Re: [Ejbca-develop] SCEP changes

[Tomas G. <to...@pr...>]

Hi,
The reason for the responceClass parameter to createCertificate() is 
that we want any kind of request message combined with any type of 
response message. It is up to the caller to decide which response 
message he wants. The response to a PKCS#10 request message can be of 
many types, so the desired implementation class is sent as parameter. 
The whole idea is that RSASignSessionBean does not need to be aware of 
the kind of reply message that is generated.

The string for the implementation class can off course be put as an 
attribute in the deployment descriptor of the caller. In SCEP this would 
be web.xml in apply.war.

What is the problem with this parameter?

Cheers,
Tomas


Jon Barber wrote:

> Dear All,
> 
> I've started work on the SCEP stuff and found one NPE which I've fixed & 
> checked back in.  I've also added some error conditions for the failinfo 
> in response messages.  I've got some questions on the actual structure 
> of the code I've been working with.
> 
> So far I've only changed se.anatom.ejbca.ca.sign.RSASignSessionBean and 
> se.anatom.ejbca.protocol.IResponseMessage.  However, I'm planning on 
> changing the possible constant values in IResponseMessage to enumerated 
> types.
> The thing that puzzles me the most is the responseClass parameter to 
> RSASignSessionBean.createCertificate(). This is passed in and then an 
> instance is created and passed back as the return value.  Could this be 
> changed to a factory of some sort, or even an attribute in the 
> deployment descriptor ?
> 
> Apologies if the question is dumb as I'm not too familier with EJB 
> development.
> 
> Regards,
> 
> Jon.