[sqlalchemy] Re: programming style | using sessions

2009-01-09 Thread Nebur
Right. There's not only the question about the session lifetime. "Where do happen the session modifications" seems a non-trivial question with large applications. I've tried a layer architecture. There's exactly one layer knowing of [and modifying] the session. But that architecture is broken in a

[sqlalchemy] Re: programming style | using sessions

2009-01-07 Thread az
i was using the antipattern of one session forever (huge non-web-app). the pros are - caching; the cons are longer: - usual cache stuff - changes to the obj in the DB won't come unless i expire it - any changes to anything keep accumulating and hence get flushed() at unexpected times. in gene