Craig R. McClanahan wrote:
Quoting Simon Kitching <[EMAIL PROTECTED]>:
On Sat, 2004-01-17 at 23:58, robert burrell donkin wrote:
the RSS 0.91 is really just an example but many people find it a very useful one. creating digestions for RSS 1.0 and/or RSS 2.0 isn't really an itch i have but i suspect that if you were to create one and donate it to the ASF then there are probably a lot of folks who'd be grateful...
While on that topic, I would like to move the RSS stuff from the main library to the "examples" section before the next Digester release.
Defintiely +1. But an even stronger +1 on new commons components for "pure data beans" that represent the data of common XML document formats like RSS. For RSS itself, such a set of beans (which covered a reasonable attempt at a union of all the variant formats, plus a way to plug in arbitrary extension) would be *broadly* useful in both client and server applications that manipulate RSS-type data, far beyond the particular scenario of parsing XML->beans (which is what the Digester example does) and the reverse beans->XML direction (which is what an example in Betwixt does).
Craig,
Good idea, but I'd suggest that maybe something like that is a good candidate for development outside of the ASF. I think it is a good idea, but starting a Jakarta Commons project to house "pure data beans" for commons data formats seems outside of the commons charter - you start to think about boundaries, how it would scale, etc.
Robert's idea of moving the RSS stuff to an examples distribution is a good practical first-step. Maybe over the next few months if someone steps up to the plate and donates and/or creates other Digester/Betwixt examples they could be moved into the same example distribution, and once enough momentum has developed we could create a new commons components.
I like the idea, but I think we should take small steps. Encourage people to take initiative, and then see if we need YACC (yet another commons component) once we've got more than "n" rulesets.
Tim
Any objection? And if not, should we bother trying to actually move the ",v" files within the cvs repository to preserve file history, or just "cvs rm" the old and add the current versions in the new location?
Attempts to maintain history in this manner (moving ,v files) does not strike me as the right approach ... the original files will still be in the attic after you "cvs delete" them, and the initial commit to the new directory can include a pointer to where to review the prior history.
I'm not sure if my access rights allow me to do the former myself...
Regards,
Simon
Craig
--------------------------------------------------------------------- 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]
