Serge Knystautas wrote: > > > Steve Brewin wrote: > > Being pass� is no reason of itself to "change the Mailet > API". I'm sure I'm > > being paranoid, but you don't mean drop the current Mailet > API do you? > > Just that there are much better lifecycle and configuration options > available now that IoC is better understood and available. > > > We could come up with a new Mailet API and deprecate the > current one, or > > enhance the current one in a backwards compatible manner > (which would most > > likely be no less pass�). It would be a dereliction of duty > to simply > > "change the Mailet API" to something new and funky leaving > current Mailets > > stranded. > > > > If we do want a new Mailet API, as long as the parameters > passed to a POJO > > DI class include those required by the current Mailet API, > it would be > > simple enough to maintain backwards compatibility via an > adapter in a POJO > > DI world. > > My preference would be to keep the name (mailets), put it in a new > package, call it 2.0, and have James support mailet/matchers > using API > 1.0, and convert all the mailets we package to 2.0.
My thoughts in a nutshell! -- Steve --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
