Jim wrote:
Based on some of the deployment work that will be done, we actually need to separate out the load phase into two, one which loads the assembly model and another phase which resolves the dependencies necessary to construct the component type metadata (e.g. Java class, WSDL, etc.). I think we should pick this work up once we have those phases separated out.
Jim or Jeremy, could you guys please let me know what's the status of this separation ? I'm in need of this for the contribution services, to be able to process a SCDL without necessary having to load dependencies on the first pass. -- Luciano Resende http://people.apache.org/~lresende On 1/17/07, Jim Marino <[EMAIL PROTECTED]> wrote:
On Jan 17, 2007, at 9:09 AM, Jeremy Boynes wrote: > On Jan 17, 2007, at 8:52 AM, Jim Marino wrote: > >> >> Based on some of the deployment work that will be done, we >> actually need to separate out the load phase into two, one which >> loads the assembly model and another phase which resolves the >> dependencies necessary to construct the component type metadata >> (e.g. Java class, WSDL, etc.). I think we should pick this work up >> once we have those phases separated out. > > Expanding on this a bit, one problem we have now is that we require > the resource loading mechanisms work during the load phase. > However, now we've added the <dependency> mechanism and are looking > at things like OSGi with its bundle dependency mechanisms we need > to set up these resource dependencies in order to get the > information needed for introspection (e.g. being able to load the > class or WSDL document). > > We may actually need to add two phases: > 1) introspection, where we load definitions referenced from the > model and fully flesh out the configuration > 2) validation, where we determine (to the extent we can) that the > planned change to the assembly is legal Sounds good > > I think validation is a more global problem - the load/introspect > phases are creating a complete, detailed proposal for a change to > the global assembly. The validation phase needs to check that the > entire proposal is valid based not just on correctness of the > proposal itself but also correctness in the context of the current > assembly. +1 on generalizing validation. Jim --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
