John Hendrikx

A new simpler test case, that clearly shows the bounds in local of the arc being incorrect. package hs.javafxbug; import java.util.ArrayList; import java.util.List; import java.util.concurrent.Executors; import java.util.concurrent.TimeUnit; import javafx.animation.KeyFrame; import...

https://bugs.java.com/bugdatabase/view_bug.do?bug_id=JDK-8228698

Rotations like that are done with a transform. The button won't report correct layout bounds when this happens, unless wrapped in a Group.

Added a fix for this in #318

Gaps can also be temporary, depending on commit order. If there are two transactions, one which claims sequence 1 and the other claims sequence 2, if the latter commits first,...

What I mean is, there can be two processes doing (for different aggregates): UPDATE ORDER_AGGREGATE SET VERSION = VERSION + 1 WHERE ID = ? AND VERSION = ? INSERT...

Yes, but you are updating the same aggregate here. If you update two different aggregates, then the step: ID for ORDER_EVENT is generated from sequence ... can happen in a...

> In my sample, the problem you describe is relevant to the Transaction Outbox pattern implementation only. Yes, exactly. > I will change the implementation of the Transaction Outbox pattern...

I'd be in favor of doing this with a `Point3D` -- the amount of garbage created should be taken in the light of the framerate of the application. At 60...

> Fundamentally, the problem does not arise from bindings, but from the fact that `ListPropertyBase` adds a `ListChangeListener` to its wrapped `ObservableList`. Yes, I would agree. The cause is the...