[sqlalchemy] Re: isolation level not supported on MySQL 3.23
Mmm. The traceback I got last (TypeError), only occurs on a client platform (redhat 8.0) using mysql 3.23 client software. When run on a platform with MySQL 5.0 (suse 10.1) software you're solution by raising the NotImplemented exception works. redhat 8.0 + mysql 3.23 client + python-mysql 1.2.2 + sqlalchemy 0.7.8 + mysql server 3.23.31 (on other host) fails suse 10.1 + mysql 5.0 client + python-mysql 1.2.2 + sqlalchemy 0.7.8 + mysql server 3.23.54 (on other host) works Thanks -- You received this message because you are subscribed to the Google Groups sqlalchemy group. To view this discussion on the web visit https://groups.google.com/d/msg/sqlalchemy/-/RXAaI5QEo48J. To post to this group, send email to sqlalchemy@googlegroups.com. To unsubscribe from this group, send email to sqlalchemy+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en.
[sqlalchemy] isolation level not supported on MySQL 3.23
We were using SQLAlchemy 0.5.1 and wanted to upgrade to 0.7.8 but ran into the following problem. When trying to create an engine, the mysql dialect tries to determine the current isolation level by issuing the SELECT @@tx_isolation; SQL statement (from dialects/mysql/base.py get_isolation_level()). However, this statement is not supported on MySQL 3.23 and therefore SQLAlchemy 0.7.8 doesn't seem to work anymore. We worked around this by not using SQLAlchemy anymore for our antique MySQL 3.23 db, but this means you could also delete 3.23 from the supported database list. As a solution you could maybe add a try/except clause around it with a version check just like you do in do_commit() in dialects/mysql/base.py Best regards, Ids -- You received this message because you are subscribed to the Google Groups sqlalchemy group. To view this discussion on the web visit https://groups.google.com/d/msg/sqlalchemy/-/uJQPXujgUCEJ. To post to this group, send email to sqlalchemy@googlegroups.com. To unsubscribe from this group, send email to sqlalchemy+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en.
[sqlalchemy] Re: isolation level not supported on MySQL 3.23
We are using MySQL 3.23.31 on that machine :-) It seems tx_isolation was added ad 4.0.3 (according to http://dev.mysql.com/doc/refman/4.1/en/server-system-variables.html#sysvar_tx_isolation). But if I add if self.server_version_info (4, 0, 3): raise NotImplementedError() I get another traceback: Traceback (most recent call last): File ./t.py, line 9, in ? con=engine.connect() File /opt/python-2.4/lib/python2.4/site-packages/sqlalchemy/engine/base.py, line 2472, in connect return self._connection_cls(self, **kwargs) File /opt/python-2.4/lib/python2.4/site-packages/sqlalchemy/engine/base.py, line 878, in __init__ self.__connection = connection or engine.raw_connection() File /opt/python-2.4/lib/python2.4/site-packages/sqlalchemy/engine/base.py, line 2558, in raw_connection return self.pool.unique_connection() File /opt/python-2.4/lib/python2.4/site-packages/sqlalchemy/pool.py, line 183, in unique_connection return _ConnectionFairy(self).checkout() File /opt/python-2.4/lib/python2.4/site-packages/sqlalchemy/pool.py, line 387, in __init__ rec = self._connection_record = pool._do_get() File /opt/python-2.4/lib/python2.4/site-packages/sqlalchemy/pool.py, line 741, in _do_get con = self._create_connection() File /opt/python-2.4/lib/python2.4/site-packages/sqlalchemy/pool.py, line 188, in _create_connection return _ConnectionRecord(self) File /opt/python-2.4/lib/python2.4/site-packages/sqlalchemy/pool.py, line 273, in __init__ pool.dispatch.first_connect.exec_once(self.connection, self) File /opt/python-2.4/lib/python2.4/site-packages/sqlalchemy/event.py, line 282, in exec_once self(*args, **kw) File /opt/python-2.4/lib/python2.4/site-packages/sqlalchemy/event.py, line 291, in __call__ fn(*args, **kw) File /opt/python-2.4/lib/python2.4/site-packages/sqlalchemy/engine/strategies.py, line 167, in first_connect dialect.initialize(c) File /opt/python-2.4/lib/python2.4/site-packages/sqlalchemy/dialects/mysql/base.py, line 2005, in initialize self._detect_ansiquotes(connection) File /opt/python-2.4/lib/python2.4/site-packages/sqlalchemy/dialects/mysql/base.py, line 2246, in _detect_ansiquotes mode = row[1] or '' File /opt/python-2.4/lib/python2.4/site-packages/sqlalchemy/dialects/mysql/base.py, line 2737, in __getitem__ item = self.rowproxy[index] TypeError: unsubscriptable object My test script is pretty simple. ... url='%(driver)s://%(user)s:%(passwd)s@%(host)s/%(db)s' % cfg engine=create_engine(url) con=engine.connect() con.close() Thanks Op maandag 17 september 2012 13:21:02 UTC+2 schreef Ids het volgende: We were using SQLAlchemy 0.5.1 and wanted to upgrade to 0.7.8 but ran into the following problem. When trying to create an engine, the mysql dialect tries to determine the current isolation level by issuing the SELECT @@tx_isolation; SQL statement (from dialects/mysql/base.py get_isolation_level()). However, this statement is not supported on MySQL 3.23 and therefore SQLAlchemy 0.7.8 doesn't seem to work anymore. We worked around this by not using SQLAlchemy anymore for our antique MySQL 3.23 db, but this means you could also delete 3.23 from the supported database list. As a solution you could maybe add a try/except clause around it with a version check just like you do in do_commit() in dialects/mysql/base.py Best regards, Ids -- You received this message because you are subscribed to the Google Groups sqlalchemy group. To view this discussion on the web visit https://groups.google.com/d/msg/sqlalchemy/-/YfpYDdO8XCwJ. To post to this group, send email to sqlalchemy@googlegroups.com. To unsubscribe from this group, send email to sqlalchemy+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en.
[sqlalchemy] delete multiple objects
Hello, I'm using the declarative extension and was wondering (after searching the docs) how to elegantly delete multiple objects at once, preferably without loading them in first. Suppose you've got a mapped class like this: class Person(Base): __tablename__ = 'persons' stamp = sa.Column(sa.DateTime) ... Now you could delete multiple objects like this (I think): Person.__table__.delete().where(stamp somedate) or like this with loading all objects: persons = session.query(Person).filter(Person.stamp somedate) for person in persons: session.delete(person) Is there something like session.delete(Person).filter(Person.stamp somedate) ?? i.e. something similar to the session.query construct but instead of selecting performing a delete? Regards, Ids --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to sqlalchemy+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Re: set_shard problems
Excellent work! Indeed r5335 did fix the problem. Thank you very much for your help. Regards, Ids --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Re: set_shard problems
Just tested with SQLAlchemy 0.4.8 and that one works fine (see log below). So something relevant must have changed between 0.4.8 and 0.5.0rc1. Regards, Ids 2008-11-27 14:06:51,941 INFO sqlalchemy.orm.mapper.Mapper: (Person| persons) _compile_property(id, Column) 2008-11-27 14:06:51,942 INFO sqlalchemy.orm.mapper.Mapper: (Person| persons) _compile_property(name, Column) 2008-11-27 14:06:51,947 INFO sqlalchemy.orm.mapper.Mapper: (Person| persons) Identified primary key columns: ColumnSet([Column('id', Integer(), table=persons, primary_key=True, nullable=False)]) 2008-11-27 14:06:51,947 INFO sqlalchemy.orm.mapper.Mapper: (Person| persons) constructed 2008-11-27 14:06:52,047 INFO sqlalchemy.orm.mapper.Mapper: (Person| persons) __initialize_properties() started 2008-11-27 14:06:52,051 INFO sqlalchemy.orm.mapper.Mapper: (Person| persons) initialize prop id 2008-11-27 14:06:52,052 INFO sqlalchemy.orm.strategies.ColumnLoader: register managed attribute id on class Person 2008-11-27 14:06:52,052 INFO sqlalchemy.orm.mapper.Mapper: (Person| persons) initialize prop name 2008-11-27 14:06:52,049 INFO sqlalchemy.orm.strategies.ColumnLoader: register managed attribute name on class Person 2008-11-27 14:06:52,049 INFO sqlalchemy.orm.mapper.Mapper: (Person| persons) __initialize_properties() complete 2008-11-27 14:06:52,050 DEBUG root: QUERY 1: SELECT persons.id AS persons_id, persons.name AS persons_name FROM persons ORDER BY persons.id LIMIT 1 2008-11-27 14:06:52,078 INFO sqlalchemy.pool.QueuePool.0x..74: Created new connection _mysql.connection open to 'localhost' at 829a20c 2008-11-27 14:06:52,078 INFO sqlalchemy.pool.QueuePool.0x..74: Connection _mysql.connection open to 'localhost' at 829a20c checked out from pool 2008-11-27 14:06:52,083 INFO sqlalchemy.engine.base.Engine.0x..94: BEGIN 2008-11-27 14:06:52,084 INFO sqlalchemy.engine.base.Engine.0x..94: SELECT persons.id AS persons_id, persons.name AS persons_name FROM persons ORDER BY persons.id LIMIT 1 2008-11-27 14:06:52,081 INFO sqlalchemy.engine.base.Engine.0x..94: [] 2008-11-27 14:06:52,086 DEBUG sqlalchemy.engine.base.Engine.0x..94: Col ('persons_id', 'persons_name') 2008-11-27 14:06:52,087 DEBUG sqlalchemy.engine.base.Engine.0x..94: Row (1L, 'bob') 2008-11-27 14:06:52,091 DEBUG sqlalchemy.orm.mapper.Mapper: (Person| persons) _instance(): identity key (class '__main__.Person', (1L,), None) not in session 2008-11-27 14:06:52,092 DEBUG sqlalchemy.orm.mapper.Mapper: (Person| persons) _instance(): created new instance [EMAIL PROTECTED] identity (class '__main__.Person', (1L,), None) 2008-11-27 14:06:52,092 DEBUG sqlalchemy.orm.strategies.ColumnLoader: Returning active column fetcher for Mapper|Person|persons id 2008-11-27 14:06:52,089 DEBUG sqlalchemy.orm.strategies.ColumnLoader: Returning active column fetcher for Mapper|Person|persons name 2008-11-27 14:06:52,089 DEBUG sqlalchemy.orm.strategies.ColumnLoader: populating [EMAIL PROTECTED] with RowProxy/id 2008-11-27 14:06:52,090 DEBUG sqlalchemy.orm.strategies.ColumnLoader: populating [EMAIL PROTECTED] with RowProxy/name 2008-11-27 14:06:52,095 DEBUG root: QUERY 1 RESULT: [__main__.Person object at 0xb78aa60c] 2008-11-27 14:06:52,095 DEBUG root: QUERY 2: SELECT persons.id AS persons_id, persons.name AS persons_name FROM persons LIMIT 1 2008-11-27 14:06:52,096 INFO sqlalchemy.engine.base.Engine.0x..94: SELECT persons.id AS persons_id, persons.name AS persons_name FROM persons LIMIT 1 2008-11-27 14:06:52,094 INFO sqlalchemy.engine.base.Engine.0x..94: {} 2008-11-27 14:06:52,099 DEBUG sqlalchemy.engine.base.Engine.0x..94: Col ('persons_id', 'persons_name') 2008-11-27 14:06:52,100 DEBUG sqlalchemy.engine.base.Engine.0x..94: Row (1L, 'bob') 2008-11-27 14:06:52,100 DEBUG root: QUERY 2: RESULT: [(1L, 'bob')] --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] set_shard problems
Hello, I think I have found a bug, but I may be doing something wrong. It looks like session.query(class).set_shard(shard_id) does not work and session.connection(shard_id=shard_id).execute does. The first does not return any result, the second one does (even when executing the same query). I've tested it with MySQL 3.23.54 and 5.0.45 and sqlalchemy 0.5.0rc1, rc2 and rc4. Here is the test database setup: CREATE TABLE persons ( id INTEGER NOT NULL AUTO_INCREMENT, name VARCHAR(20) NOT NULL, PRIMARY KEY (id), UNIQUE (name) ); insert into persons (name) values('bob'); insert into persons (name) values('alice'); Here is the test code: #!/opt/python-2.4/bin/python import sys import logging import sqlalchemy as sa from sqlalchemy.orm.shard import ShardedSession from sqlalchemy.orm import sessionmaker from sqlalchemy.ext.declarative import declarative_base logging.basicConfig(stream=sys.stdout, format='%(asctime)s %(levelname) s %(name)s: %(message)s') logging.getLogger('sqlalchemy').setLevel(logging.DEBUG) logging.getLogger().setLevel(logging.DEBUG) Session = sessionmaker(class_=ShardedSession) Base = declarative_base() class Person(Base): __tablename__ = 'persons' id = sa.Column(sa.Integer, primary_key=True) name = sa.Column(sa.String(20), unique=True, nullable=False) def __str__(self): return 'Person(%s, %s)' % (self.id, self.name) def shard_chooser(mapper, instance, clause=None): raise NotImplementedError def id_chooser(query, ident): raise NotImplementedError def query_chooser(query): raise NotImplementedError Session.configure(shard_chooser=shard_chooser, id_chooser=id_chooser, query_chooser=query_chooser) session = Session() shard_id='test' engine = sa.create_engine('mysql://[EMAIL PROTECTED]/%s' % shard_id) session.bind_shard(shard_id, engine) q = session.query(Person).set_shard(shard_id).limit(1) logging.debug(QUERY 1: %s, q) rows = list(q.all()) logging.debug(QUERY 1 RESULT: %s % rows) # # now to it manually: # q = '''SELECT persons.id AS persons_id, persons.name AS persons_name FROM persons LIMIT 1 ''' logging.debug(QUERY 2: %s, q) rows = session.connection(shard_id=shard_id).execute(q) rows = list(rows) logging.debug(QUERY 2: RESULT: %s % rows) And here is the code output: 2008-11-26 10:52:26,043 INFO sqlalchemy.orm.strategies.ColumnLoader: Person.id register managed attribute 2008-11-26 10:52:26,044 INFO sqlalchemy.orm.strategies.ColumnLoader: Person.name register managed attribute 2008-11-26 10:52:26,045 DEBUG root: QUERY 1: SELECT persons.id AS persons_id, persons.name AS persons_name FROM persons LIMIT 1 2008-11-26 10:52:26,061 INFO sqlalchemy.pool.QueuePool.0x...8bf4: Created new connection _mysql.connection open to 'localhost' at 82b02ec 2008-11-26 10:52:26,062 INFO sqlalchemy.pool.QueuePool.0x...8bf4: Connection _mysql.connection open to 'localhost' at 82b02ec checked out from pool 2008-11-26 10:52:26,062 INFO sqlalchemy.engine.base.Engine.0x...8a14: BEGIN 2008-11-26 10:52:26,060 INFO sqlalchemy.engine.base.Engine.0x...8a14: SELECT persons.id AS persons_id, persons.name AS persons_name FROM persons LIMIT 1 2008-11-26 10:52:26,064 INFO sqlalchemy.engine.base.Engine.0x...8a14: [] 2008-11-26 10:52:26,066 DEBUG sqlalchemy.engine.base.Engine.0x...8a14: Col ('persons_id', 'persons_name') 2008-11-26 10:52:26,070 DEBUG root: QUERY 1 RESULT: [] 2008-11-26 10:52:26,070 DEBUG root: QUERY 2: SELECT persons.id AS persons_id, persons.name AS persons_name FROM persons LIMIT 1 2008-11-26 10:52:26,071 INFO sqlalchemy.engine.base.Engine.0x...8a14: SELECT persons.id AS persons_id, persons.name AS persons_name FROM persons LIMIT 1 2008-11-26 10:52:26,071 INFO sqlalchemy.engine.base.Engine.0x...8a14: {} 2008-11-26 10:52:26,073 DEBUG sqlalchemy.engine.base.Engine.0x...8a14: Col ('persons_id', 'persons_name') 2008-11-26 10:52:26,073 DEBUG sqlalchemy.engine.base.Engine.0x...8a14: Row (1L, 'bob') 2008-11-26 10:52:26,074 DEBUG root: QUERY 2: RESULT: [(1L, 'bob')] There are two things I notice in the sqlalchemy.Engine logs; the second SELECT statement seems to have an additional newline and the next log (which seem to be the parameters for the select statement) contain a {} instead of a []. Am I doing something wrong here or is this supposed to work? Regards, Ids --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---
[sqlalchemy] Changing loglevels
Hi, I'm using SQLA 0.5.0rc2 and am running into a problem. I've got a piece of code like this: import sys import logging import sqlalchemy as sa from sqlalchemy.orm import sessionmaker logging.basicConfig(stream=sys.stderr) logging.getLogger('sqlalchemy.pool').setLevel(logging.ERROR) engine = sa.create_engine(.) logging.getLogger('sqlalchemy.pool').setLevel(logging.INFO) In the example above I won't see any INFO log messages. However, if you change the first setLevel from logging.ERROR to logging.INFO you _do_ get INFO log messages. It looks like you have to specify a loglevel below logging.WARN *before* creating the engine. Otherwise I can't get any logging to work. Maybe it's my lack of understanding the logging module, but does someone know what is going on? Regards, Ids --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups sqlalchemy group. To post to this group, send email to sqlalchemy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en -~--~~~~--~~--~--~---