On Wed, Mar 5, 2008 at 2:26 PM, Matt Benson <[EMAIL PROTECTED]> wrote: > My component ideas: > > * I am still interested in incubating Morph as another > shot at Commons-convert, though I'm not sure how it > will be received given its overlap with BeanUtils, to > which Niall has been giving quite some attention > lately.
It'd be interesting to know if Niall still thinks BeanUtils has lots of life. I was pondering MethodUtils the other day and how it 'should' be in Lang [I was looking for invokeStaticMethod in Lang and after a while realized it was in Beanutils]. +1 to the idea of Morph. It's gone further than Convert did. > * I am for all intents and purposes clear to release a > flat file processing library I have developed for my > employer. I'll follow up with a code dump soon, but > the basic idea is that many Java developers still have > to deal with flat file I/O for integration with legacy > systems, and until the recent release of the Spring > batch stuff I had never seen the problem addressed > with any great amount of thought. My package includes > a DSL declaration language for flat files that is > superficially akin to COBOL copybooks, that being my > language of origin, so to speak. It also implements > Morph's *Reflector interfaces for easy transformation > to/from Java object graphs. Probably replaces a tiny bit I had then :) http://www.osjava.org/genjava/multiproject/gj-core/apidocs/com/generationjava/io/FixedWidthReader.html again, +1. > * Finally, I'd like to create a component whose > principal component is a DSL interpreter for tersely > represented Java object mappings. I haven't finished > planning the entire model, but my intent is to tie an > object parsing context to an arbitrary Map instance. > This would be a from-scratch reimplementation of > another $work project, whose source code I will also > be free to release, but it is my belief that only > certain small pieces of that existing code (probably < > 5 classes) would be necessary for the rewrite. > Because of this I'm not sure if this is more a sandbox > with a grant of some code that -might- be used, or an > incubating project with a beginning "scratch" > codebase. IF we get something along the lines of > Incubator Commons as discussed the distinction might > not be very significant. Grant and Sandbox I think. With an Incubator Commons, I think an expectation would be that there was not lots of development to do for a 1.0. Hen --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
