I've decided I don't need to run the nerdErg Jira instance any longer since I'm happy enough to just use the github issue tracker for my OS projects. So if you're looking for a place to leave a bug report or suggestion head on over to https://github.com/pmcneil and bobs your uncle.

One Ring 0.7 has just been released featuring the ability to just include plain old Groovy code in the test Closures, letting you access deep map structures in facts, and fixing "Groovy Truth" for facts converted from JSON.

So now you can do this type of thing in tests:-

test(name: [firstName:'Peter', secondName: 'McNeil') {
  println fact.name.firstName.class //for debugging
  assert fact.name.firstName == 'peter'
}

The fix for Groovy Truth comes about because facts sent as JSON would return JSONObject.NULL for non exiting map elements instead of just null. This breaks when clauses that check for the existence of something the Groovy way, e.g.

which would work if you tested it, but would break when called remotely due to the JSONObject.NULL.

The other fix you can see in that last example. Previously you couldn't call name.firstName because it would cause an error when loading the rule, now you can :-)

A name change for Groupie has been on the cards for a little while now for a number of reasons, not least because groupie is a rather overloaded term, and a google search may show up some rather undesirable results. There is also www.groupie.co in this space. So the look for a new name is on the cards, if you have any suggestions ping me on twitter @pmcneil with it.

One Ring 0.6 has been released. It's a minor bug fix and clean up release on top of 0.5. The main improvements are the removal of the hsql database file that was still being created and setting the production logging level to info from debug.

We've just released version 0.5 of One Ring. This is a complete re-jig of how One Ring goes together, gone is the online editing of rules in favour of just editing your rules as files in your IDE or favourite editor and putting them under version control using Git, Mercurial Subversion or any other VCS. Under the hood we've broken out the rule engine into a separate project that produces a simple jar that can be used in your IDE for unit testing the rules (Instructions on the way). This leads to a much simpler deployment and better control over rule updates, including using your existing continuous integration server. ...more