Community
Participate
Working Groups
From the I7 planning doc at http://wiki.eclipse.org/EDT:0.7.0_Planning:Iteration_7 data base I/O - needs to be designed, including how JPA will play in, vs straight SQL. In I7, we will target having the direction / design set.
JPA is the way to go. A good initial implementation would be to have the user define all their JPA stuff in Java, and make EGL external types from the POJOs. (Joe is working on code to make ETs from Java objects. See 348119.) EDT would support basic CRUD stuff on the ETs. Straight SQL can be used alongside JPAs, so adding that capability would be the second step. Finally, we can support SQL records. They can optionally include embeds of the ETs for the JPAs, and we'd have to keep track of the relationships. For example, you could embed an ET for a row of one table, and an ET for a row of another table. Doing "add rec;" would insert into both tables. (This is another way of saying an SQL record could be a view.)
Closing: Done.
Reopening, this needs to be more specific.
*** This bug has been marked as a duplicate of bug 355573 ***