Maven and snapshots?

I'm working on a rapid project, of which, I'm directly working on a module that is continuously changing. Others have a direct dependency on the module, and as such, I stubbed out the public interface and deployed it to our local Nexus repository as 0.0-SNAPSHOT for use.

Now that I've almost completed my first iteration of the module, I've attempted to redeploy the updated artifact. Reading about snapshots, others claim that a snapshot should represent the current head trunk. Is this true?

Maven automatically increments my snapshot version upon redeploy - so, going from 0.0-SNAPSHOT, I'm still at 0.0-SNAPSHOT, only, it's iteration 3 or 4 of the same snapshot. When should I roll over to 0.1-SNAPSHOT? Is there a plugin I can use to automate the 开发者_开发百科version change, instead of manually editing my pom?

After integration testing and the deployment of our first system release, 1.0-RELEASE, how should my module progress? Should I move my module to 1.0-SNAPSHOT and continue thereon? Is there a methodology one should follow, or is it left to the discretion of the developer?

---

The X.Y.Z-SNAPSHOT notation identifies temporary versions leading up to release X.Y.Z, so you usually do not move from X.Y.Z-SNAPSHOT to X.Y.Z+1-SNAPSHOT unless you release X.Y.Z. If you adhere to this convention the maven-release-plugin may help you with the full release process.

Note that the most common Maven convention uses 3-number release identifiers without any suffix (i.e. no -RELEASE). Suffixes are usually used to distinguish variants of the same release.

Maven is all about conventions, so there's little chance you'll go very far without reading about it: This book is a good starting point.

---

For updating your POM version without manually updating, you may take a look at Maven Release plugin (Although I wrote myself a little script to do the POM update as I find Release plugin don't fit that good in my work flow)

It then come to your version number issue. It is more a release procedure issue. Normally a planned release is denoted by increment of Major or Minor version in the version number. SNAPSHOT version denote that certain release is in progress. For example, I will prefer doing something like this for your case:

Assume I am planning to release first iteration as 0.1, then I will make my head trunk in SCM (e.g. trunk in SVN) with 0.1-SNAPSHOT as version. Which denotes that all development is in fact contributing to release of version 0.1. Upon finish, I'll update the POM version from 0.1-SNAPSHOT to 0.1, perform an actual release of version 0.1 (including release branching, tagging, deploying the artifact), and then change the POM version to SNAPSHOT of next planned release (for example, 0.2-SNAPSHOT).

Similarly, after releasing 1.0 (or 1.0-RELEASE in your example), POM version in head trunk should then be updated to snapshot of your next release version, for example, 1.1-SNAPSHOT.

Just bear in mind that there should no longer be SNAPSHOTs of certain version, if that version is actually released.

---

Understand the way maven interprets SNAPSHOTS is to clear any doubt you can have.

Extract from http://books.sonatype.com/mvnref-book/reference/pom-relationships-sect-pom-syntax.html

For example, if your project has a version of “1.0-SNAPSHOT” and you deploy this project’s artifacts to a Maven repository, Maven would expand this version to “1.0-20080207-230803-1” if you were to deploy a release at 11:08 PM on February 7th, 2008 UTC

First, Let me suggest to stick on maven conventions and change your version to 0.1 as maven archetype:generate propose.

So SNAPSHOT helps others to stay up-to-date easily with your active project. On every compilation, their projects will check new releases of SNAPSHOTS dependencies (based on that pseudo-datetime-version they have on its .m2 directory).

When you finish work on 0.1-SNAPSHOT you deploy an 0.1 and start a 0.2-SNAPSHOT or 1.0-SNAPSHOT