remove special casing for legacy tokenizer from BDD tests

This commit is contained in:
Sarah Hoffmann
2024-09-21 17:07:32 +02:00
parent 74c39267d9
commit d4f3eda314
13 changed files with 7 additions and 282 deletions

View File

@@ -34,7 +34,6 @@ class NominatimEnvironment:
self.api_test_file = config['API_TEST_FILE']
self.tokenizer = config['TOKENIZER']
self.import_style = config['STYLE']
self.server_module_path = config['SERVER_MODULE_PATH']
self.reuse_template = not config['REMOVE_TEMPLATE']
self.keep_scenario_db = config['KEEP_TEST_DB']
@@ -48,9 +47,6 @@ class NominatimEnvironment:
raise RuntimeError(f"Unknown API engine '{config['API_ENGINE']}'")
self.api_engine = getattr(self, f"create_api_request_func_{config['API_ENGINE']}")()
if self.tokenizer == 'legacy' and self.server_module_path is None:
raise RuntimeError("You must set -DSERVER_MODULE_PATH when testing the legacy tokenizer.")
def connect_database(self, dbname):
""" Return a connection to the database with the given name.
Uses configured host, user and port.
@@ -100,9 +96,6 @@ class NominatimEnvironment:
if self.import_style is not None:
self.test_env['NOMINATIM_IMPORT_STYLE'] = self.import_style
if self.server_module_path:
self.test_env['NOMINATIM_DATABASE_MODULE_PATH'] = self.server_module_path
if self.website_dir is not None:
self.website_dir.cleanup()
@@ -111,7 +104,6 @@ class NominatimEnvironment:
def get_test_config(self):
cfg = Configuration(Path(self.website_dir.name), environ=self.test_env)
cfg.set_libdirs(module=self.server_module_path)
return cfg
def get_libpq_dsn(self):
@@ -190,12 +182,8 @@ class NominatimEnvironment:
self.run_nominatim('add-data', '--tiger-data', str(testdata / 'tiger'))
self.run_nominatim('freeze')
if self.tokenizer == 'legacy':
phrase_file = str(testdata / 'specialphrases_testdb.sql')
run_script(['psql', '-d', self.api_test_db, '-f', phrase_file])
else:
csv_path = str(testdata / 'full_en_phrases_test.csv')
self.run_nominatim('special-phrases', '--import-from-csv', csv_path)
csv_path = str(testdata / 'full_en_phrases_test.csv')
self.run_nominatim('special-phrases', '--import-from-csv', csv_path)
except:
self.db_drop_database(self.api_test_db)
raise
@@ -278,7 +266,7 @@ class NominatimEnvironment:
if self.website_dir is not None:
cmdline = list(cmdline) + ['--project-dir', self.website_dir.name]
cli.nominatim(module_dir=self.server_module_path,
cli.nominatim(module_dir=None,
osm2pgsql_path=None,
cli_args=cmdline,
environ=self.test_env)

View File

@@ -28,9 +28,8 @@ def check_database_integrity(context):
assert cur.fetchone()[0] == 0, "Duplicates found in place_addressline"
# word table must not have empty word_tokens
if context.nominatim.tokenizer != 'legacy':
cur.execute("SELECT count(*) FROM word WHERE word_token = ''")
assert cur.fetchone()[0] == 0, "Empty word tokens found in word table"
cur.execute("SELECT count(*) FROM word WHERE word_token = ''")
assert cur.fetchone()[0] == 0, "Empty word tokens found in word table"
@@ -324,13 +323,8 @@ def check_word_table_for_postcodes(context, exclude, postcodes):
plist.sort()
with context.db.cursor() as cur:
if nctx.tokenizer != 'legacy':
cur.execute("SELECT word FROM word WHERE type = 'P' and word = any(%s)",
(plist,))
else:
cur.execute("""SELECT word FROM word WHERE word = any(%s)
and class = 'place' and type = 'postcode'""",
(plist,))
cur.execute("SELECT word FROM word WHERE type = 'P' and word = any(%s)",
(plist,))
found = [row['word'] for row in cur]
assert len(found) == len(set(found)), f"Duplicate rows for postcodes: {found}"