avoid LookupAny with address and too many name tokens

The index for nameaddress_vector has grown so large that PostgreSQL
will resort to a sequential scan if there are too many items
in the LookupAny list.

diff --git a/nominatim/api/search/db_search_builder.py b/nominatim/api/search/db_search_builder.py
index 3d5796c526c18494242efa442c46813aa6b02b14..f2b653f2c2def3c3657abdc9159c3be66d215d5f 100644 (file)
--- a/nominatim/api/search/db_search_builder.py
+++ b/nominatim/api/search/db_search_builder.py
@@ -176,11 +176,12 @@ class SearchBuilder:
             sdata.lookups.append(dbf.FieldLookup('nameaddress_vector',
                                                  list(partials), lookups.LookupAll))
         else:
             sdata.lookups.append(dbf.FieldLookup('nameaddress_vector',
                                                  list(partials), lookups.LookupAll))
         else:

+            addr_fulls = [t.token for t
+                          in self.query.get_tokens(address[0], TokenType.WORD)]
+            if len(addr_fulls) > 5:
+                return

             sdata.lookups.append(
             sdata.lookups.append(

-                dbf.FieldLookup('nameaddress_vector',
-                                [t.token for t
-                                 in self.query.get_tokens(address[0], TokenType.WORD)],
-                                lookups.LookupAny))
+                dbf.FieldLookup('nameaddress_vector', addr_fulls, lookups.LookupAny))

 
         sdata.housenumbers = dbf.WeightedStrings([], [])
         yield dbs.PlaceSearch(0.05, sdata, expected_count)
 
         sdata.housenumbers = dbf.WeightedStrings([], [])
         yield dbs.PlaceSearch(0.05, sdata, expected_count)