Need help porting Prevayler (Java) to Python

[Patrick K. O'Brien]

> I have briefly looked at the code, it should be easy to do a
straightforward
> port.

Great!

> If I'm correct:
> - you need to be able to keep all your data in memory
> - disk usage [1] is data snapshot size + commands sequence between
snapshots
> - it seems that commands should be executed sequentially
> - in order to keep [1] under control, you need to take snapshots,
snapshots'
> cost is proportional to data in memory,
>  and while you are taking a snapshot you can't do anything else.

Those all look correct, based on my understanding as well.

> What kind of applications have you in mind?

Medium-sized applications that need a simple, but robust, persistence
mechanism without the overhead of a relational database. I think this
persistence system would be useful in combination with Pyro (Python Remote
Objects). I have a SourceForge project set up at
http://sourceforge.net/projects/pypersyst/ and I can give you developer
access to CVS if you like. And, of course, I will be forever in your debt
for any help you can offer.

Thank you,

Pat

--
Patrick K. O'Brien
Orbtech