On Jun 19, 2008, at 5:57 PM, Martijn Faassen wrote:
> > Hi there, > > I just noticed on the trunk that on sessionmaker(), autocommit > defaults > to False, but autocommit defaults to True when you use create_session. > autoflush is also True for sessionmaker, but for create_session, it's > False. Finally autoexpire is True for sessionmaker, but False for > create_session. > > Is this intentional? > create_session() is the "original" session creation function, which creates a session with none of the "newer" things added on, i.e. autocommit=False, autoexpire, autoflush. This is for backwards compatibility reasons, both within the unit test suite as well as lots of examples, user scripts, etc. The unit tests in particular were built around the notion of an "unadulterated" session that isn't doing things implicitly. The Session's behavior has changed over the course of SQLA's lifespan from being a relatively "dumb" container that doesn't make any decisions whatsoever to one which does what the large majority of use cases would prefer; this is a product of us being around for awhile and watching lots of people use it for a couple of years. I just noticed that the 0.5 docs still have a prominent "create_session()" section, which I'm assuming is how you came across it. I think the solution here at the least would be to remove prominent advertisement of "create_session()". I'm not sure yet about deprecation/non-publicizing it. I still think its a useful function for ad-hoc creation of sessions with no extra behavior but this may very well be just my skewed view of it. --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
