X-Git-Url: https://git.openstreetmap.org./nominatim.git/blobdiff_plain/d7a3039c2a4bd26d05f08ee3140b8dfaecd68f02..105597be056bf9cbf3bb4c15f9bc516479e48ee2:/nominatim/tokenizer/legacy_tokenizer.py?ds=sidebyside

diff --git a/nominatim/tokenizer/legacy_tokenizer.py b/nominatim/tokenizer/legacy_tokenizer.py
index e09700d9..1b68a494 100644
--- a/nominatim/tokenizer/legacy_tokenizer.py
+++ b/nominatim/tokenizer/legacy_tokenizer.py
@@ -256,6 +256,16 @@ class LegacyTokenizer(AbstractTokenizer):
         return LegacyNameAnalyzer(self.dsn, normalizer)
 
 
+    def most_frequent_words(self, conn: Connection, num: int) -> List[str]:
+        """ Return a list of the `num` most frequent full words
+            in the database.
+        """
+        with conn.cursor() as cur:
+            cur.execute(""" SELECT word FROM word WHERE word is not null
+                              ORDER BY search_name_count DESC LIMIT %s""", (num,))
+            return list(s[0] for s in cur)
+
+
     def _install_php(self, config: Configuration, overwrite: bool = True) -> None:
         """ Install the php script for the tokenizer.
         """