Questions about the pluto source code.

[Zhong ZHENG]

Hi pluto developers,

i am reading the pluto source code, and here are some questions.

1, Container service interfaces are defined in the SPI. According to the doc, LogService is mendatory. The goal of the LogService is to create Logger instances for components, and i found that the Logger interface extends ContainerService too. That structure seems a little strange to me. I commented the "extends ContainerService" in the Logger class and rebuild pluto, and everything goes on well. So, is it necessary that Logger must extends ContainerService?

2, The pluto container defines a set of Factory interfaces to create certain classes, such as classes in the "javax.portlet" package. In the FactoryManagerService interface, the following method is defined:

public Factory getFactory(Class theClass);

The method signature seems to mean that: get a factory instance that i can use to create the specified class (theClass). For example, pluto will call FactoryManagerService.getFactory(RenderRequest.class) to get a render request factory instance which may be used to create RenderRequest instances. But...

2.1 In the "org.apache.pluto.util.ObjectIDAccess" class, pluto calls FactoryManagerService.getFactory(ObjectIDFactory.class) to get an ObjectIDFactory instance, which may be used to create ObjectID instances. Is it more logical to get the ObjectIDFactory by calling FactoryManagerService.getFactory(ObjectID.class)?

2.2 In the "org.apache.pluto.om.ControllerObjectAccess" class, pluto calls FactoryManagerService.getFactory(ControllerFactory.class) to get a ControllerFactory instance, which may be used to create Controller instances. Is it more logical to get the ControllerFactory by calling FactoryManagerService.getFactory(Controller.class)?

Thanks in advance for your reply.

Regards