In reply to Re: A way to test that circular references are actually freed by perrin in thread A way to test that circular references are actually freed by Jeffrey Kegler