[SNMP4J] getPeerAddress?

[Frank Fock]

Hi Mathias,

Thanks for your suggestions!

Changing Target from abstract class to an interface is a good idea.
If no one has a strong opinion against it, then it will be in release 1.2.
The change should not have any consequences to existing code if
one has not subclassed Target directly yet.

What would like to have instead of the getPeerAddress() method?
The address is provided by the transport. The ResponseEvent
also provides the userObject which can be used to hold the target
used to send the request.

We would, of course, let other developers work on it - however
up to now we have not received any serious offers. For the 1.2 release
there will be a RC version so that mailing list members can comment
on it. In addition there will be also several alpha releases for  the
SNMP4J-Agent API. The first one should be released end of March.

In general, I appreciate any suggestions for improvements and I will
do my best to implement them.

Best regards,
Frank

Mathias Bogaert wrote:

> Hi,
>
> Why was the choice for getPeerAddress() made? Now one still has to go 
> and construct a Target (with its own specific properties) in order to 
> implement a request-response-request-response chain. Also the fact 
> that org.snmp4j.Target is a concrete class bothers me. Make it an 
> interface and let the applications implement it (e.g. population from 
> a database). Let CommunityTarget and SecureTarget extend a 
> TargetSupport or AbstractTarget class, which implements Target.
>
> Will SNMP4J's development process be torn open in order to let other 
> developers work on it? Will there be beta and RC versions in the 
> future? I'd like to see a more 'open source' approach to the development.
>
> /rantmode
>
> Mathias Bogaert
>
> _______________________________________________
> SNMP4J mailing list
> SNMP4J at agentpp.org
> http://lists.agentpp.org/mailman/listinfo/snmp4j
>