Error in ThreadLocal's JavaDoc?

Question

I'm currently reading "Java Concurrency In Practice" (an excellent book,
BTW), and was looking at the passage on ThreadLocal. I then went to the
JavaDoc, and it sure looks to me like there's an error in the example code
in the JavaDoc. Can someone confirm or contradict me?
See http://java.sun.com/javase/6/docs/api/java/lang/ThreadLocal.html
The example code looks like this:
public class UniqueThreadIdGenerator {
private static final AtomicInteger uniqueId = new AtomicInteger(0);
private static final ThreadLocal < Integer > uniqueNum =
new ThreadLocal < Integer > () {
@Override protected Integer initialValue() {
return uniqueId.getAndIncrement();
public static int getCurrentThreadId() {
return uniqueId.get();
} // UniqueThreadIdGeneratorIt seems to me that getCurrentThreadId() should call uniqueNum.get() , not
uniqueId.get() . uniqueId.get() would return the last unique ID generated
in any thread, since it isn't ThreadLocal, while uniqueNum is the
ThreadLocal ID for this thread.
By the way, if you look at the JavaDoc for 1.4.2 and 1.5.0, it's a very similar
example, that appears correct to me (except for the counter not being
and AtomicInteger, since that didn't exist in 1.4.2, but the 1.5.0 example has no excuse for not avoiding that race condition).
[1.4.2 JavaDoc for ThreadLocal.html|AtomicInteger]
[1.5.0 JavaDoc for ThreadLocal.html|http://java.sun.com/j2se/1.5.0/docs/api/java/lang/ThreadLocal.html]
Am I right? Thanks.

Answer 1

Heh. I did search for it before sending the bug report, but I think I searched for ThreadLocal Example or Doc or something like that. Thanks for finding that but report.
The original bug report was filed almost three years ago, but I see someone submitted a comment two days ago mentioning it still was wrong in the canonical JavaDoc location. Odd coincidence, that.
It seems this example code has been a problem for a while. See bugs 4850726 and 6251838.