From 0045203092fbe4c630a4c5b823567912fdd618d2 Mon Sep 17 00:00:00 2001
From: Sarah Hoffmann <lonvia@denofr.de>
Date: Wed, 6 Aug 2025 17:27:52 +0200
Subject: [PATCH] don't restrict to viewbox for frequent terms

All searched places may be outside the viewbox in which case the
restriction means that there are no results at all. Add the penalty for
being outside the viewbox earlier instead and then cut the list.
---
 .../search/db_searches/place_search.py        | 22 ++++++++-----------
 test/python/api/search/test_search_places.py  |  8 ++++---
 2 files changed, 14 insertions(+), 16 deletions(-)

diff --git a/src/nominatim_api/search/db_searches/place_search.py b/src/nominatim_api/search/db_searches/place_search.py
index 3ef223be..e647632d 100644
--- a/src/nominatim_api/search/db_searches/place_search.py
+++ b/src/nominatim_api/search/db_searches/place_search.py
@@ -63,8 +63,7 @@ class PlaceSearch(base.AbstractSearch):
                         t.c.name_vector, t.c.nameaddress_vector,
                         sa.case((t.c.importance > 0, t.c.importance),
                                 else_=0.40001-(sa.cast(t.c.search_rank, sa.Float())/75))
-                          .label('importance'),
-                        penalty.label('penalty'))
+                          .label('importance'))
 
         for lookup in self.lookups:
             sql = sql.where(lookup.sql_condition(t))
@@ -88,10 +87,10 @@ class PlaceSearch(base.AbstractSearch):
                 sql = sql.where(t.c.centroid
                                    .intersects(VIEWBOX_PARAM,
                                                use_index=details.viewbox.area < 0.2))
-            elif not self.postcodes and self.expected_count >= 10000:
-                sql = sql.where(t.c.centroid
-                                   .intersects(VIEWBOX2_PARAM,
-                                               use_index=details.viewbox.area < 0.5))
+            else:
+                penalty += sa.case((t.c.centroid.intersects(VIEWBOX_PARAM, use_index=False), 0.0),
+                                   (t.c.centroid.intersects(VIEWBOX2_PARAM, use_index=False), 0.5),
+                                   else_=1.0)
 
         if details.near is not None and details.near_radius is not None:
             if details.near_radius < 0.1:
@@ -110,6 +109,8 @@ class PlaceSearch(base.AbstractSearch):
             sql = sql.where(sa.or_(t.c.address_rank <= MAX_RANK_PARAM,
                                    t.c.search_rank <= MAX_RANK_PARAM))
 
+        sql = sql.add_columns(penalty.label('penalty'))
+
         inner = sql.limit(5000 if self.qualifiers else 1000)\
                    .order_by(sa.desc(sa.text('importance')))\
                    .subquery()
@@ -121,8 +122,8 @@ class PlaceSearch(base.AbstractSearch):
         # If the query is not an address search or has a geographic preference,
         # preselect most important items to restrict the number of places
         # that need to be looked up in placex.
-        if (details.viewbox is None or details.bounded_viewbox)\
-           and (details.near is None or details.near_radius is not None)\
+        if (details.viewbox is None or not details.bounded_viewbox)\
+           and (details.near is None or details.near_radius is None)\
            and not self.qualifiers:
             sql = sql.add_columns(sa.func.first_value(inner.c.penalty - inner.c.importance)
                                     .over(order_by=inner.c.penalty - inner.c.importance)
@@ -166,11 +167,6 @@ class PlaceSearch(base.AbstractSearch):
                 # The postcode search needs to get priority here.
                 penalty += sa.case((t.c.postcode.in_(self.postcodes.values), 0.0), else_=1.0)
 
-        if details.viewbox is not None and not details.bounded_viewbox:
-            penalty += sa.case((t.c.geometry.intersects(VIEWBOX_PARAM, use_index=False), 0.0),
-                               (t.c.geometry.intersects(VIEWBOX2_PARAM, use_index=False), 0.5),
-                               else_=1.0)
-
         if details.near is not None:
             sql = sql.add_columns((-tsearch.c.centroid.ST_Distance(NEAR_PARAM))
                                   .label('importance'))
diff --git a/test/python/api/search/test_search_places.py b/test/python/api/search/test_search_places.py
index 31d1a778..7ea5d28e 100644
--- a/test/python/api/search/test_search_places.py
+++ b/test/python/api/search/test_search_places.py
@@ -156,9 +156,11 @@ class TestNameOnlySearches:
         assert result.place_id == 333
         assert len(geom['coordinates']) == npoints
 
-    @pytest.mark.parametrize('viewbox', ['5.0,4.0,6.0,5.0', '5.7,4.0,6.0,5.0'])
-    @pytest.mark.parametrize('wcount,rids', [(2, [100, 101]), (20000, [100])])
-    def test_prefer_viewbox(self, apiobj, frontend, viewbox, wcount, rids):
+    @pytest.mark.parametrize('viewbox,rids', [('5.0,4.0,6.0,5.0', [100]),
+                                              ('5.7,4.0,6.0,5.0', [100, 101]),
+                                              ('10.0,10.0,11.0,11.0', [101, 100])])
+    @pytest.mark.parametrize('wcount', [2, 50000])
+    def test_prefer_viewbox(self, apiobj, frontend, viewbox, rids, wcount):
         lookup = FieldLookup('name_vector', [1, 2], LookupAll)
         ranking = FieldRanking('name_vector', 0.2, [RankedTokens(0.0, [21])])
 
-- 
2.39.5