On Sunday, September 22, 2013 9:30:06 PM UTC+3, Aymeric Augustin wrote: > > On 22 sept. 2013, at 19:38, Anssi Kääriäinen > <[email protected]<javascript:>> > wrote: > > > There is no explicit use of savepoint=False. It is reasonable to expect > that the code succeeds on databases that allow this coding pattern. > > Yes, you're right… > > I looked at your version of the patch: > > 1) I believe the low-level APIs (rollback and savepoint_rollback) > shouldn't be aware of the high level API (atomic), eg. they shouldn't > manipulate the needs_rollback flag. The user should explicitly manipulate > the flag with set_rollback if he starts messing with the low level APIs. > > I know the low-level APIs are calling `validate_no_atomic_block`, but > that's a bit different. It's making a check and possibly raising an > exception, not changing the internal state of the high level API. > > Yes, this seems like a good idea to me. Atomic.__exit__ should be able to handle this.
2) You had to do many changes to the tests; it looks like some weren't very > well written and your changes are justified. I'd like to double check. > Thanks your the additional test cases, too. > > It must be said that some of the changes were written in bit of a haste - so double check is definitely welcome. - Anssi -- You received this message because you are subscribed to the Google Groups "Django developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/django-developers. For more options, visit https://groups.google.com/groups/opt_out.
