indexing: precompute row counts

[nominatim.git] / test / python / tokenizer / sanitizers / test_clean_housenumbers.py

diff --git a/test/python/tokenizer/sanitizers/test_clean_housenumbers.py b/test/python/tokenizer/sanitizers/test_clean_housenumbers.py
index 5784619b59a584c35fa982617484de1d74a8a780..01fb162b8e78ae4b90b7ac6253ff43dd1212875d 100644 (file)
--- a/test/python/tokenizer/sanitizers/test_clean_housenumbers.py
+++ b/test/python/tokenizer/sanitizers/test_clean_housenumbers.py
@@ -1,26 +1,26 @@
-# SPDX-License-Identifier: GPL-2.0-only
+# SPDX-License-Identifier: GPL-3.0-or-later
 #
 # This file is part of Nominatim. (https://nominatim.org)
 #
-# Copyright (C) 2022 by the Nominatim developer community.
+# Copyright (C) 2024 by the Nominatim developer community.
 # For a full list of authors see the git log.
 """
 Tests for the sanitizer that normalizes housenumbers.
 """
 import pytest
 
-from nominatim.tokenizer.place_sanitizer import PlaceSanitizer
-from nominatim.indexer.place_info import PlaceInfo
+from nominatim_db.tokenizer.place_sanitizer import PlaceSanitizer
+from nominatim_db.data.place_info import PlaceInfo
 
 @pytest.fixture
-def sanitize(request):
+def sanitize(request, def_config):
     sanitizer_args = {'step': 'clean-housenumbers'}
     for mark in request.node.iter_markers(name="sanitizer_params"):
         sanitizer_args.update({k.replace('_', '-') : v for k,v in mark.kwargs.items()})
 
     def _run(**kwargs):
         place = PlaceInfo({'address': kwargs})
-        _, address = PlaceSanitizer([sanitizer_args]).process_names(place)
+        _, address = PlaceSanitizer([sanitizer_args], def_config).process_names(place)
 
         return sorted([(p.kind, p.name) for p in address])
 
@@ -42,3 +42,27 @@ def test_housenumber_lists(sanitize, number):
 def test_filter_kind(sanitize):
     assert sanitize(housenumber='34', number='4', badnumber='65') == \
             [('badnumber', '65'), ('housenumber', '34'), ('housenumber', '4')]
+
+
+@pytest.mark.parametrize('number', ('6523', 'n/a', '4'))
+def test_convert_to_name_converted(def_config, number):
+    sanitizer_args = {'step': 'clean-housenumbers',
+                      'convert-to-name': (r'\d+', 'n/a')}
+
+    place = PlaceInfo({'address': {'housenumber': number}})
+    names, address = PlaceSanitizer([sanitizer_args], def_config).process_names(place)
+
+    assert ('housenumber', number) in set((p.kind, p.name) for p in names)
+    assert 'housenumber' not in set(p.kind for p in address)
+
+
+@pytest.mark.parametrize('number', ('a54', 'n.a', 'bow'))
+def test_convert_to_name_unconverted(def_config, number):
+    sanitizer_args = {'step': 'clean-housenumbers',
+                      'convert-to-name': (r'\d+', 'n/a')}
+
+    place = PlaceInfo({'address': {'housenumber': number}})
+    names, address = PlaceSanitizer([sanitizer_args], def_config).process_names(place)
+
+    assert 'housenumber' not in set(p.kind for p in names)
+    assert ('housenumber', number) in set((p.kind, p.name) for p in address)