[ZODB-Dev] Re: Programaticallly managing the cache in stand alone ZODB

Chris Cioffi evenprimes at gmail.com
Tue Jan 25 13:28:30 EST 2005


Ignore this.  get_transaction().abort() works much better when placed
in the appropriate scope.  Doh!

Chris


On Tue, 25 Jan 2005 11:59:01 -0500, Chris Cioffi <evenprimes at gmail.com> wrote:
> Hello all,
> 
> I have a program using stand alone ZODB.  Currently it's a bit more than 2GB.
> 
> After normal processing, I go through and extract data for reports.  I
> don't modify the data, and generally process the data in a single
> pass.  Once used, the data is not accessed again during the run.
> 
> Is there a way to programmatically say "flush the cache, I don't need
> anything currently there?"
> 
> get_transaction().abort() doesn't seem to quite do this (maybe I need
> to make a setting change?)
> 
> I'm running v3.2.5 if that makes a difference.  Thanks!
> 
> Chris
> --
> "It is our responsibilities, not ourselves, that we should take
> seriously." -- Peter Ustinov
> 


-- 
"It is our responsibilities, not ourselves, that we should take
seriously." -- Peter Ustinov


More information about the ZODB-Dev mailing list