|Adding const to java email@example.com (Ghulam Lashari) (2002-10-24)|
|deepfinal modifier for Java firstname.lastname@example.org (Ghulam Lashari) (2002-11-13)|
|Re: deepfinal modifier for Java email@example.com (Joachim Durchholz) (2002-11-15)|
|Re: deepfinal modifier for Java firstname.lastname@example.org (Eric Eide) (2002-11-20)|
|From:||"Joachim Durchholz" <email@example.com>|
|Date:||15 Nov 2002 00:37:18 -0500|
|Posted-Date:||15 Nov 2002 00:37:18 EST|
Ghulam Lashari wrote:
> In fact, I had implemented a "deepfinal" modifier for Java.
> Is there any usefulness of this thing in Program analysis (static
> analysis), Invariant detection etc?
It would be useful if it worked.
The problem is if some deepfinal object is referring to another object
which is accessible via a non-deepfinal reference.
In other words: to make use of "deepfinal", the compiler will have to do
data-flow analysis - and if the compiler is smart enough to do that, I
suspect that it is also smart enough to deduce whether an object is
deepfinal or not.
Return to the
Search the comp.compilers archives again.