MCache is on beta 6, but what is a beta? While a lot of projects have differing views on what constitutes a beta, our view of a beta is that it is "fully functional" and should be usable, but may have a few small issues. In fact, it doesn't get put up as a beta unless it is "in use" at Mohawk Software.

This web site, for instance, always uses the latest beta of MCache. So, simply by using this site, you are providing useful QA data.

MCache Beta 6 with Java support has been released. Some changes have been made, so here's a list:

The java/ directory contains code and classe to interface to MCache from Java. There is a new interface class that should work with Java's local HttpSession strategy.

The C/C++ API has been moved to its own directory, api/. A couple new objects, MCacheInterface and MCacheBase, mirror the Java class API. MCache_conn has been renamed to MCacheConnector to bring it in line with the Java API. (This should not be a problem as the C++ interface has been largely undocumented.)

The PHP API is basically unchanged, but provides a "Classname" during PS_READ_FUNC(). Squirrelmail assumes that it can lock a session that has been deleted.

While Beta5 was destined to be the release candidate, Java support became a requirement, and with Java support came some interesting changes to the locking system and two new API functions, LockObject and UnlockObject, the two APIs deal with both the data string and name/value pairs in the object, additionally, session versioning has been added so that Java's built in local cache didn't need to recreate the object each time. If the local version of the object is the same as the MCache version, then no data is sent from the server.

There are some issues to resolve between the PHP and Java in

This week, before we spin the release candidate, a small API change will happen to the locking calls, and two new locking calls will be added.

If you've checked out the source code from CVS you'll notice that there is a new directory called "java," yes MCache will be supported under Java. While a proper HttpSession interface will take a while, the underpinnings for it should happen sooner rather than later.

The issue to be "fixed" is an optimization of the locking protocol which will allow an MCache client to locally cache their data. This will allow the Java HttpSession to use it's own internal cache when creating sessions, and only report changes. This could also lead to a way to optimize PHP session handling.