Skip to main content

Please review fix for 6513256 - tree update problems.

1 reply [Last post]
Joined: 2004-12-13


- replace deprecated call with the non-deprecated version (just a method rename)

Primary problem was that during insert/remove event processing, some of that code caused out of order operations to the data. Some queries done _before_ the insertion took place caused the data to change - so code was moved into the lower class so that the operation is more atomic (TreeModelEvent now return type). Primary case of root cause was calling TT_BasicNode.findByName() - which caused population of the node which was an unexpected effect that the code was not written to assume.

Both insertion and removal events have been repaired.

Things found not to be a root cause problem:
- events correctly transition from underlying threads to the event thread
- Events from core data model seem to be symmetric, which is necessary

Other changes:
- synchronization upgraded to use some "new" API in import java.util.concurrent.atomic.*.
- some utility methods added
- removed redudent field declaration in TT_TestNode

Reply viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
Joined: 2004-12-13

Sergey looked, agrees in principle. Hard for someone to review though. Some people have tested it and it seems not to be DOA.

Integrated rev 1524 to trunk for b07.