I'd much rather focus my efforts on the 1.0 branch. Yes, it's taking time, but it has the least chance of suprising us with unexpected issues.
I've seen radical implementation switches cost us lots of time before too, and I don't believe we have anywhere near enough tests to cover the current jelly plugin contract with maven core. -- dIon Gillard, Multitask Consulting Blog: http://blogs.codehaus.org/people/dion/ Jason van Zyl <[EMAIL PROTECTED]> wrote on 05/12/2003 04:45:47 AM: > Howdy, > > Just wanted to share a few ideas about some things I would like to do > over the next couple of weeks. > > I think it is apparent that we are in a bit of deadlock wrt to a release > so I would like to do the following to try and get a stable release out > in the next little while. > > 1. Get maven-model working with 100% coverage (done). > 2. Get maven-model-tools working with 100% coverage. > 3. Get maven-project working with 100% coverage while. > excluding artifact handling code to give Michal some time > to clean up Wagon. > 4. Integrate Wagon into maven-project and get the resulting code > working with 100% coverage. > > After these steps these three components can be wrapped with a plexus > embedder to create a small tool that will take care of the entire Maven > project building concern and that's what I would like to integrate into > Maven for the 1.0 release because we're just floundering at this point. > > In order to use this it means that all notions of Jelly/Ant must be > pushed into PluginManager. I have code in the RC2_REFACTOR which strips > the Ant/Jelly code out of Cli/App classes which I believe will be enough > when combined with the new project builder. > > So at this point I am going to work on the components and then if no one > has beaten me to it I will strip Jelly/Ant out of the necessary places > in order to use the new project builder. > > With the use of the new project builder recursive inheritance will work > on all levels with respect to the model and properties and we can then > focus on the Jelly related stuff. > > I believe it will take less time with this approach as the components I > want to use are cleanly separated and it will be much easier to create > comprehensive tests with Jelly/Ant out of the way. Ultimately this is > the only solution that will work for the long-term and I think it's > obvious our attempts on the various branches isn't working so I would > like to focus on the long-term solution as I think it will actually be > faster and easier to do. > > -- > jvz. > > Jason van Zyl > [EMAIL PROTECTED] > http://tambora.zenplex.org > > In short, man creates for himself a new religion of a rational > and technical order to justify his work and to be justified in it. > > -- Jacques Ellul, The Technological Society > > > --------------------------------------------------------------------- > 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]
