Merge pull request #3011 from lonvia/fix-flex-scripts

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

diff --git a/nominatim/tokenizer/base.py b/nominatim/tokenizer/base.py
index 5a3d3b1276aa7c6bf3bb36787967d519fe70aa79..afbd1914b35d84219812afdd64f3061d306944f5 100644 (file)
--- a/nominatim/tokenizer/base.py
+++ b/nominatim/tokenizer/base.py
@@ -5,16 +5,16 @@
 # Copyright (C) 2022 by the Nominatim developer community.
 # For a full list of authors see the git log.
 """
 # Copyright (C) 2022 by the Nominatim developer community.
 # For a full list of authors see the git log.
 """

-Abstract class defintions for tokenizers. These base classes are here
+Abstract class definitions for tokenizers. These base classes are here

 mainly for documentation purposes.
 """
 from abc import ABC, abstractmethod
 mainly for documentation purposes.
 """
 from abc import ABC, abstractmethod

-from typing import List, Tuple, Dict, Any
+from typing import List, Tuple, Dict, Any, Optional, Iterable
+from pathlib import Path

 
 from nominatim.config import Configuration
 from nominatim.data.place_info import PlaceInfo
 
 from nominatim.config import Configuration
 from nominatim.data.place_info import PlaceInfo

-
-# pylint: disable=unnecessary-pass
+from nominatim.typing import Protocol

 
 class AbstractAnalyzer(ABC):
     """ The analyzer provides the functions for analysing names and building
 
 class AbstractAnalyzer(ABC):
     """ The analyzer provides the functions for analysing names and building


@@ -80,7 +80,8 @@ class AbstractAnalyzer(ABC):
 
 
     @abstractmethod
 
 
     @abstractmethod

-    def update_special_phrases(self, phrases: List[Tuple[str, str, str, str]],
+    def update_special_phrases(self,
+                               phrases: Iterable[Tuple[str, str, str, str]],

                                should_replace: bool) -> None:
         """ Update the tokenizer's special phrase tokens from the given
             list of special phrases.
                                should_replace: bool) -> None:
         """ Update the tokenizer's special phrase tokens from the given
             list of special phrases.


@@ -112,7 +113,7 @@ class AbstractAnalyzer(ABC):
             the search index.
 
             Arguments:
             the search index.
 
             Arguments:

-                place: Place information retrived from the database.
+                place: Place information retrieved from the database.

 
             Returns:
                 A JSON-serialisable structure that will be handed into
 
             Returns:
                 A JSON-serialisable structure that will be handed into


@@ -140,7 +141,7 @@ class AbstractTokenizer(ABC):
 
               init_db: When set to False, then initialisation of database
                 tables should be skipped. This option is only required for
 
               init_db: When set to False, then initialisation of database
                 tables should be skipped. This option is only required for

-                migration purposes and can be savely ignored by custom
+                migration purposes and can be safely ignored by custom

                 tokenizers.
 
             TODO: can we move the init_db parameter somewhere else?
                 tokenizers.
 
             TODO: can we move the init_db parameter somewhere else?


@@ -186,7 +187,7 @@ class AbstractTokenizer(ABC):
 
 
     @abstractmethod
 
 
     @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.
 
         """ 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.
         """
             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.
+        """