RE: currentClassLoader problem

[Andrew Haley]

David Holmes writes:
 > Andrew,
 > 
 > I just realized that this code isn't in the Classpath version at all.

You would not expect it to be, would you?  VMSecurityManager is part
of the VM-specific layer.  It's a good deal more helpful than

  static native ClassLoader currentClassLoader();

Andrew.

 > 
 > David 
 > 
 > > Index: java/lang/VMSecurityManager.java
 > > ===================================================================
 > > RCS file: /cvs/gcc/gcc/libjava/java/lang/VMSecurityManager.java,v
 > > retrieving revision 1.2
 > > diff -c -2 -p -r1.2 VMSecurityManager.java
 > > *** java/lang/VMSecurityManager.java       5 Dec 2002 00:49:29 -0000       
 > > 1.2
 > > --- java/lang/VMSecurityManager.java       15 Mar 2004 09:54:54 -0000
 > > *************** class VMSecurityManager
 > > *** 55,63 ****
 > >       // FIXME this implementation is a bit wrong too -- the docs say we
 > >       // must also consider ancestors of the system class loader.
 > >       Class[] classStack = getClassContext ();
 > >       for (int i = 0; i < classStack.length; i++)
 > >         {
 > >    ClassLoader loader = classStack[i].getClassLoader();
 > > !  if (loader != null)
 > >      return loader;
 > >         }
 > > --- 55,64 ----
 > >       // FIXME this implementation is a bit wrong too -- the docs say we
 > >       // must also consider ancestors of the system class loader.
 > > +     ClassLoader systemClassLoader = 
 > > VMClassLoader.getSystemClassLoader();
 > >       Class[] classStack = getClassContext ();
 > >       for (int i = 0; i < classStack.length; i++)
 > >         {
 > >    ClassLoader loader = classStack[i].getClassLoader();
 > > !  if (loader != null && loader != systemClassLoader)
 > >      return loader;
 > >         }
 > >