FYI: Patch: java/lang/SecurityManager.java

[Dalibor Topic]

Hi all,

I've committed the attached patch to clean up a direct use of an 
internal field in java.lang.Thread  instead of using the API method for 
it. I've tried to improve the docs a little bit, too.

2004-04-09  Dalibor Topic  <address@hidden>

        * java/lang/SecurityManager.java: (checkAccess): Use
        getThreadGroup(). Improved documentation.

cheers,
dalibor topic

Index: java/lang/SecurityManager.java
===================================================================
RCS file: /cvsroot/classpath/classpath/java/lang/SecurityManager.java,v
retrieving revision 1.16
diff -u -r1.16 SecurityManager.java
--- java/lang/SecurityManager.java      6 Jan 2004 09:56:04 -0000       1.16
+++ java/lang/SecurityManager.java      8 Apr 2004 15:31:04 -0000
@@ -375,9 +375,9 @@
    * <code>RuntimePermission("modifyThread")</code>, return silently, so that
    * core classes (the Classpath library!) can modify any thread.
    *
-   * @param t the other Thread to check
+   * @param thread the other Thread to check
    * @throws SecurityException if permission is denied
-   * @throws NullPointerException if t is null
+   * @throws NullPointerException if thread is null
    * @see Thread#stop()
    * @see Thread#suspend()
    * @see Thread#resume()
@@ -385,9 +385,10 @@
    * @see Thread#setName(String)
    * @see Thread#setDaemon(boolean)
    */
-  public void checkAccess(Thread t)
+  public void checkAccess(Thread thread)
   {
-    if (t.group != null && t.group.getParent() != null)
+    if (thread.getThreadGroup() != null 
+       && thread.getThreadGroup().getParent() != null)
       checkPermission(new RuntimePermission("modifyThread"));
   }