Commits

Author Commit Message Labels Comments Date
jason kirtland
- Be a little smarter about aliased funcs/methods by ignoring func_name
jason kirtland
- Another namespace cleanup tweak, why not.
jason kirtland
- Docstring fix.
jason kirtland
Duh.
jason kirtland
- Removed deprecated append(val, **kw) - dict/set/list proxies are now docstring'd like their python counterparts
jason kirtland
- More uses of exc.NO_STATE
jason kirtland
- Centralized 'x is not mapped' reporting into sa.orm.exc. - Guards are now present on all public Session methods and passing in an unmapped hoho anywhere yields helpful exception messages, going to some effort to provide hints for debugging situations that would otherwise seem hopeless, such as broken user instrumentation or half-pickles.
jason kirtland
- ...and added bind.py into the orm suite
jason kirtland
- Moved an ORM test out of engine...
jason kirtland
- 2.3 compat.
jason kirtland
Refactor-o fix.
Jonathan Ellis
handle null tablespace_name
jason kirtland
- Fleshed out Session.get_bind(), generating a couple todos: [ticket:1053], [ticket:1054], [ticket:1055] - Trotted out util.pending_deprecation, replacing some 'TODO: deprecate's - Big session docstring content edit fiesta - session.py line length and whitespace non-fiesta
Jonathan Ellis
add CHAR to ischema_names map; some minor cleanup
jason kirtland
- Quick cleanup of defaults.py. The main DefaultTest is still a mess.
jason kirtland
Split out a couple true autoincrement/identity tests from emulated-with-sequences autoincrement=True tests.
jason kirtland
Formatting.
Mike Bayer
- changed char_length() to use a fake, neutral "generic function" - assert_compile() reports the dialect in use
Mike Bayer
- zoomark/zoomark_orm seem to work with pool_threadlocal turned off, [ticket:1050] becomes WORKSFORME - fixed probably errenous unique=True checkin on unitofwork.py
jason kirtland
- Implemented generic CHAR_LENGTH for sqlite (-> LENGTH()) - Updated .requires for firebird
Lele Gaifax
Remove some noise from the uow test
Mike Bayer
pool_threadlocal is off by default [ticket:1049]
Mike Bayer
-removed useless log statement (merge garbage?) - clarified autocommit mechanism
Mike Bayer
put a cleanup handler on the "echo" property to try preventing log garbage in the buildbot
Mike Bayer
added ORM version of zoomark, python 2.5 only for starters
Mike Bayer
some order by's failing on the buildbot
Mike Bayer
- added test for threadlocal not supporting begin_nested() - removed query.compile(); use explicit query.with_labels().statement instead - moved statement annotation step upwards from query._compile_context() to outliers from_self()/statement. speeds zoomark.step_6_editing by 16%
Lele Gaifax
The column default has been renamed `server_default` in 0.5
Mike Bayer
begin() pre-issues a flush() in all cases, better fix for [ticket:1046] and allows rollback to work properly with autocommit=True/begin()
Mike Bayer
dont raise assertions when in autocommit mode [ticket:1046]
  1. Prev
  2. Next