X-Git-Url: https://git.openstreetmap.org./nominatim.git/blobdiff_plain/bb175cc95898de420781867973b38d033c187e81..f92e580293a527d167ddb8b12a2690967c19b26b:/nominatim/tokenizer/base.py?ds=inline

diff --git a/nominatim/tokenizer/base.py b/nominatim/tokenizer/base.py
index c88c39f0..12c826eb 100644
--- a/nominatim/tokenizer/base.py
+++ b/nominatim/tokenizer/base.py
@@ -1,14 +1,21 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# This file is part of Nominatim. (https://nominatim.org)
+#
+# 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
-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.indexer.place_info import PlaceInfo
-
-# pylint: disable=unnecessary-pass
+from nominatim.db.connection import Connection
+from nominatim.data.place_info import PlaceInfo
+from nominatim.typing import Protocol
 
 class AbstractAnalyzer(ABC):
     """ The analyzer provides the functions for analysing names and building
@@ -22,7 +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()
 
 
@@ -46,8 +53,8 @@ class AbstractAnalyzer(ABC):
 
             Returns:
                 The function returns the list of all tuples that could be
-                found for the given words. Each list entry is a tuple of
-                (original word, word token, word id).
+                    found for the given words. Each list entry is a tuple of
+                    (original word, word token, word id).
         """
 
 
@@ -74,7 +81,8 @@ class AbstractAnalyzer(ABC):
 
 
     @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.
@@ -89,7 +97,7 @@ 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:
@@ -106,11 +114,11 @@ class AbstractAnalyzer(ABC):
             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
-                the database via the `token_info` field.
+                    the database via the `token_info` field.
         """
 
 
@@ -134,10 +142,8 @@ class AbstractTokenizer(ABC):
 
               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?
         """
 
 
@@ -180,7 +186,7 @@ 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.
 
@@ -189,13 +195,13 @@ class AbstractTokenizer(ABC):
 
             Returns:
               If an issue was found, return an error message with the
-              description of the issue as well as hints for the user on
-              how to resolve the issue. If everything is okay, return `None`.
+                  description of the issue as well as hints for the user on
+                  how to resolve the issue. If everything is okay, return `None`.
         """
 
 
     @abstractmethod
-    def update_statistics(self) -> None:
+    def update_statistics(self, config: Configuration, threads: int = 1) -> None:
         """ Recompute any tokenizer statistics necessary for efficient lookup.
             This function is meant to be called from time to time by the user
             to improve performance. However, the tokenizer must not depend on
@@ -203,6 +209,13 @@ class AbstractTokenizer(ABC):
         """
 
 
+    @abstractmethod
+    def update_word_tokens(self) -> None:
+        """ Do house-keeping on the tokenizers internal data structures.
+            Remove unused word tokens, resort data etc.
+        """
+
+
     @abstractmethod
     def name_analyzer(self) -> AbstractAnalyzer:
         """ Create a new analyzer for tokenizing names and queries
@@ -217,3 +230,24 @@ class AbstractTokenizer(ABC):
             When used outside the with construct, the caller must ensure to
             call the close() function before destructing the analyzer.
         """
+
+
+    @abstractmethod
+    def most_frequent_words(self, conn: Connection, num: int) -> List[str]:
+        """ Return a list of the most frequent full words in the database.
+
+            Arguments:
+              conn: Open connection to the database which may be used to
+                    retrieve the words.
+              num: Maximum number of words to return.
+        """
+
+
+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.
+        """