Re: Proposal for CNI/JNI problems

[Jon Olson <olson@mmsi.com>]

On Wed, 12 Jan 2000, Jochen Hoenicke wrote:

>Another point is how to put pointers to native structures into a
>classfile.  Sun didn't solve it well.  If I understand the code in
>japhar's java/util/zip correctly sun used an int field to store the
>pointer and later changed it to long to support 64bit architectures.
>libgcj declares natives fields as "gnu.gcj.RawData", but this is not
>portable to other jvms, where the garbage collector doesn't know that
>this class is special.  My solution was to put the structure into a
>java byte array, which imposes a little overhead, but should be
>portable (and you get it freed automatically).
>
>  Jochen

Interesting idea, putting the native peer class into a byte array.
Seems like kind of a hack, but I guess it solves the problem.
Sun's solution of using a long to store a pointer is a real kludge.

My solution was to store an `int' as the peer and to make any
Java class which require a native peer a subclass of this Peer
class:

public class Peer
{
  protected int _peer;
  public Peer(int peer)	{ _peer = peer; }
  public native void destroy();
  protected void finalize() {
    destroy();
  }
}

Native code stores peer objects in a C++ Vector and manages
recovering space when finalized in destroy().

-- 
Jon Olson, Modular Mining Systems
	   3289 E. Hemisphere Loop
	   Tucson, AZ 85706
INTERNET:  olson@mmsi.com
PHONE:     (520)746-9127
FAX:       (520)889-5790