On Monday 04 October 2004 01:29, Stefano Mazzocchi wrote:
So, what do you guys think: should gump metadata still reside in xml files or should it reside in a database and we have a web application on top that takes care of managing it?
What is known as Magic, an Ant 1.6 extension system with a strong project model similar to Maven, generates the current Gump XML from the project definition. Currently, this generation has to be run whenever the model changes (which sometimes is forgotten), instead of just handing Gump the model "on demand".
What would be really cool, is that Gump could query each module for the meta data somehow, so that in case of Magicthe 'intermediary' step of creating the xml of today could be skipped.
well, I'm sorry but I don't get it. Gump *is* querying the metadata for each module. If you choose to encode your data in a different way and need to transform it to the gump markup format, that's your problem.
But I have the impression I'm missing your point.
That would also mean that the current xml, new DB way, Magic, Maven and possibly other ways can co-exist.
hmmm, that's interesting.
In effect, it could be a 'pre-stage' to the 'build-stage' and that a DB connection is handed over to some generator/updater.
say more, I'm not sure I follow you entirely.
-- Stefano.
smime.p7s
Description: S/MIME Cryptographic Signature
