geronimo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Aaron Mulder <>
Subject Re: J2CA ActivationSpec Data Types
Date Sun, 03 Oct 2004 00:06:51 GMT
Here's what I'm looking at:

The archives are kind of hard to read, but if you look at the September 
traffic, you should see someone asking about whether they need to be 
Strings, and someone from Sun identifying it as a bug that their product 
doesn't handle other data types, and then a note that it's been fixed in 
an upcoming maintenance release of their product.  Whether those Sun guys 
speak for the JSR, I don't know.


On Sat, 2 Oct 2004, David Jencks wrote:
> I wondered.... as far as I can tell the spec says nothing about the 
> types for an ActivationSpec.  Do you have any quotes from the connector 
> list?
> Basically I think this is a good idea if it is spec compliant. It 
> certainly makes the ActivationSpec more useful.  I concluded that it 
> probably wasn't spec compliant from reading the spec, but since I 
> couldn't find a discussion I'd be very happy to be wrong.
> Apparently, looking at the list of allowed types, primitives are 
> explicitly not allowed for ResourceAdapter or ManagedConnectionFactory 
> config-properties, so I'd be a bit reluctant to allow them for 
> ActivationSpecs without a clear statement from the spec committee.
> thanks
> david jencks
> On Oct 2, 2004, at 4:14 PM, Aaron Mulder wrote:
> > 	So part of my previous problem deploying a J2EE connector was that
> > my ActivationSpec class had a property of type Integer.  There's no
> > setting in the DD for the data type, and Geronimo was blowing up during
> > deployment with a ClassCastException.  I think this was recently 
> > discussed
> > on the connector mailing list, with the implication that other data 
> > types
> > should be supported.  I'm thinking we should try to convert the String 
> > to
> > the appropriate data type for the ActivationSpec property if it has a
> > constructor that takes a String (Integer, etc.) or perhaps if it's a
> > primitive.  Any thoughts on this?
> >
> > Aaron
> >

View raw message