#coding:utf-8 """ ID: issue-1713 ISSUE: 1713 TITLE: Can't create table using long username and UTF8 as attachment charset DESCRIPTION: JIRA: CORE-1292 FBTEST: bugs.core_1292 NOTES: [23.08.2024] pzotov 1. Removed LIST() from initial query because it displays tokens in unpredictable order. This can cause fail if we change OptimizeForFirstRows = true config parameter. 2. Found oddities when try to use non-ascii user name and substitute it using f-notation: at least REVOKE and GRANT commands reflect this user name in the trace as encoded in cp1251 instead of utf8. This causes: 335544321 : arithmetic exception, numeric overflow, or string truncation 335544565 : Cannot transliterate character between character sets To be investigated further. """ import locale import pytest from firebird.qa import * db = db_factory(charset = 'utf8') act = python_act('db', substitutions = [ ('[ \t]+', ' '), ]) tmp_user = user_factory('db', name='Nebuchadnezzar_The_Babylon_Lord', password='123', plugin = 'Srp') #tmp_user = user_factory('db', name='"НавохудоносорВластелинВавилона2"', password='123', plugin = 'Srp') @pytest.mark.version('>=3') def test_1(act: Action, tmp_user: User, capsys): test_sql = f""" set bail on; set list on; set wng off; connect '{act.db.dsn}' user {act.db.user} password '{act.db.password}'; revoke all on all from {tmp_user.name}; grant create table to {tmp_user.name}; commit; connect '{act.db.dsn}' user {tmp_user.name} password '{tmp_user.password}'; select a.mon$user as who_am_i, c.rdb$character_set_name as my_connection_charset from mon$attachments a join rdb$character_sets c on a.mon$character_set_id = c.rdb$character_set_id where a.mon$attachment_id = current_connection; create table test(n int); commit; connect '{act.db.dsn}' user {act.db.user} password '{act.db.password}'; select p.rdb$user usr_name ,p.rdb$grantor grantor ,p.rdb$grant_option can_grant ,p.rdb$relation_name tab_name ,p.rdb$user_type usr_type ,p.rdb$object_type obj_type ,trim(p.rdb$privilege) priv from rdb$user_privileges p where upper(trim(p.rdb$relation_name)) = upper('test') and p.rdb$user = _utf8 '{tmp_user.name}' collate unicode_ci order by priv; commit; """ expected_stdout = f""" WHO_AM_I {tmp_user.name.upper()} MY_CONNECTION_CHARSET UTF8 USR_NAME {tmp_user.name.upper()} GRANTOR {tmp_user.name.upper()} CAN_GRANT 1 TAB_NAME TEST USR_TYPE 8 OBJ_TYPE 0 PRIV D USR_NAME {tmp_user.name.upper()} GRANTOR {tmp_user.name.upper()} CAN_GRANT 1 TAB_NAME TEST USR_TYPE 8 OBJ_TYPE 0 PRIV I USR_NAME {tmp_user.name.upper()} GRANTOR {tmp_user.name.upper()} CAN_GRANT 1 TAB_NAME TEST USR_TYPE 8 OBJ_TYPE 0 PRIV R USR_NAME {tmp_user.name.upper()} GRANTOR {tmp_user.name.upper()} CAN_GRANT 1 TAB_NAME TEST USR_TYPE 8 OBJ_TYPE 0 PRIV S USR_NAME {tmp_user.name.upper()} GRANTOR {tmp_user.name.upper()} CAN_GRANT 1 TAB_NAME TEST USR_TYPE 8 OBJ_TYPE 0 PRIV U """ act.expected_stdout = expected_stdout act.isql(switches = ['-q'], input = test_sql, charset = 'utf8', connect_db=False, credentials = False, combine_output = True, io_enc = locale.getpreferredencoding()) assert act.clean_stdout == act.clean_expected_stdout