On Fri, 31 Jan 2025 19:36:02 GMT, Kevin Rushforth <k...@openjdk.org> wrote:
>> Created a test that validates various Nodes can be initialized in a
>> background thread.
>
> tests/system/src/test/java/test/robot/javafx/scene/NodeInitializationBackgroundThreadTest.java
> line 779:
>
>> 777:
>> 778: private static boolean nextBoolean() {
>> 779: return new Random().nextBoolean();
>
> It is a best practice when using Random to share a single instance and not
> allocate a new one each time. It is more performant and more random.
while true, the shared instance will add synchronization and this is something
I would like to avoid. The degree of randomness is not that important in this
case.
the only thing that we lose is reproducibility, but in the case of a headful
test it's probably impossible anyway, due to many other factors.
In any case, I think the main purpose of this test is a sort of sanity check of
the requirement, not an exhaustive test of all the properties and all the
methods that the application can call from a background thread - at least that
what I thought at the time of writing. We can always add more stress to it.
-------------
PR Review Comment: https://git.openjdk.org/jfx/pull/1690#discussion_r1937935407
