Support for optimistic locking [DATAJDBC-219]

[spring-projects-issues]

Jens Schauder opened DATAJDBC-219 and commented

If there is a @Version  annotated attribute in an Aggregate Root the update on the Aggregate Root should check that the version has the current value and increment that value.

If the update fails due to that version check an exception should get thrown

---

Issue Links:

• DATAJDBC-312 Support for optimistic locking on Aggregate Roots
  ("is duplicated by")

Referenced from: pull request #166

3 votes, 7 watchers

[spring-projects-issues]

Tom Hombergs commented

I'd like to give this one a try. This is what I came up with:

• on insert, if a @Version field is present, set version value to 1 (in DefaultDataAccessStrategy.insert())
• on update, if a @Version field is present
  • add condition "WHERE version_field = :oldVersion" to update statement (add a new method getUpdateWithVersion() or similar to SqlGenerator)
  • if no row was updated, someone else must have updated the row in the meantime and we throw an OptimisticLockingFailureException (in DefaultDataAccessStrategy.update())

Does that make sense? Is DefaultDataAccessStrategy the right place to attack? 

BTW: if you currently add a @Version field and perform an update on an entity, Spring Data JDBC will try to do an insert instead (and fail due to a unique constraint if you have a unique ID column). The reason is that the entity is interpreted as a new entity because the version is still 0. This would be fixed by the above approach because the version field would be properly updated

[spring-projects-issues]

Tom Hombergs commented

In the list above, I forgot:

• on update
  • add "set version_field = version_field + 1" to update statement
  • probably we also want to increment the version in the entity object itself so the client will receive the new version

[spring-projects-issues]

Jens Schauder commented

Yes, that sounds right. 

And we definitly need to update the version in the entity

[spring-projects-issues]

Tom Hombergs commented

Thanks, I'll get to it, then :)

[spring-projects-issues]

Tom Hombergs commented

Pull request: #124

[spring-projects-issues]

Tyler K Van Gorder commented

I have an updated PR that includes the work that Tom Hombergs started that can be found here: #166