It does make sense... but I'd still like to figure out a way to
circumnavigate the CFINCLUDE issue without rebuilding the app from scratch.
;^)

I'm actually (as I suspect may CFers are) trying to add some CFC-based
features to an existing procedural application.  In the end the session
manager and much of the foundational elements will be CFC-based but the
presentation and content will still be procedurally-based.

So it isn't an "OO" implementation (or at least not fully), but it is a CFC
implementation.  ;^)

Another problem is that CFCs just don't lend themselves all that well to
full OO implementations.  A simple example is that if I use inheritance to
create the components as you suggest CF doesn't allow me to do automated
type-validation.

The application, now, heavily relies on type validation.  If I expect a
component of type "CFC_DepressedPress.Framework.DP_Session" I CAN'T pass a
component of type "CFC_DepressedPress.Framework.DP_Session4SQLServer" - even
if the latter IS an extended instance of the former.

I think my first option (a dedicated broker object instantiated with each
business object) is the "most OO" of the bunch - and does allow for type
validation during composition at the business-logic layer.  But it also
duplicates the methods in both the parent and the broker (unless I want to
call the broker directly whenever I need the method).

I dunno... maybe I'll still give it a try, but the lack of type validation
will be a huge pain the ass.

Jim



----------------------------------------------------------
You are subscribed to cfcdev. To unsubscribe, send an email
to [EMAIL PROTECTED] with the words 'unsubscribe cfcdev' 
in the message of the email.

CFCDev is run by CFCZone (www.cfczone.org) and supported
by Mindtool, Corporation (www.mindtool.com).

An archive of the CFCDev list is available at www.mail-archive.com/[EMAIL PROTECTED]

Reply via email to