RE: [cp-patches] [RFC/JDWP] ClassLoaderReferenceCommandSet

2005-08-09 Thread Jeroen Frijters
Archie Cobbs wrote:
 Aaron Luchko wrote:
  There's only a single command which asks for a list of 
 every class the
  ClassLoader has been requested to load. The first way to do 
 this is just
  use a hook in the vm. I don't know enough with how ClassLoaders are
  handled within the vm to know how easy this would be to implement.
  
  The second method involves a patch to ClassLoader which 
 adds a single
  variable loadRequests which will store every Class that is 
 requested to
  be loaded.
 
 I'd strongly vote for option #1, i.e., just ask the VM directly.
 VMs that want to support this stuff are going to have to implement
 a lot of other native methods too, and this one should be fairly easy.

I strongly agree with Archie. Also, approach #2 is incorrect, because
the classes that should be returned are the classes for which this
class loader has been recorded as the initiating loader. The class
loader doesn't know whether it is the initiating loader or not, only the
VM knows this.

Regards,
Jeroen


___
Classpath-patches mailing list
Classpath-patches@gnu.org
http://lists.gnu.org/mailman/listinfo/classpath-patches


RE: [cp-patches] [RFC/JDWP] ClassLoaderReferenceCommandSet

2005-08-09 Thread Jeroen Frijters
Archie Cobbs wrote:
 Jeroen Frijters wrote:
  I strongly agree with Archie. Also, approach #2 is 
 incorrect, because
  the classes that should be returned are the classes for which this
  class loader has been recorded as the initiating loader. The class
  loader doesn't know whether it is the initiating loader or 
 not, only the
  VM knows this.
 
 Tiny clarification.. that should be an initiating loader rather
 than the initiating loader... there can be more than one for
 a given class, right?

Yeah, you're absolutely right. I copied the quoted text from the Sun
docs, but it's definitely misleading to talk about the.

Regards,
Jeroen


___
Classpath-patches mailing list
Classpath-patches@gnu.org
http://lists.gnu.org/mailman/listinfo/classpath-patches


Re: [cp-patches] [RFC/JDWP] ClassLoaderReferenceCommandSet

2005-08-09 Thread Archie Cobbs

Jeroen Frijters wrote:

I strongly agree with Archie. Also, approach #2 is incorrect, because
the classes that should be returned are the classes for which this
class loader has been recorded as the initiating loader. The class
loader doesn't know whether it is the initiating loader or not, only the
VM knows this.


Tiny clarification.. that should be an initiating loader rather
than the initiating loader... there can be more than one for
a given class, right?

-Archie

__
Archie Cobbs  *CTO, Awarix*  http://www.awarix.com


___
Classpath-patches mailing list
Classpath-patches@gnu.org
http://lists.gnu.org/mailman/listinfo/classpath-patches


[cp-patches] [RFC/JDWP] ClassLoaderReferenceCommandSet

2005-08-08 Thread Aaron Luchko
Here's two different proposals for the ClassLoaderRefererenceCommandSet
http://java.sun.com/j2se/1.5.0/docs/guide/jpda/jdwp/jdwp-protocol.html#JDWP_ClassLoaderReference

There's only a single command which asks for a list of every class the
ClassLoader has been requested to load. The first way to do this is just
use a hook in the vm. I don't know enough with how ClassLoaders are
handled within the vm to know how easy this would be to implement.

The second method involves a patch to ClassLoader which adds a single
variable loadRequests which will store every Class that is requested to
be loaded.

The patch has the advantages of being vm-agnostic and avoiding yet
another hook into the virtual machine.

On the downside we're putting what seems a useless variable in
ClassLoader and we may induce a small performance penalty (although this
could easily be present in the VM-hook as well).

The ClassLoader patch doesn't need to go in until we're further along if
we use it.

Aaron
--- /dev/null	2005-06-09 16:29:11.371620296 -0400
+++ gnu/classpath/jdwp/processor/ClassLoaderReferenceCommandSet.java	2005-08-08 14:35:35.0 -0400
@@ -0,0 +1,114 @@
+/* ClassLoaderReferenceCommandSet.java -- class to implement the 
+   ClassLoaderReference Command Set
+   Copyright (C) 2005 Free Software Foundation
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+
+package gnu.classpath.jdwp.processor;
+
+import gnu.classpath.jdwp.IVirtualMachine;
+import gnu.classpath.jdwp.Jdwp;
+import gnu.classpath.jdwp.JdwpConstants;
+import gnu.classpath.jdwp.exception.JdwpException;
+import gnu.classpath.jdwp.exception.JdwpInternalErrorException;
+import gnu.classpath.jdwp.exception.NotImplementedException;
+import gnu.classpath.jdwp.id.IdManager;
+import gnu.classpath.jdwp.id.ObjectId;
+import gnu.classpath.jdwp.id.ReferenceTypeId;
+
+import java.io.DataOutputStream;
+import java.io.IOException;
+import java.nio.ByteBuffer;
+import java.util.ArrayList;
+import java.util.Iterator;
+
+/**
+ * A class representing the ClassLoaderReference Command Set.
+ * 
+ * @author Aaron Luchko [EMAIL PROTECTED]
+ */
+public class ClassLoaderReferenceCommandSet implements CommandSet
+{
+  // Our hook into the jvm
+  private final IVirtualMachine vm = Jdwp.getIVirtualMachine();
+
+  // Manages all the different ids that are assigned by jdwp
+  private final IdManager idMan = Jdwp.getIdManager();
+
+  public boolean runCommand(ByteBuffer bb, DataOutputStream os, byte command)
+  throws JdwpException
+  {
+
+// Although there's only a single command to choose from we still use
+// a switch to maintain consistency with the rest of the CommandSets
+   try
+  {
+switch (command)
+  {
+  case JdwpConstants.CommandSet.ClassLoaderReference.VISIBLE_CLASSES:
+executeVisibleClasses(bb, os);
+break;
+  default:
+throw new NotImplementedException(Command  + command +
+ not found in ClassLoaderReference Command Set.);
+  }
+  }
+catch (IOException ex)
+  {
+// The DataOutputStream we're using isn't talking to a socket at all
+// So if we throw an IOException we're in serious trouble
+throw new JdwpInternalErrorException(ex);
+  }
+return true;
+  }
+