[infinispan-dev] StateTransferControlCommand initialisation?

[Galder Zamarreno]

Manik Surtani wrote:
> 
> On 25 Mar 2009, at 19:23, Galder Zamarreno wrote:
> 
>> Hi,
>>
>> I'm looking at RemoteCommandFactory in Infinispan as part as mapping the 
>> marshalling stuff to JBoss Marshalling and here's something that has 
>> confused me:
>>
>> fromStream() method says:
>>
>> * Creates an un-initialized command.  Un-initialized in the sense that 
>> parameters will be set, but any components
>> * specific to the cache in question will not be set.
>> * <p/>
>> * You would typically set these parameters using {@link 
>> org.infinispan.commands.CommandsFactory#initializeReplicableCommand(ReplicableCommand)} 
>>
>> * <p/>
>>
>> And then you do the following:
>>
>>         case StateTransferControlCommand.METHOD_ID:
>>            command = new StateTransferControlCommand();
>>            ((StateTransferControlCommand) command).init(rpcManager);
>>            break;
>>
>> Shouldn't ((StateTransferControlCommand) command).init(rpcManager) go 
>> into 
>> org.infinispan.commands.CommandsFactory#initializeReplicableCommand(ReplicableCommand). 
>>
>>
>> initializeReplicableCommand() javadoc says:
>>
>> * Initializes a {@link org.infinispan.commands.ReplicableCommand} read 
>> from * a data stream with components specific to the target cache 
>> instance.
>>
>> The rpcManager seems to be one those components?
> 
> No.  The RPCManager is scoped as GLOBAL:
> 
>     https://svn.jboss.org/repos/infinispan/trunk/src/main/java/org/infinispan/remoting/RPCManager.java 
> 
> 
> Refer to the @Scope annotation on the interface.
> 
> But I understand your confusion here.  We should either:
> 
> 1.  Change the Javadocs on fromStream() to be more explicit, i.e., only 
> dependent components of GLOBAL scope are injected.  NAMED_CACHE scope 
> components are only injected later with 
> CommandsFactory.initializeReplicableCommand().
> 2.  Remove the injection in fromStream(); have 
> CommandsFactory.initializeReplicableCommand() handle all of the 
> injection, of GLOBAL and NAMED_CACHE scoped components.
> 
> I have no problem with 2, if it makes your integration easier.  In fact, 
> I actually think 2 is probably cleaner.

Yeah, 2 would be perfect for me. I can implement 2 and run the full 
testsuite and see whether all works fine.

If it looks good, I'll create a JIRA so that we keep a separate track of 
this change.

> 
> Cheers
> -- 
> Manik Surtani
> Lead, JBoss Cache
> http://www.jbosscache.org
> manik at jboss.org
> 
> 
> 
> 
> 

-- 
Galder Zamarreño
Sr. Software Maintenance Engineer
JBoss, a division of Red Hat