penalize search with frequent partials

[nominatim.git] / nominatim / api / search / db_search_builder.py

diff --git a/nominatim/api/search/db_search_builder.py b/nominatim/api/search/db_search_builder.py
index 9ff8c03c90c3d6ef4b7f1ff1c038e24bdb165171..ee06dba5ef95f916bceb10edc9148366a1ba0d8b 100644 (file)
--- a/nominatim/api/search/db_search_builder.py
+++ b/nominatim/api/search/db_search_builder.py
@@ -163,6 +163,7 @@ class SearchBuilder:
         sdata.lookups = [dbf.FieldLookup('name_vector', [t.token for t in hnrs], 'lookup_any'),
                          dbf.FieldLookup('nameaddress_vector', partial_tokens, 'lookup_all')
                         ]
+        sdata.housenumbers = dbf.WeightedStrings([], [])
         yield dbs.PlaceSearch(0.05, sdata, sum(t.count for t in hnrs))
 
 
@@ -210,21 +211,21 @@ class SearchBuilder:
         exp_addr_count = min(t.count for t in addr_partials) if addr_partials else exp_name_count
         if exp_addr_count < 1000 and partials_indexed:
             # Lookup by address partials and restrict results through name terms.
-            yield penalty, exp_addr_count,\
+            # Give this a small penalty because lookups in the address index are
+            # more expensive
+            yield penalty + exp_addr_count/5000, exp_addr_count,\
                   [dbf.FieldLookup('name_vector', [t.token for t in name_partials], 'restrict'),
                    dbf.FieldLookup('nameaddress_vector', addr_tokens, 'lookup_all')]
             return
 
         # Partial term to frequent. Try looking up by rare full names first.
         name_fulls = self.query.get_tokens(name, TokenType.WORD)
-        rare_names = list(filter(lambda t: t.count < 1000, name_fulls))
+        rare_names = list(filter(lambda t: t.count < 10000, name_fulls))
         # At this point drop unindexed partials from the address.
         # This might yield wrong results, nothing we can do about that.
         if not partials_indexed:
             addr_tokens = [t.token for t in addr_partials if t.is_indexed]
-            log().var_dump('before', penalty)
             penalty += 1.2 * sum(t.penalty for t in addr_partials if not t.is_indexed)
-            log().var_dump('after', penalty)
         if rare_names:
             # Any of the full names applies with all of the partials from the address
             lookup = [dbf.FieldLookup('name_vector', [t.token for t in rare_names], 'lookup_any')]
@@ -233,19 +234,25 @@ class SearchBuilder:
             yield penalty, sum(t.count for t in rare_names), lookup
 
         # To catch remaining results, lookup by name and address
-        if all(t.is_indexed for t in name_partials):
-            lookup = [dbf.FieldLookup('name_vector',
-                                      [t.token for t in name_partials], 'lookup_all')]
-        else:
-            # we don't have the partials, try with the non-rare names
-            non_rare_names = [t.token for t in name_fulls if t.count >= 1000]
-            if not non_rare_names:
-                return
-            lookup = [dbf.FieldLookup('name_vector', non_rare_names, 'lookup_any')]
-        if addr_tokens:
-            lookup.append(dbf.FieldLookup('nameaddress_vector', addr_tokens, 'lookup_all'))
-        yield penalty + 0.1 * max(0, 5 - len(name_partials) - len(addr_tokens)),\
-              min(exp_name_count, exp_addr_count), lookup
+        # We only do this if there is a reasonable number of results expected.
+        if min(exp_name_count, exp_addr_count) < 10000:
+            if all(t.is_indexed for t in name_partials):
+                lookup = [dbf.FieldLookup('name_vector',
+                                          [t.token for t in name_partials], 'lookup_all')]
+            else:
+                # we don't have the partials, try with the non-rare names
+                non_rare_names = [t.token for t in name_fulls if t.count >= 1000]
+                if not non_rare_names:
+                    return
+                lookup = [dbf.FieldLookup('name_vector', non_rare_names, 'lookup_any')]
+            if addr_tokens:
+                lookup.append(dbf.FieldLookup('nameaddress_vector', addr_tokens, 'lookup_all'))
+            penalty += 0.1 * max(0, 5 - len(name_partials) - len(addr_tokens))
+            if len(rare_names) == len(name_fulls):
+                # if there already was a search for all full tokens,
+                # avoid this if anything has been found
+                penalty += 0.25
+            yield penalty, min(exp_name_count, exp_addr_count), lookup
 
 
     def get_name_ranking(self, trange: TokenRange) -> dbf.FieldRanking: