test/python/tokenizer/sanitizers/test_strip_brace_terms.py

   1 """
   2 Tests for the sanitizer that handles braced suffixes.
   3 """
   4 import pytest
   5
   6 from nominatim.tokenizer.place_sanitizer import PlaceSanitizer
   7 from nominatim.indexer.place_info import PlaceInfo
   8
   9 def run_sanitizer_on(**kwargs):
  10     place = PlaceInfo({'name': kwargs})
  11     name, _ = PlaceSanitizer([{'step': 'strip-brace-terms'}]).process_names(place)
  12
  13     return sorted([(p.name, p.kind, p.suffix) for p in name])
  14
  15
  16 def test_no_braces():
  17     assert run_sanitizer_on(name='foo', ref='23') == [('23', 'ref', None),
  18                                                       ('foo', 'name', None)]
  19
  20
  21 def test_simple_braces():
  22     assert run_sanitizer_on(name='Halle (Saale)', ref='3')\
  23       == [('3', 'ref', None), ('Halle', 'name', None), ('Halle (Saale)', 'name', None)]
  24     assert run_sanitizer_on(name='ack ( bar')\
  25       == [('ack', 'name', None), ('ack ( bar', 'name', None)]
  26
  27
  28 def test_only_braces():
  29     assert run_sanitizer_on(name='(maybe)') == [('(maybe)', 'name', None)]
  30
  31
  32 def test_double_braces():
  33     assert run_sanitizer_on(name='a((b))') == [('a', 'name', None),
  34                                                ('a((b))', 'name', None)]
  35     assert run_sanitizer_on(name='a (b) (c)') == [('a', 'name', None),
  36                                                   ('a (b) (c)', 'name', None)]
  37
  38
  39 def test_no_names():
  40     place = PlaceInfo({'address': {'housenumber': '3'}})
  41     name, address = PlaceSanitizer([{'step': 'strip-brace-terms'}]).process_names(place)
  42
  43     assert not name
  44     assert len(address) == 1