add address counts to tokens

nominatim/api/search/icu_tokenizer.py

@@ -97,6 +97,7 @@ class ICUToken(qmod.Token):
         """ Create a ICUToken from the row of the word table.
         """
         count = 1 if row.info is None else row.info.get('count', 1)
+        addr_count = 1 if row.info is None else row.info.get('addr_count', 1)
 
         penalty = 0.0
         if row.type == 'w':
@@ -123,7 +124,8 @@ class ICUToken(qmod.Token):
 
         return ICUToken(penalty=penalty, token=row.word_id, count=count,
                         lookup_word=lookup_word, is_indexed=True,
-                        word_token=row.word_token, info=row.info)
+                        word_token=row.word_token, info=row.info,
+                        addr_count=addr_count)
 
 
 
@@ -257,7 +259,7 @@ class ICUQueryAnalyzer(AbstractQueryAnalyzer):
             if len(part.token) <= 4 and part[0].isdigit()\
                and not node.has_tokens(i+1, qmod.TokenType.HOUSENUMBER):
                 query.add_token(qmod.TokenRange(i, i+1), qmod.TokenType.HOUSENUMBER,
-                                ICUToken(0.5, 0, 1, part.token, True, part.token, None))
+                                ICUToken(0.5, 0, 1, 1, part.token, True, part.token, None))
 
 
     def rerank_tokens(self, query: qmod.QueryStruct, parts: QueryParts) -> None:

nominatim/api/search/legacy_tokenizer.py

@@ -210,6 +210,7 @@ class LegacyQueryAnalyzer(AbstractQueryAnalyzer):
 
         return LegacyToken(penalty=penalty, token=row.word_id,
                            count=row.search_name_count or 1,
+                           addr_count=1, # not supported
                            lookup_word=lookup_word,
                            word_token=row.word_token.strip(),
                            category=(rowclass, row.type) if rowclass is not None else None,
@@ -226,7 +227,7 @@ class LegacyQueryAnalyzer(AbstractQueryAnalyzer):
             if len(part) <= 4 and part.isdigit()\
                and not node.has_tokens(i+1, qmod.TokenType.HOUSENUMBER):
                 query.add_token(qmod.TokenRange(i, i+1), qmod.TokenType.HOUSENUMBER,
-                                LegacyToken(penalty=0.5, token=0, count=1,
+                                LegacyToken(penalty=0.5, token=0, count=1, addr_count=1,
                                             lookup_word=part, word_token=part,
                                             category=None, country=None,
                                             operator=None, is_indexed=True))

nominatim/api/search/query.py

@@ -99,10 +99,10 @@ class Token(ABC):
     penalty: float
     token: int
     count: int
+    addr_count: int
     lookup_word: str
     is_indexed: bool
 
-    addr_count: int = 1
 
     @abstractmethod
     def get_category(self) -> Tuple[str, str]:

nominatim/tokenizer/base.py

@@ -201,7 +201,7 @@ class AbstractTokenizer(ABC):
 
 
     @abstractmethod
-    def update_statistics(self, config: Configuration) -> 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

nominatim/tokenizer/legacy_tokenizer.py

@@ -210,7 +210,7 @@ class LegacyTokenizer(AbstractTokenizer):
             self._save_config(conn, config)
 
 
-    def update_statistics(self, _: Configuration) -> None:
+    def update_statistics(self, config: Configuration, threads: int = 1) -> None:
         """ Recompute the frequency of full words.
         """
         with connect(self.dsn) as conn: