"""
Tests for ICU tokenizer.
"""
-import shutil
import yaml
import itertools
sqldir.mkdir()
(sqldir / 'tokenizer').mkdir()
(sqldir / 'tokenizer' / 'icu_tokenizer.sql').write_text("SELECT 'a'")
- shutil.copy(str(project_env.lib_dir.sql / 'tokenizer' / 'icu_tokenizer_tables.sql'),
- str(sqldir / 'tokenizer' / 'icu_tokenizer_tables.sql'))
project_env.lib_dir.sql = sqldir
def test_finalize_import(tokenizer_factory, temp_db_conn,
temp_db_cursor, test_config, sql_preprocessor_cfg):
- func_file = test_config.lib_dir.sql / 'tokenizer' / 'legacy_tokenizer_indices.sql'
- func_file.write_text("""CREATE FUNCTION test() RETURNS TEXT
- AS $$ SELECT 'b'::text $$ LANGUAGE SQL""")
-
tok = tokenizer_factory()
tok.init_new_db(test_config)
+ assert not temp_db_conn.index_exists('idx_word_word_id')
+
tok.finalize_import(test_config)
- temp_db_cursor.scalar('SELECT test()') == 'b'
+ assert temp_db_conn.index_exists('idx_word_word_id')
def test_check_database(test_config, tokenizer_factory,
assert tok.check_database(test_config) is None
-def test_update_statistics_reverse_only(word_table, tokenizer_factory):
+def test_update_statistics_reverse_only(word_table, tokenizer_factory, test_config):
tok = tokenizer_factory()
- tok.update_statistics()
+ tok.update_statistics(test_config)
-def test_update_statistics(word_table, table_factory, temp_db_cursor, tokenizer_factory):
+def test_update_statistics(word_table, table_factory, temp_db_cursor,
+ tokenizer_factory, test_config):
word_table.add_full_word(1000, 'hello')
+ word_table.add_full_word(1001, 'bye')
table_factory('search_name',
- 'place_id BIGINT, name_vector INT[]',
- [(12, [1000])])
+ 'place_id BIGINT, name_vector INT[], nameaddress_vector INT[]',
+ [(12, [1000], [1001])])
tok = tokenizer_factory()
- tok.update_statistics()
+ tok.update_statistics(test_config)
assert temp_db_cursor.scalar("""SELECT count(*) FROM word
- WHERE type = 'W' and
- (info->>'count')::int > 0""") > 0
+ WHERE type = 'W' and word_id = 1000 and
+ (info->>'count')::int > 0""") == 1
+ assert temp_db_cursor.scalar("""SELECT count(*) FROM word
+ WHERE type = 'W' and word_id = 1001 and
+ (info->>'addr_count')::int > 0""") == 1
def test_normalize_postcode(analyzer):
def test_process_place_nonexisting_street(self):
info = self.process_address(street='Grand Road')
- assert 'street' not in info
+ assert info['street'] == '{}'
def test_process_place_multiple_street_tags(self):
def test_process_place_street_empty(self):
info = self.process_address(street='🜵')
- assert 'street' not in info
+ assert info['street'] == '{}'
def test_process_place_street_from_cache(self):
def test_process_place_place(self):
info = self.process_address(place='Honu Lulu')
- assert eval(info['place']) == self.name_token_set('HONU', 'LULU')
+ assert eval(info['place']) == self.name_token_set('HONU', 'LULU', '#HONU LULU')
def test_process_place_place_extra(self):
suburb='Zwickau', street='Hauptstr',
full='right behind the church')
- city = self.name_token_set('ZWICKAU')
- state = self.name_token_set('SACHSEN')
+ city = self.name_token_set('ZWICKAU', '#ZWICKAU')
+ state = self.name_token_set('SACHSEN', '#SACHSEN')
result = {k: eval(v) for k,v in info['addr'].items()}
result = {k: eval(v) for k,v in info['addr'].items()}
- assert result == {'city': self.name_token_set('Bruxelles')}
+ assert result == {'city': self.name_token_set('Bruxelles', '#Bruxelles')}
def test_process_place_address_terms_empty(self):