On Jul 24, 2005, at 5:02 AM, Hans Gilde wrote:
So I'm looking at your test, and it looks like one should expect this behavior (the test should always fail). Have you tried this test on a pre-leak-fix version of Jelly? I'm going to do so tomorrow.
Yes. This test mimics a regression from an earlier RC that I ran into when upgrading Jelly in my application.
Previous behavior was for each call to TagScript.run() to use a new tag instance.
This isn't the same caching problem reported by other users. That problem is caused not by repeated invocation of the same script, but by something that is causing the same TagScript to be sued for two XML tags. This makes thesecond instance use the variables from the first instance.
When we fixed the ThreadLocal memory leak, JellyContext.isCacheTags() was also removed. It is that function and the code in TagScript.run() that checks it that I restored.
-pete -- peter royal
smime.p7s
Description: S/MIME cryptographic signature
