On 24/03/2008, Rahul Akolkar <[EMAIL PROTECTED]> wrote: > On 3/23/08, Phil Steitz <[EMAIL PROTECTED]> wrote: > > On Sun, Mar 23, 2008 at 4:07 AM, James Carman > > <[EMAIL PROTECTED]> wrote: > > > On Sun, Mar 23, 2008 at 2:23 AM, Phil Steitz <[EMAIL PROTECTED]> wrote: > > > > Right. I would not put the to-be-voted-on candidate there, just the > > > > RCs leading up the the final, all of which have "RC" in their > version > > > > specs. > > > > > > From what I understand, we're not supposed to cut release candidates > > > with "rc" in their version numbers. If you're going to cut a release > > > candidate, then it's going to be up for a vote. That's why the > > > version says something like "1.0" so that those exact bits can be > > > deployed. At least, that's the way it was described to me when doing > > > proxy. > > > > > > > We have to VOTE on the final bits. It is perfectly fine - and IMO > > advisable - to make RCs available for review prior to final VOTE. The > > only hard and fast rule is that we vote on the final bits. Partly > > for the reason that people's local repos end up with integrity > > problems, I think it is a bad idea to have final version specs in > > candidates used for testing and validation. One of the best things > > about the Maven pom and repo structure is that got us away from > > "commmons-foo.jar" naming and enforced the discipline that artifact > > names, built into jar names, are unique and defining. Even just among > > the development community we should try to stick to that, IMO. > > > > I may be odd man out here, but I see no reason that we should force > > everyone to stop creating RCs for inspection prior to VOTE. If I have > > to change the names to SNAPSHOT everywhere to make people happy, I > > will do that, but as an RM I am not going to produce a sequence of > > "candidate" jars all with the release name unless something really bad > > surfaces in the final VOTE. > > > > <snip/> > > Thats perfectly fine, IMO. > > Strictly from the perspective of a simpler m2 build, your offer of > using "SNAPSHOT" (in addition to "RC") -- exact version details TBD -- > will work very well. That way, (the [B] equivalents of) ... > > mvn -Prc deploy > > ... will: > > * deploy to the m2-snap-repo for RCs with the above versioning > scheme, which is what you'd prefer for the earlier RCs > > * deploy to a pao/builds special repo for the "final" RCs, which is > what everyone should(!) prefer > > And for those creating all RCs with final versions, the first bullet > doesn't hold. > > That way, on the topic of which style we should recommend to RMs, this > thread (and subsequently, the build) can acknowledge both and remain > agnostic. >
I think someone needs to turn this thread into a document so it does not get lost ... > > -Rahul > > > > > > > Phil > > > > > > --------------------------------------------------------------------- > 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]
