do not count words when in reverse-only mode

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

diff --git a/nominatim/tokenizer/icu_tokenizer.py b/nominatim/tokenizer/icu_tokenizer.py
index 2ece10f2ccd28aecb3a181ceafabeb86c9f289a7..3331a3210aaba70d49b602299c2ce9e88238a3a0 100644 (file)
--- a/nominatim/tokenizer/icu_tokenizer.py
+++ b/nominatim/tokenizer/icu_tokenizer.py
@@ -2,7 +2,6 @@
 Tokenizer implementing normalisation as used before Nominatim 4 but using
 libICU instead of the PostgreSQL module.
 """
 Tokenizer implementing normalisation as used before Nominatim 4 but using
 libICU instead of the PostgreSQL module.
 """

-from collections import Counter

 import itertools
 import json
 import logging
 import itertools
 import json
 import logging


@@ -68,10 +67,13 @@ class LegacyICUTokenizer(AbstractTokenizer):
             self.term_normalization = get_property(conn, DBCFG_TERM_NORMALIZATION)
 
 
             self.term_normalization = get_property(conn, DBCFG_TERM_NORMALIZATION)
 
 

-    def finalize_import(self, _):
+    def finalize_import(self, config):

         """ Do any required postprocessing to make the tokenizer data ready
             for use.
         """
         """ Do any required postprocessing to make the tokenizer data ready
             for use.
         """

+        with connect(self.dsn) as conn:
+            sqlp = SQLPreprocessor(conn, config)
+            sqlp.run_sql_file(conn, 'tokenizer/legacy_tokenizer_indices.sql')

 
 
     def update_sql_functions(self, config):
 
 
     def update_sql_functions(self, config):


@@ -93,6 +95,26 @@ class LegacyICUTokenizer(AbstractTokenizer):
         return None
 
 
         return None
 
 

+    def update_statistics(self):
+        """ Recompute frequencies for all name words.
+        """
+        with connect(self.dsn) as conn:
+            if conn.table_exists('search_name'):
+                with conn.cursor() as cur:
+                    cur.drop_table("word_frequencies")
+                    LOG.info("Computing word frequencies")
+                    cur.execute("""CREATE TEMP TABLE word_frequencies AS
+                                     SELECT unnest(name_vector) as id, count(*)
+                                     FROM search_name GROUP BY id""")
+                    cur.execute("CREATE INDEX ON word_frequencies(id)")
+                    LOG.info("Update word table with recomputed frequencies")
+                    cur.execute("""UPDATE word
+                                   SET info = info || jsonb_build_object('count', count)
+                                   FROM word_frequencies WHERE word_id = id""")
+                    cur.drop_table("word_frequencies")
+            conn.commit()
+
+

     def name_analyzer(self):
         """ Create a new analyzer for tokenizing names and queries
             using this tokinzer. Analyzers are context managers and should
     def name_analyzer(self):
         """ Create a new analyzer for tokenizing names and queries
             using this tokinzer. Analyzers are context managers and should


@@ -142,45 +164,6 @@ class LegacyICUTokenizer(AbstractTokenizer):
             sqlp.run_sql_file(conn, 'tokenizer/icu_tokenizer_tables.sql')
             conn.commit()
 
             sqlp.run_sql_file(conn, 'tokenizer/icu_tokenizer_tables.sql')
             conn.commit()
 

-            LOG.warning("Precomputing word tokens")
-
-            # get partial words and their frequencies
-            words = self._count_partial_terms(conn)
-
-            # copy them back into the word table
-            with CopyBuffer() as copystr:
-                for term, cnt in words.items():
-                    copystr.add('w', term, json.dumps({'count': cnt}))
-
-                with conn.cursor() as cur:
-                    copystr.copy_out(cur, 'word',
-                                     columns=['type', 'word_token', 'info'])
-                    cur.execute("""UPDATE word SET word_id = nextval('seq_word')
-                                   WHERE word_id is null and type = 'w'""")
-
-            conn.commit()
-
-    def _count_partial_terms(self, conn):
-        """ Count the partial terms from the names in the place table.
-        """
-        words = Counter()
-        name_proc = self.loader.make_token_analysis()
-
-        with conn.cursor(name="words") as cur:
-            cur.execute(""" SELECT v, count(*) FROM
-                              (SELECT svals(name) as v FROM place)x
-                            WHERE length(v) < 75 GROUP BY v""")
-
-            for name, cnt in cur:
-                terms = set()
-                for word in name_proc.get_variants_ascii(name_proc.get_normalized(name)):
-                    if ' ' in word:
-                        terms.update(word.split())
-                for term in terms:
-                    words[term] += cnt
-
-        return words
-

 
 class LegacyICUNameAnalyzer(AbstractAnalyzer):
     """ The legacy analyzer uses the ICU library for splitting names.
 
 class LegacyICUNameAnalyzer(AbstractAnalyzer):
     """ The legacy analyzer uses the ICU library for splitting names.


@@ -209,14 +192,14 @@ class LegacyICUNameAnalyzer(AbstractAnalyzer):
     def _search_normalized(self, name):
         """ Return the search token transliteration of the given name.
         """
     def _search_normalized(self, name):
         """ Return the search token transliteration of the given name.
         """

-        return self.token_analysis.get_search_normalized(name)
+        return self.token_analysis.search.transliterate(name).strip()

 
 
     def _normalized(self, name):
         """ Return the normalized version of the given name with all
             non-relevant information removed.
         """
 
 
     def _normalized(self, name):
         """ Return the normalized version of the given name with all
             non-relevant information removed.
         """

-        return self.token_analysis.get_normalized(name)
+        return self.token_analysis.normalizer.transliterate(name).strip()

 
 
     def get_word_token_info(self, words):
 
 
     def get_word_token_info(self, words):


@@ -456,6 +439,7 @@ class LegacyICUNameAnalyzer(AbstractAnalyzer):
         if addr_terms:
             token_info.add_address_terms(addr_terms)
 
         if addr_terms:
             token_info.add_address_terms(addr_terms)
 

+

     def _compute_partial_tokens(self, name):
         """ Normalize the given term, split it into partial words and return
             then token list for them.
     def _compute_partial_tokens(self, name):
         """ Normalize the given term, split it into partial words and return
             then token list for them.


@@ -492,19 +476,25 @@ class LegacyICUNameAnalyzer(AbstractAnalyzer):
         partial_tokens = set()
 
         for name in names:
         partial_tokens = set()
 
         for name in names:

+            analyzer_id = name.get_attr('analyzer')

             norm_name = self._normalized(name.name)
             norm_name = self._normalized(name.name)

-            full, part = self._cache.names.get(norm_name, (None, None))
+            if analyzer_id is None:
+                token_id = norm_name
+            else:
+                token_id = f'{norm_name}@{analyzer_id}'
+
+            full, part = self._cache.names.get(token_id, (None, None))

             if full is None:
             if full is None:

-                variants = self.token_analysis.get_variants_ascii(norm_name)
+                variants = self.token_analysis.analysis[analyzer_id].get_variants_ascii(norm_name)

                 if not variants:
                     continue
 
                 with self.conn.cursor() as cur:
                     cur.execute("SELECT (getorcreate_full_word(%s, %s)).*",
                 if not variants:
                     continue
 
                 with self.conn.cursor() as cur:
                     cur.execute("SELECT (getorcreate_full_word(%s, %s)).*",

-                                (norm_name, variants))
+                                (token_id, variants))

                     full, part = cur.fetchone()
 
                     full, part = cur.fetchone()
 

-                self._cache.names[norm_name] = (full, part)
+                self._cache.names[token_id] = (full, part)

 
             full_tokens.add(full)
             partial_tokens.update(part)
 
             full_tokens.add(full)
             partial_tokens.update(part)