Merge pull request #2757 from lonvia/filter-postcodes

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

diff --git a/nominatim/tokenizer/base.py b/nominatim/tokenizer/base.py
index 94fac1fc4e1cdcab553af27f6fe1a94b21a9f95f..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.


@@ -30,7 +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


@@ -50,7 +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


@@ -66,7 +70,6 @@ class AbstractAnalyzer(ABC):
             Returns:
                 The given postcode after normalization.
         """
             Returns:
                 The given postcode after normalization.
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod


@@ -74,7 +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


@@ -90,7 +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


@@ -102,7 +103,6 @@ class AbstractAnalyzer(ABC):
                               refer to.
                 names: Dictionary of name type to name.
         """
                               refer to.
                 names: Dictionary of name type to name.
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod


@@ -145,7 +145,6 @@ 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


@@ -158,7 +157,6 @@ class AbstractTokenizer(ABC):
             Arguments:
               config: Read-only object with configuration options.
         """
             Arguments:
               config: Read-only object with configuration options.
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod


@@ -171,7 +169,6 @@ class AbstractTokenizer(ABC):
             Arguments:
               config: Read-only object with configuration options.
         """
             Arguments:
               config: Read-only object with configuration options.
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod


@@ -186,7 +183,6 @@ class AbstractTokenizer(ABC):
             Arguments:
               config: Read-only object with configuration options.
         """
             Arguments:
               config: Read-only object with configuration options.
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod


@@ -202,7 +198,6 @@ class AbstractTokenizer(ABC):
               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`.
         """

-        pass

 
 
     @abstractmethod
 
 
     @abstractmethod


@@ -212,7 +207,13 @@ class AbstractTokenizer(ABC):
             to improve performance. However, the tokenizer must not depend on
             it to be called in order to work.
         """
             to improve performance. However, the tokenizer must not depend on
             it to be called in order to work.
         """

-        pass
+
+
+    @abstractmethod
+    def update_word_tokens(self) -> None:
+        """ Do house-keeping on the tokenizers internal data structures.
+            Remove unused word tokens, resort data etc.
+        """

 
 
     @abstractmethod
 
 
     @abstractmethod


@@ -229,4 +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