I don't want to go too much further on the verification and JPar2 main class layout until I know exactly how JPar2 will setup the RS matrix and solve it.
I've fixed a few bugs in the GF Calculator, and added a method to JPar2 to try some repairing. I found that there is a pretty big difference from Plank's paper and the PAR2 spec. I'm spending most of my time trying to figure out the best way to setup the matrix.
The first set of code has been commited to CVS. The API is likely to change a bit before the first offical release.