Knut, I think this is a great idea. I think it would go quite a ways towards seperating the Hivemind xml syntax from Hivemind, which is a good thing. As I recall, the hibernate project did something similar for reasons xml syntax pain.
Richard -----Original Message----- From: Knut Wannheden [mailto:[EMAIL PROTECTED] Sent: Thursday, May 19, 2005 1:50 AM To: [email protected]; Howard Lewis Ship Subject: Re: Using Java code to instantiate services Hi all, I've been thinking quite a bit along these lines lately. Especially since there have been many other questions and comments which seem to be pushing in this direction. What I believe HiveMind needs is a terse API to construct a HiveMind registry. I know we already have an API for defining the constituents of the registry: RegistryBuilder along with ModuleDescriptorProvider and all the o.a.h.parse.*Descriptor classes. IMO this API has a lot of concepts which are specific to HiveMind's XML syntax: schemas, schema rules, conditional contributions (and conditional expressions), and possibly more. The API I had in mind would define interfaces for the central concepts Module, ServicePoint, Implementation, Interceptor, ConfigurationPoint, and Contribution. The XML parsing with the *Descriptor classes would then be using this API. What do you think about this? Should I try to document a change proposal on the Wiki? --knut On 5/6/05, Howard Lewis Ship <[EMAIL PROTECTED]> wrote: > So I had an idea for an alternate way of getting objects instantiated, > that borrows from the Ruby approach. This is an addition to the > traditional HiveMind (XML) approach. > --------------------------------------------------------------------- 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]
