David,
First apologies for accidently taking this thread off the lift group.
It was entirely accidental.
So my question with regards to the scala language that I can't seem to
find an answer for online anywhere is what exactly is an object nested
inside a class i.e. :
class ObjectWrapper {
object obj extends AnyRef{var prop = "prop"}
}
versus a class with a final val that's set to an instance of an
anonymous class i.e. :
class AnonClassWrapper{
final val obj = new AnyRef{var prop="prop"}
}
I've tried to discover some sort of behavioral difference both in
simple tests in the scala console, and couldn't find any. In both
clases obj responds just as if it was a val and not a type. This
analogy towards static classes in java breaks down because a static
class in java is one that doesn't have a containing instance, and yet:
scala> class ObjectWrapper{var outerProp = "prop"; object obj extends
AnyRef{var prop=outerProp}}
defined class ObjectWrapper
lastly, in a desperate attempt to prove to myself that there is an
actual difference I went against your rule and changed a RequestVar
from an object extending RequestVar to a value set to an instance of
an anonymous extention and my lift app ran without any changes in
behavior???
On Aug 4, 10:42 am, David Pollak <feeder.of.the.be...@gmail.com>
wrote:
> In this case, if you change the type/behavior of the SessionVar and/or
> RequestVar, you could mess up code that depends on these.
>
> You also must *always* make RequestVar and SessionVar as object rather than
> val so that they can make the determination of their unique string ID.
>
> I view ProtoUser much like I view Rails' scaffolding... at some point the
> code becomes too heavy to keep subclassing and it's easier to copy the
> MegaProtoUser code into your own User class and go from there.
>
>
>
>
>
> On Tue, Aug 4, 2009 at 10:27 AM, E. Biggs <tacoban...@gmail.com> wrote:
>
> > I have extended Mega*ProtoUser to achieve cookie-based perpetual
> > login.. and it was fairly easy to do except I ended up having to do
> > something hackish simply because there are two private object members
> > in the class that are core to the functionality I've extended.
>
> > Namely:
>
> > private object curUserId extends SessionVar[Box[String]](Empty)
>
> > private object curUser extends RequestVar[Box[ModelType]]
> > (currentUserId.flatMap(id => getSingleton.find(id)))
>
> > I have no idea if this would be an acceptable change, but at least for
> > one user of lift it would make much more sense that in the spirit of
> > *ProtoUser's clear intent for extensibility that these members simply
> > become:
>
> > protected val curUserId = new SessionVar[Box[String]](Empty){}
>
> > protected val curUser = new RequestVar[Box[ModelType]]
> > (currentUserId.flatMap(id => getSingleton.find(id))){}
>
> > Unless I'm not properly understanding what object members actually
> > are? I'm interpreting them as being final vals assigned to instances
> > of anonymous classes (rough tests in the scala console seem to verify
> > this)
>
> --
> Lift, the simply functional web frameworkhttp://liftweb.net
> Beginning Scalahttp://www.apress.com/book/view/1430219890
> Follow me:http://twitter.com/dpp
> Git some:http://github.com/dpp
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Lift" group.
To post to this group, send email to liftweb@googlegroups.com
To unsubscribe from this group, send email to
liftweb+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/liftweb?hl=en
-~----------~----~----~----~------~----~------~--~---
