From: Jack M. Thompson (JMThompson_at_[hidden])
Date: 2002-01-30 17:09:52
Ya, that's what I'm trying to say. That's one of the main advantages.
You build the program with simple scripts, and only code where you need
to specify actions (the guts of your program). When the development
efforts are complete, and periodically during the dev cycle, you just
compile the scripts, generate some enumerations to be compiled by the C
/ C++ compiler, and load the data. You don't dare change anything. It
should be a finish product.
This concept works.
From: quendezus [mailto:quendez_at_[hidden]]
Sent: Monday, January 28, 2002 9:17 AM
Subject: Re: Persistence Library Needed in Boost? (Was: [boost] A tip
from the Dr.)
--- In boost_at_y..., Emily Winch <emily_at_b...> wrote:
> quendezus said:
> > All these persistence libraries are intrusive (you have to use a
> > class). Some of them don't use serialization. They all heavily
> > on runtime polymorphism. I think a boost persistence library
> > be non intrusive and should use template programming.
> External polymorphism would give you the runtime polymorphism
> intrusiveness. It would be nice to be able to turn that off,
> circumstances where the structure is known at compile time: and
> since "the structure will be known at compile time" sounds like one
> things in requirements documents that gets changed when you already
> everything :)
Yes, I have tried external polymorphism successfully for a
persistence system. I will think about your idea of being able to
switch easily between a compile time mechanism and a runtime one.
Info: http://www.boost.org Send unsubscribe requests to:
Your use of Yahoo! Groups is subject to
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk