Re: java.util.Properties -- make it thread-safe?

[Mark Wielaard]

Hi,

On Fri, 2002-02-15 at 15:44, Jeff Sturm wrote:
> On 15 Feb 2002, Mark Wielaard wrote:
> > When javadoc puts a synchronized in the javadoc you can be sure that the
> > thread will need to acquire the lock for that object.
> 
> Not so.  For instance, Java allows overriding a synchronized method with a
> non-sychronized one.  Relying on the information from javadoc could be
> misleading.
Jikes. You (and Artur) are right and I was wrong. I have been relying on
this information. Hmmmm. Time to review all my code again... :}

> Also, the inverse doesn't apply:  a method not identified as synchronized
> in javadoc may synchronize "this" anyway.
That is true, but as Artur pointed out this should be in the
documentation.

> For these reasons, I believe "synchornized" is part of a method's
> implementation but not its interface, and I am glad javadoc ignores it.

I have to agree now. Lets hope that the synchronization behaviour (and
thread safety) of methods is always described correctly in the
documentation...

Cheers,

Mark