def install_legacy_tokenizer(conn, config, **_):
""" Setup legacy tokenizer.
If no other tokenizer has been configured yet, then create the
configuration for the backwards-compatible legacy tokenizer
"""
if properties.get_property(conn, 'tokenizer') is None:
with conn.cursor() as cur:
for table in ('placex', 'location_property_osmline'):
has_column = cur.scalar("""SELECT count(*) FROM information_schema.columns
WHERE table_name = %s
and column_name = 'token_info'""",
(table, ))
if has_column == 0:
cur.execute('ALTER TABLE {} ADD COLUMN token_info JSONB'.format(table))
tokenizer = tokenizer_factory.create_tokenizer(config, init_db=False,
module_name='legacy')
tokenizer.migrate_database(config)
def migrate(config, paths):
""" Check for the current database version and execute migrations,
if necesssary.
"""
with connect(config.get_libpq_dsn()) as conn:
if conn.table_exists('nominatim_properties'):
db_version_str = properties.get_property(conn, 'database_version')
else:
db_version_str = None
if db_version_str is not None:
parts = db_version_str.split('.')
db_version = tuple([int(x) for x in parts[:2] + parts[2].split('-')])
if db_version == NOMINATIM_VERSION:
LOG.warning("Database already at latest version (%s)", db_version_str)
return 0
LOG.info("Detected database version: %s", db_version_str)
else:
db_version = _guess_version(conn)
has_run_migration = False
for version, func in _MIGRATION_FUNCTIONS:
if db_version <= version:
LOG.warning("Runnning: %s (%s)", func.__doc__.split('\n', 1)[0],
'{0[0]}.{0[1]}.{0[2]}-{0[3]}'.format(version))
kwargs = dict(conn=conn, config=config, paths=paths)
func(**kwargs)
has_run_migration = True
if has_run_migration:
LOG.warning('Updating SQL functions.')
refresh.create_functions(conn, config, paths.sqllib_dir)
properties.set_property(conn, 'database_version',
'{0[0]}.{0[1]}.{0[2]}-{0[3]}'.format(NOMINATIM_VERSION))
conn.commit()
return 0
def test_update_sql_functions(sql_preprocessor, temp_db_conn,
tokenizer_factory, test_config, table_factory,
monkeypatch, temp_db_cursor):
monkeypatch.setenv('NOMINATIM_MAX_WORD_FREQUENCY', '1133')
monkeypatch.setattr(legacy_tokenizer, '_check_module' , lambda m, c: None)
tok = tokenizer_factory()
tok.init_new_db(test_config)
monkeypatch.undo()
assert properties.get_property(temp_db_conn, legacy_tokenizer.DBCFG_MAXWORDFREQ) == '1133'
table_factory('test', 'txt TEXT')
func_file = test_config.lib_dir.sql / 'tokenizer' / 'legacy_tokenizer.sql'
func_file.write_text("""INSERT INTO test VALUES ('{{max_word_freq}}'),
('{{modulepath}}')""")
tok.update_sql_functions(test_config)
test_content = temp_db_cursor.row_set('SELECT * FROM test')
assert test_content == set((('1133', ), (str(test_config.project_dir / 'module'), )))
def test_get_property_unknown(prop_table, temp_db_conn, temp_db_cursor):
temp_db_cursor.execute(
"INSERT INTO nominatim_properties VALUES('other', 'bar')")
assert properties.get_property(temp_db_conn, 'foo') is None
def test_get_property_existing(prop_table, temp_db_conn, temp_db_cursor):
temp_db_cursor.execute(
"INSERT INTO nominatim_properties VALUES('foo', 'bar')")
assert properties.get_property(temp_db_conn, 'foo') == 'bar'
def _get_db_property(name):
return properties.get_property(temp_db_conn, name)
def test_get_property_unknown(property_factory, temp_db_conn):
property_factory('other', 'bar')
assert properties.get_property(temp_db_conn, 'foo') is None
def test_get_property_existing(property_factory, temp_db_conn):
property_factory('foo', 'bar')
assert properties.get_property(temp_db_conn, 'foo') == 'bar'
def init_from_project(self, _):
""" Initialise the tokenizer from the project directory.
"""
with connect(self.dsn) as conn:
self.normalization = properties.get_property(
conn, DBCFG_NORMALIZATION)
def _get_db_property(name):
return properties.get_property(temp_db_conn,
getattr(legacy_icu_tokenizer, name))
def get(self, name):
""" Set a property in the table to the given value.
"""
return properties.get_property(self.conn, name)