-1, type=result.type))
-
+            found = pc.not_equal(result, pa.scalar(-1, type=result.type))
+            if end is None:
+                end = length
+            elif end < 0:
+                end = pc.add(end, length)
+                end = pc.if_else(pc.less(end, 0), 0, end)
+            found = pc.and_(
+                found,
+                pc.and_(
+                    pc.less_equal(start_offset, end),
+                    pc.less_equal(start_offset, length),
+                ),
+            )
             offset_result = pc.add(result, start_offset)
-            result = pc.if_else(not_found, result, offset_result)
+            result = pc.if_else(found, offset_result, -1)
         return type(self)(result)
 
     def _str_join(self, sep: str) -> Self:

pandas/tests/extension/test_arrow.py

@@ -23,6 +23,7 @@
     BytesIO,
     StringIO,
 )
+from itertools import combinations
 import operator
 import pickle
 import re
@@ -1975,6 +1976,27 @@ def test_str_find_large_start():
         tm.assert_series_equal(result, expected)
 
 
+def _get_all_substrings(string):
+    length = len(string) + 1
+    return [string[x:y] for x, y in combinations(range(length), r=2)]
+
+
+@pytest.mark.xfail(
+    pa_version_under13p0, reason="https://github.com/apache/arrow/issues/36311"
+)
+def test_str_find_e2e():
+    string = "abcdefgh"
+    s = pd.Series([string], dtype=ArrowDtype(pa.string()))
+    substrings = _get_all_substrings(string) + ["", "az", "abce"]
+    offsets = list(range(-15, 15)) + [None]
+    for start in offsets:
+        for end in offsets:
+            for sub in substrings:
+                result = s.str.find(sub, start, end)
+                expected = pd.Series([string.find(sub, start, end)], dtype=result.dtype)
+                tm.assert_series_equal(result, expected)
+
+
 def test_str_find_negative_start_negative_end_no_match():
     # GH 56791
     ser = pd.Series(["abcdefg", None], dtype=ArrowDtype(pa.string()))