On Jul 2, 2006, at 1:07 PM, Tzahi Fadida wrote: > > I am a little confused about your answer. My main concern though, > is that > the result of the implicit and explicit transactions to the > database will be > the same as if i executed flush() after each query/insert/delete. > I.e. this seems a little non-deterministic, isn't that risky? > 10x.
i think flush()'s behavior is very deterministic. like i said, it might still be relying on dictionary ordering for the order of UPDATE/ DELETE statements within rows that arent dependent on each other inside a single table, which is something that will eventually be fixed; and as i also said i wont consider that "guaranteed" until I work up test cases that positively assure its true, which is the bigger job. i doubt youd be able to observe it actually doing the same statements with two different orderings right now, and id be pretty surprised if you can get it to deadlock itself based on two flush()es (especially if youre setting transaction isolation lower). i think your best bet is to try it out and watch how it organizes statements. if youre really building something with zero tolerance for failures, something like Hibernate is much more mature; as SA is hardly even beta, i cant say very much is "guaranteed" about it right now. Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Sqlalchemy-users mailing list Sqlalchemy-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/sqlalchemy-users
