Sorry, that's tag SJSAS-9_0-B32G-BETA-10_Feb_2006 ! This corresponds to the most recent Milestone build 5 of the Glassfish project.
So I just checked out from CVS, compiled, and compared my generated binaries (classes only) to those provided by Glassfish for the persistence-api and transaction-api modules only. As far as SFV is concerned, these are identical files/directories. This is as we expected, but we needed to confirm it. So the binaries (Sun binary license) being delivered by Glassfish are identical to the source files (CDDL licensed). So... given this, what's the next step? I can revert my entire local CVS root to the B32G build, compile all the APIs, create the appropriate POMs (assuming we can make a decision on groupId and artifactId), and create bundles for all these Glassfish APIs, if we agree on this approach. Wayne On 2/28/06, Wayne Fay <[EMAIL PROTECTED]> wrote: > On 2/28/06, Steve Loughran <[EMAIL PROTECTED]> wrote: > > Oh, this nice. javax.persistence especially. I've been doing some EJB3 > > stuff and that is a sore point. JTA too. > > Since you expressed an interest in persistence and transaction, I > focused efforts on those two modules. > > I pulled down the Glassfish CVS code, compiled into classes, and > compared to the classes delivered by Glassfish distribution using SFV > (its like MD5 for a group of files/directories). > > Found my compiled classes had a number of differences including new files. > > Realized I was using CVS head rather than tag SJAS-9_0-B32G-BETA-10_Feb_2006. > > So now I'm pulling the tag code from CVS, will repeat the compile and > compare steps, and report back shortly... > > Wayne > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
