add type annotations for legacy tokenizer

[nominatim.git] / nominatim / tokenizer / base.py

diff --git a/nominatim/tokenizer/base.py b/nominatim/tokenizer/base.py
index 70a54bfdc28141e62e0e5e63e5e52342044b5a14..afcb0864214eb6a2c7e2c6378b0dca0c7843c4e3 100644 (file)
--- a/nominatim/tokenizer/base.py
+++ b/nominatim/tokenizer/base.py
@@ -9,13 +9,14 @@ Abstract class defintions for tokenizers. These base classes are here
 mainly for documentation purposes.
 """
 from abc import ABC, abstractmethod
-from typing import List, Tuple, Dict, Any
+from typing import List, Tuple, Dict, Any, Optional
+from pathlib import Path
+
+from typing_extensions import Protocol
 
 from nominatim.config import Configuration
 from nominatim.data.place_info import PlaceInfo
 
-# pylint: disable=unnecessary-pass
-
 class AbstractAnalyzer(ABC):
     """ The analyzer provides the functions for analysing names and building
         the token database.
@@ -28,7 +29,7 @@ class AbstractAnalyzer(ABC):
         return self
 
 
-    def __exit__(self, exc_type, exc_value, traceback) -> None:
+    def __exit__(self, exc_type: Any, exc_value: Any, traceback: Any) -> None:
         self.close()
 
 
@@ -95,7 +96,7 @@ class AbstractAnalyzer(ABC):
 
 
     @abstractmethod
-    def add_country_names(self, country_code: str, names: Dict[str, str]):
+    def add_country_names(self, country_code: str, names: Dict[str, str]) -> None:
         """ Add the given names to the tokenizer's list of country tokens.
 
             Arguments:
@@ -186,7 +187,7 @@ class AbstractTokenizer(ABC):
 
 
     @abstractmethod
-    def check_database(self, config: Configuration) -> str:
+    def check_database(self, config: Configuration) -> Optional[str]:
         """ Check that the database is set up correctly and ready for being
             queried.
 
@@ -230,3 +231,13 @@ class AbstractTokenizer(ABC):
             When used outside the with construct, the caller must ensure to
             call the close() function before destructing the analyzer.
         """
+
+
+class TokenizerModule(Protocol):
+    """ Interface that must be exported by modules that implement their
+        own tokenizer.
+    """
+
+    def create(self, dsn: str, data_dir: Path) -> AbstractTokenizer:
+        """ Factory for new tokenizers.
+        """