I'm wary of introducing unnecessary synchronization into critical code
paths.

I'm pretty sure I can make ReferenceMap not start the thread unless you try
to use a weak or soft reference.

Bob

On Thu, Dec 25, 2008 at 9:39 PM, [email protected] <[email protected]>wrote:

>
> I've prepared a change that strips out our dear friend,
> FinalizableReferenceQueue. This limits ReferenceCache to
> STRONG,STRONG.
>
> Benefits: no extra thread, so our redepoy in Java EE leakiness goes
> away. Also good for Java-like environments where starting threads is
> forbidden.
>
> Drawbacks: the caches in BytecodeGen (classloaders) and
> StackTraceElements are a bit uglier. For example, a WeakHashMap<Class<?
> >,SoftReference<LineNumbers>> is now used instead.
>
> Despite the ugliness, I figure I'll check this in as it makes an ugly
> problem go away for our users. Let me know if you've got any
> reservations, I could go either way on this one.
> >
>

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"google-guice-dev" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/google-guice-dev?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to