Author aymeric.augustin
Recipients Jeremy Banks, Mark.Bucciarelli, Ronny.Pfannschmidt, adamtj, asvetlov, aymeric.augustin, dholth, ghaering, monsanto, r.david.murray, scott.urban, torsten, tshepang, zzzeek
Date 2014-04-17.20:43:46
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1397767426.6.0.13137648705.issue10740@psf.upfronthosting.co.za>
In-reply-to
Content
That patch solves the problem, at the cost of introducing an unwieldy API, "operation_needs_transaction_callback".

I'm very skeptical of the other API, "in_transaction". Other database backends usually provide an "autocommit" attribute.

"autocommit" is the opposite of "in_transaction" for all practical purposes. There's only two situations where they may be equal:

- before the first query
- after an explicit commit

Then you aren't in a transaction and you aren't in autocommit. But in these cases, in practice, the question you want to ask is "is the next query going to create a transaction?" (and if not, I may want to create one.)

So the semantic of "connection.autocommit" is much more useful than the semantic of "connection.in_transaction".

While you're there, it would be cool to provide "connection.autocommit = True" as an API to enable autocommit, because "connection.isolation_level = None" isn't a good API at all -- it's very obscure and has nothing to do with isolation level whatsoever.
History
Date User Action Args
2014-04-17 20:43:46aymeric.augustinsetrecipients: + aymeric.augustin, ghaering, Jeremy Banks, r.david.murray, zzzeek, asvetlov, adamtj, dholth, torsten, monsanto, scott.urban, tshepang, Ronny.Pfannschmidt, Mark.Bucciarelli
2014-04-17 20:43:46aymeric.augustinsetmessageid: <1397767426.6.0.13137648705.issue10740@psf.upfronthosting.co.za>
2014-04-17 20:43:46aymeric.augustinlinkissue10740 messages
2014-04-17 20:43:46aymeric.augustincreate