I think it would be wrong to think in terms of controlling CFC
serialization, but might be better to think in terms of "storing CFCs
in a persistent store", which may or may not be implemented via
serialization.

For example, the Java Data Objects (JDO) spec defines annotations to
Java objects to manage persistence, but doesn't define what the
persistence mechanism is (it could be a relational database, or an
object database, or the GAE datastore, etc.). We should do the same
thing for CFCs. I've never looked at any of the CFC ORMs (Transfer?
Reactor?) but planned to looked at these for ideas. Also, I've heard
CF9 (Centaur?) is implementing some sort of Hibernate integration, so
maybe there are some ideas to be gleaned there.

The main idea is that the "persistence annotations" for CFCs should be
compatible and produce similar results across different
implementations, whether that be an ORM, the GAE datastore, or some
other future yet-to-be-determined persistence mechanism. This will
help in portability of CFML applications, and could also help in
migrating data from one form of datastore to another.

It seems to me this is an area where the CFML Advisory Committee could
really add value.

Vince

On Jun 2, 2:34 pm, "Peter J. Farrell" <[email protected]> wrote:
> [email protected] said the following on 06/02/2009 12:18 PM:
>
> > Yes, the current implementation will probably end up being the default
> > behavior: serialize everything, and make all simple variables in the
> > "this" and "variables" scopes searchable. Later we can add some
> > mechanism to limit and control this; adding new attributes to
> > CFPROPERTY seems like a good candidate (you've newly joined the CFML
> > Advisory Committee, right? maybe you could run this by them?).
>
> Yes, I am that person (goans please ;-).  Anyways, I know if I propose
> something I should have an idea on syntax because that will be the first
> question that will be asked.  Any ideas on that?
>
> I was also thinking that Adobe CF allows for serialization /
> deserialization of objects to a byte stream -- albeit it's limited and a
> bit buggy (and no limitation on what is serialized.  I'm wondering if
> Open BD has the same functionality or will do so in the future?  I'm
> thinking that the same cfproperty flag or directive for serialization
> should work the same for the Google Datastore and
> serialization/deserialization of objects to / from byte streams.
>
> Best,
> .Peter
--~--~---------~--~----~------------~-------~--~----~
Open BlueDragon Public Mailing List
 http://groups.google.com/group/openbd?hl=en
 official site @ http://www.openbluedragon.org/

!! save a network - trim replies before posting !!
-~----------~----~----~----~------~----~------~--~---

Reply via email to