> Ahmed is largely just repeating what we've said, and we've > just been unclear about where the fault lies. We were not > expecting to use another Avalon container besides Phoenix, so > we would just use the broad name Avalon to explain why > reloading isn't available. > > POJO's do not in and of itself support reloading, but it > would give us an ability to use other IoC containers, so of > which might support reloading.
You can already use our avalon components inside PicoContainer and it will be easier to create the needed wrapper to another container than to remove the already present interfaces. Avalon has been splitted in API + Implementation so I think we can safely continue to use its interfaces to describe our own lifecycle. Avalon does not force us to extends a common/basic class so we are currently using POJOs. We are simply using Avalon interfaces to describe our own lifecycle: we can remove the "implements" word from our classes and create specific wrappers for each component*container but I think it will be easier to create the wrapper for avaloninterface*container. Maybe I'm missing something: is my concept of POJO different from what you are proposing? > There was that one experimental offshot, but as you said, > there hasn't been a lot of movement. This is why I'm trying to do just little refactoring steps. IMHO it is the only way to move forward. If we had a few million dollars from IBM to do "james-the-next-revenge" then I would be in favor of a complete rewrite from start :-) Stefano --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
