Merge remote-tracking branch 'upstream/master'

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

diff --git a/nominatim/tokenizer/base.py b/nominatim/tokenizer/base.py
index 00ecae447c5843eb0fc772960a421e271a82cb1d..f81b3bc262ed9f690b0eb6bc794e2cab076de822 100644 (file)
--- a/nominatim/tokenizer/base.py
+++ b/nominatim/tokenizer/base.py
@@ -1,3 +1,9 @@
+# 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
 mainly for documentation purposes.
 """
 Abstract class defintions for tokenizers. These base classes are here
 mainly for documentation purposes.


@@ -6,6 +12,7 @@ from abc import ABC, abstractmethod
 from typing import List, Tuple, Dict, Any
 
 from nominatim.config import Configuration
 from typing import List, Tuple, Dict, Any
 
 from nominatim.config import Configuration

+from nominatim.indexer.place_info import PlaceInfo

 
 # pylint: disable=unnecessary-pass
 
 
 # pylint: disable=unnecessary-pass
 


@@ -29,7 +36,6 @@ class AbstractAnalyzer(ABC):
     def close(self) -> None:
         """ Free all resources used by the analyzer.
         """
     def close(self) -> None:
         """ Free all resources used by the analyzer.
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod


@@ -49,7 +55,6 @@ class AbstractAnalyzer(ABC):
                 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).
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod


@@ -65,7 +70,6 @@ class AbstractAnalyzer(ABC):
             Returns:
                 The given postcode after normalization.
         """
             Returns:
                 The given postcode after normalization.
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod


@@ -73,7 +77,6 @@ class AbstractAnalyzer(ABC):
         """ Update the tokenizer's postcode tokens from the current content
             of the `location_postcode` table.
         """
         """ Update the tokenizer's postcode tokens from the current content
             of the `location_postcode` table.
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod


@@ -89,7 +92,6 @@ class AbstractAnalyzer(ABC):
                                 When false, just add the given phrases to the
                                 ones that already exist.
         """
                                 When false, just add the given phrases to the
                                 ones that already exist.
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod


@@ -101,24 +103,16 @@ class AbstractAnalyzer(ABC):
                               refer to.
                 names: Dictionary of name type to name.
         """
                               refer to.
                 names: Dictionary of name type to name.
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod

-    def process_place(self, place: Dict) -> Any:
+    def process_place(self, place: PlaceInfo) -> Any:

         """ Extract tokens for the given place and compute the
             information to be handed to the PL/pgSQL processor for building
             the search index.
 
             Arguments:
         """ Extract tokens for the given place and compute the
             information to be handed to the PL/pgSQL processor for building
             the search index.
 
             Arguments:

-                place: Dictionary with the information about the place. Currently
-                       the following fields may be present:
-
-                       - *name* is a dictionary of names for the place together
-                         with the designation of the name.
-                       - *address* is a dictionary of address terms.
-                       - *country_feature* is set to a country code when the
-                         place describes a country.
+                place: Place information retrived from the database.

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


@@ -142,7 +136,7 @@ class AbstractTokenizer(ABC):
             the tokenizer remains stable over updates.
 
             Arguments:
             the tokenizer remains stable over updates.
 
             Arguments:

-              config: Read-only object with configuration obtions.
+              config: Read-only object with configuration options.

 
               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


@@ -151,17 +145,18 @@ class AbstractTokenizer(ABC):
 
             TODO: can we move the init_db parameter somewhere else?
         """
 
             TODO: can we move the init_db parameter somewhere else?
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod

-    def init_from_project(self) -> None:
+    def init_from_project(self, config: Configuration) -> None:

         """ Initialise the tokenizer from an existing database setup.
 
             The function should load all previously saved configuration from
             the project directory and/or the property table.
         """ Initialise the tokenizer from an existing database setup.
 
             The function should load all previously saved configuration from
             the project directory and/or the property table.

+
+            Arguments:
+              config: Read-only object with configuration options.

         """
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod


@@ -172,9 +167,8 @@ class AbstractTokenizer(ABC):
             during query time.
 
             Arguments:
             during query time.
 
             Arguments:

-              config: Read-only object with configuration obtions.
+              config: Read-only object with configuration options.

         """
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod


@@ -187,24 +181,39 @@ class AbstractTokenizer(ABC):
             data structures or data itself must not be changed by this function.
 
             Arguments:
             data structures or data itself must not be changed by this function.
 
             Arguments:

-              config: Read-only object with configuration obtions.
+              config: Read-only object with configuration options.

         """
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod

-    def check_database(self) -> str:
+    def check_database(self, config: Configuration) -> 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.
 

+            Arguments:
+              config: Read-only object with configuration options.
+

             Returns:
               If an issue was found, return an error message with the
               description of the issue as well as hints for the user on
             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.
+              how to resolve the issue. If everything is okay, return `None`.
+        """

 
 

-              Return `None`, if no issue was found.
+
+    @abstractmethod
+    def update_statistics(self) -> 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
+            it to be called in order to work.
+        """
+
+
+    @abstractmethod
+    def update_word_tokens(self) -> None:
+        """ Do house-keeping on the tokenizers internal data structures.
+            Remove unused word tokens, resort data etc.

         """
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod


@@ -221,4 +230,3 @@ 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.
         """

-        pass