Hi Brian
On 17/12/2008, at 10:03 AM, Brian Aker wrote:
On Dec 16, 2008, at 3:25 PM, Roland Bouman wrote:
I fail to see how this has got anything at all to do with
transactions. I mean, NOT NULL is a constraint - no more, no less.
Historically the answer always was "no matter what to not let MyISAM
fail midway through a statement". So a lot of logic in the server
was "muddle through".
So it has nothing to do with transactions, but instead it has to do
with MyISAM behavior.
Goodness, not you too.
Of course it has everything to do with transactions, or rather the
lack of them in the original storage engines: MyISAM and HEAP.
The non-transactional nature (behaviour) of these engines necessitated
having some kind of coping mechanism.
It's not about any engine in particular, it's about architectural
aspects in those engines that make the additional logic necessary.
And apart from just knowing my MySQL history (and heck we can ask
Monty, too!), my hypothesis can be tested:
- If MyISAM had been transactional, there would have been no reason
for this kind of logic.
- HEAP has the exact same issues, so why would we hang all the blame
on MyISAM?
Cheers,
Arjen.
--
Arjen Lentz, Director @ Open Query (http://openquery.com.au)
Training and Expertise for MySQL and related tools
OurDelta: free enhanced builds for MySQL @ http://ourdelta.org
_______________________________________________
Mailing list: https://launchpad.net/~drizzle-discuss
Post to : [email protected]
Unsubscribe : https://launchpad.net/~drizzle-discuss
More help : https://help.launchpad.net/ListHelp
