reintroduce cutoffs when searching for very frequent words

[nominatim.git] / nominatim / tokenizer / sanitizers / split_name_list.py

diff --git a/nominatim/tokenizer/sanitizers/split_name_list.py b/nominatim/tokenizer/sanitizers/split_name_list.py
index 13921c3e104385f27489b4404034e523aa389ef3..7d0667b4e323ce1aa060c3f0ae2738505173a476 100644 (file)
--- a/nominatim/tokenizer/sanitizers/split_name_list.py
+++ b/nominatim/tokenizer/sanitizers/split_name_list.py
@@ -11,15 +11,18 @@ Arguments:
     delimiters: Define the set of characters to be used for
                 splitting the list. (default: ',;')
 """
-from nominatim.tokenizer.sanitizers.helpers import create_split_regex
+from typing import Callable
 
-def create(func):
+from nominatim.tokenizer.sanitizers.base import ProcessInfo
+from nominatim.tokenizer.sanitizers.config import SanitizerConfig
+
+def create(config: SanitizerConfig) -> Callable[[ProcessInfo], None]:
     """ Create a name processing function that splits name values with
         multiple values into their components.
     """
-    regexp = create_split_regex(func)
+    regexp = config.get_delimiter()
 
-    def _process(obj):
+    def _process(obj: ProcessInfo) -> None:
         if not obj.names:
             return