add support for external sanitizer modules

This commit is contained in:
Sarah Hoffmann
2022-07-25 16:10:19 +02:00
parent 7b7203c149
commit 6d41046b15
9 changed files with 142 additions and 94 deletions

View File

@@ -13,14 +13,14 @@ from nominatim.tokenizer.place_sanitizer import PlaceSanitizer
from nominatim.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])
@@ -45,24 +45,24 @@ def test_filter_kind(sanitize):
@pytest.mark.parametrize('number', ('6523', 'n/a', '4'))
def test_convert_to_name_converted(number):
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]).process_names(place)
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(number):
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]).process_names(place)
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)