[CDBI] Re: uniqueness of objects in memory??

Michael Peters mpeters at plusthree.com
Mon Oct 24 21:23:10 BST 2005

Phil Mitchell wrote:
> On 10/24/05, Edward J. Sabol <sabol at alderaan.gsfc.nasa.gov> wrote:
>>Perrin wrote:
>>>All good guesses, but none of them apply here.  If you load an object of
>>>the same class with the same primary key twice, both references should
>>>point to the same object, so you should see changes in both places with
>>>no update() calls needed.
>>That's very true, but I guess I'm not yet convinced that's exactly what Phil
>>is doing/experiencing though, seeing as how we haven't actually seen his code
>>and can't say for certain. What he's describing sure sounds symptomatic of
>>the classic "I forgot to call update() or dbi_commit()" problem to me.
> No, I'm definitely calling update(), auto_commit is on, and I've not
> monkeyed with db_main. I can see the updated values in the DB,
> Nonetheless the old values appear to be cached via the has_a r'ship
> (user->prefs).

Probably not the case, but just thought I'd throw it out there anyway:
any possibility that the browser is caching the page if you are issuing
a redirect. IE has a pretty aggressive caching policy.

Michael Peters
Plus Three, LP

More information about the ClassDBI mailing list