OT: I too am surprised that people can dev an app with ever needing
Are you suggesting that application scoped CFC's shouldn't/don't need
to be initialised on application start? If not, where?
I've found <appinit> most helpful in collecting app initialisation
into Fusebox rather than Application.cfc and I was hoping to do the
same with session initialisation. To me it makes sense to have Fusebox
describe the flow of initialisation rather then having to say "now,
have I set this var/object in Fusebox or Application.cfc????".
On a more fundamental question, should the framework be constrained by
people that may find aspects of it to complex? I'm starting to play
with Reactor and I don't understand it all yet but I'm chipping away
at it and more and more of its functions are being revealed as my
I've not played with Mach II so can't comment on how it handles
integration with Application.cfc but I'd lament the loss of <appinit>.
--- In email@example.com, Sean Corfield <seanc@...> wrote:
> On Feb 26, 2007, at 1:58 AM, dickbob57 wrote:
> > We have a appinit point in globalfuseactions but do people feel there
> > is a need for a sessioninit point?
> Quite a few sites have session scope disabled (I was surprised by
> this - it came up on the Model-Glue list when Joe implemented events
> being carried across requests). That means there is no way to do it
> reliably without generating exceptions - which is something people
> don't like.
> I also think that adding fusebox.sessioninit.cfm and/or a
> <sessioninit> global fuseaction just adds unnecessary complexity -
> and despite the necessity (in some people's minds) of adding a hook
> for application initialization to handle CFC creation etc, a number
> of people have complained about fusebox.appinit.cfm and <appinit>
> adding complexity.
> > I know there is one in Application.cfc but I'd like to have one in
> > Fusebox so that I can use the features of the framework.
> One thing we might look at it finding a way to integrate
> Application.cfc and Fusebox - in a manner similar to Mach II's
> integration. If I can figure out a suitable way to deal with that, I
> would probably deprecate the appinit mechanism introduced in 5.0 (it
> would remain, for backward compatibility, but labeled as "not
> Sean A Corfield -- (904) 302-SEAN
> An Architect's View -- http://corfield.org/
> "If you're not annoying somebody, you're not really alive."
> -- Margaret Atwood