FIX-#5972: compute correct dtype for Series.str.find/index/rfind/rindex

Signed-off-by: Anatoly Myachev <[email protected]>

modin/core/storage_formats/pandas/query_compiler.py

@@ -1804,10 +1804,10 @@ def isin_func(df, values):
     str_contains = Map.register(_str_map("contains"), dtypes=np.bool_)
     str_count = Map.register(_str_map("count"), dtypes=int)
     str_endswith = Map.register(_str_map("endswith"), dtypes=np.bool_)
-    str_find = Map.register(_str_map("find"), dtypes="copy")
+    str_find = Map.register(_str_map("find"), dtypes=np.int64)
     str_findall = Map.register(_str_map("findall"), dtypes="copy")
     str_get = Map.register(_str_map("get"), dtypes="copy")
-    str_index = Map.register(_str_map("index"), dtypes="copy")
+    str_index = Map.register(_str_map("index"), dtypes=np.int64)
     str_isalnum = Map.register(_str_map("isalnum"), dtypes=np.bool_)
     str_isalpha = Map.register(_str_map("isalpha"), dtypes=np.bool_)
     str_isdecimal = Map.register(_str_map("isdecimal"), dtypes=np.bool_)
@@ -1847,8 +1847,8 @@ def str_extract(self, pat, flags, expand):
         return qc
 
     str_replace = Map.register(_str_map("replace"), dtypes="copy", shape_hint="column")
-    str_rfind = Map.register(_str_map("rfind"), dtypes="copy", shape_hint="column")
-    str_rindex = Map.register(_str_map("rindex"), dtypes="copy", shape_hint="column")
+    str_rfind = Map.register(_str_map("rfind"), dtypes=np.int64, shape_hint="column")
+    str_rindex = Map.register(_str_map("rindex"), dtypes=np.int64, shape_hint="column")
     str_rjust = Map.register(_str_map("rjust"), dtypes="copy", shape_hint="column")
     _str_rpartition = Map.register(
         _str_map("rpartition"), dtypes="copy", shape_hint="column"

modin/pandas/test/test_series.py

@@ -4338,8 +4338,6 @@ def test_str_find(data, sub, start, end):
         modin_series,
         pandas_series,
         lambda series: series.str.find(sub, start=start, end=end),
-        # https://github.com/modin-project/modin/issues/5972
-        comparator_kwargs={"check_dtypes": False},
     )
 
 
@@ -4353,8 +4351,6 @@ def test_str_rfind(data, sub, start, end):
         modin_series,
         pandas_series,
         lambda series: series.str.rfind(sub, start=start, end=end),
-        # https://github.com/modin-project/modin/issues/5972
-        comparator_kwargs={"check_dtypes": False},
     )
 
 
@@ -4368,8 +4364,6 @@ def test_str_index(data, sub, start, end):
         modin_series,
         pandas_series,
         lambda series: series.str.index(sub, start=start, end=end),
-        # https://github.com/modin-project/modin/issues/5972
-        comparator_kwargs={"check_dtypes": False},
     )
 
 
@@ -4383,8 +4377,6 @@ def test_str_rindex(data, sub, start, end):
         modin_series,
         pandas_series,
         lambda series: series.str.rindex(sub, start=start, end=end),
-        # https://github.com/modin-project/modin/issues/5972
-        comparator_kwargs={"check_dtypes": False},
     )