Jose,
We are getting ready to use commons-chain in a production system. I think it works quite well for certain problems. You'll want to make sure you have the latest build because we just added a few things to it yesterday (also the javadoc on the commons-lang site does not mention these new classes.)
I think the main drawback to the chain of responsibility (COR) pattern is the lack of type safety (and transparency) that pops up in a typical implementation. As you mentioned, commons-chain requires everything to be "jammed" into a context.
I have a few thoughts on how I plan to use commons-chain after considering the pros and cons. You can take 'em or leave em.
I think chain is most appropriate when you can use it with only a limited number of objects in the context. My application deals with documents (which are beans with lots of properties, some of which are beans themselves.) So when a user wants to save changes to a document, I have a chain for "saving". Each command in the chain can check the part of the document that it is interested in and handle changes as it sees fit. I've got one or two other objects in the context (such as "user") but its mostly just document. None of my commands add anything or remove anything to the context because there is no assumption about what command is next in the chain.
I also have a chain for reading a document from the database. I create an empty document object and put it into my context and then kick off the chain. Each command populates the portion of the document its concerned with. I don't really like the idea of my commands "returning" information through the context. I limit myself to modifying at most one object in the context and make it clear that the chains only function is to modify that document (in the read case) or to read that document and perform some function (in the save case.)
Well there's some ideas for you.
HTH, sean
Jos� Antonio P�rez Testa wrote:
Hello all,
I'm evaluating commons-chain to implement some sort of components.
The idea is to have components fire events and those events be processed in a xml configurable chain of execution.
This schema allows me to insert processing commands in the flow of execution dinamically and change from SimulatorExecutor to RealExecutor, for example, or to insert a LoggerCommand that writes a detailed log of the current event.
In order to give flexibility to the system I have to put almost everything in the Context of the Command Chain, that is implemented as a Map. This way leads to a lack of robustness and ellegance in the overall implementation.
Anybody has faced the same dilema ??
-------------------------------------------------------------------------------------------------------------------
Este correo electr�nico y, en su caso, cualquier fichero anexo al mismo, contiene informaci�n de car�cter confidencial exclusivamente dirigida a su destinatario o destinatarios. Queda prohibida su divulgaci�n, copia o distribuci�n a terceros sin la previa autorizaci�n escrita de Indra. En el caso de haber recibido este correo electr�nico por error, se ruega notificar inmediatamente esta circunstancia mediante reenv�o a la direcci�n electr�nica del remitente.
The information in this e-mail and in any attachments is confidential and solely for the attention and use of the named addressee(s). You are hereby notified that any dissemination, distribution or copy of this communication is prohibited without the prior written consent of Indra. If you have received this communication in error, please, notify the sender by reply e-mail
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
