X-Git-Url: https://git.openstreetmap.org/nominatim.git/blobdiff_plain/261e0cfd5a084dc5de333eda1093a9e7635c4859..faeee7528f710dc98bf14a08e592137d3e6d37f2:/nominatim/tokenizer/icu_tokenizer.py

diff --git a/nominatim/tokenizer/icu_tokenizer.py b/nominatim/tokenizer/icu_tokenizer.py
index b6e64637..39c1cbc6 100644
--- a/nominatim/tokenizer/icu_tokenizer.py
+++ b/nominatim/tokenizer/icu_tokenizer.py
@@ -183,6 +183,18 @@ class ICUTokenizer(AbstractTokenizer):
                                self.loader.make_token_analysis())
 
 
+    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, sum((info->'count')::int) as count
+                             FROM word WHERE type = 'W'
+                             GROUP BY word
+                             ORDER BY count DESC LIMIT %s""", (num,))
+            return list(s[0].split('@')[0] for s in cur)
+
+
     def _install_php(self, phpdir: Path, overwrite: bool = True) -> None:
         """ Install the php script for the tokenizer.
         """