Rob Eden

TCollections.synchronizedSet is type agnostic, simply wrapping whatever implementation is given to it (as Collections.unmodifiableSet does from the JDK). Implementing a concurrent set isn't trivial and I am by no means an expert. I know the JDK implements a concurrent skip list for sets and ConcurrentHashMap uses a system of buckets.

Quite a few people have talked about it, but I don't think I've seen an actual implementation (or beginnings of one). Doing it properly would be quite an undertaking, but it is obviously doable. Side note: rather than extending TCustomHashSet, you should be able to use TCollections.synchronizedSet.

Quite a few people have talked about it, but I don't think I've seen an actual implementation (or beginnings of one). Doing it properly would be quite an undertaking, but it is obviously doable. Side note: rather than extending TCustomHashSet, would should be able to use TCollections.synchronizedSet.

Let's discuss in the Stack Overflow question you raised: https://stackoverflow.com/questions/48170039/gnu-trove-classes-throws-java-lang-arithmeticexception-divide-by-zero

Hi Bill, welcome to Trove. I believe the source jar in maven should have everything....

Hello all - First off, the official Trove repo has moved locations on BitBucket....

Code and the issue database has moved to BitBucket. There's a link on the front SF...

IANAL, so take this with a grain of salt. Trove doesn't fall into any category of...