java.security expert?

[Andrew Haley]

Tom Tromey writes:
 > While debugging gcjx, I ran across some strange code in Prime.java.
 > I think the intention was for the code to resemble the result of the
 > appended patch.  The patch looks big, but it just wraps the body in a
 > `while (true)' and changes a `break' to `continue'.
--text follows this line--
`diff -w' is your friend.

Index: Prime.java
===================================================================
RCS file: /cvs/gcc/gcc/libjava/gnu/java/security/util/Prime.java,v
retrieving revision 1.3
diff -u -w -r1.3 Prime.java
--- Prime.java  11 Aug 2002 16:34:44 -0000      1.3
+++ Prime.java  6 Mar 2004 11:24:58 -0000
@@ -1,5 +1,5 @@
 /* Prime.java --- Prime number generation utilities
-   Copyright (C) 1999 Free Software Foundation, Inc.
+   Copyright (C) 1999, 2004 Free Software Foundation, Inc.
 
 This file is part of GNU Classpath.
 
@@ -105,14 +105,19 @@
   /*
     See IEEE P1363 A.15.5 (10/05/98 Draft)
   */
-  public static BigInteger generateRandomPrime( BigInteger r, BigInteger a, 
int pmin, int pmax, BigInteger f )
+  public static BigInteger generateRandomPrime( BigInteger r, BigInteger a,
+                                               int pmin, int pmax,
+                                               BigInteger f )
   {
     BigInteger d, w;
 
     //Step 1 - generate prime
     BigInteger p = new BigInteger( (pmax + pmin)/2, new Random() );
 
-  steptwo:{ //Step 2
+  steptwo:
+    while (true)
+      {
+       //Step 2
       w = p.mod( r.multiply( BigInteger.valueOf(2) ));
 
       //Step 3
@@ -121,7 +126,8 @@
       p = p.add(a);
 
       //Step 4 - test for even
-      if( p.mod( BigInteger.valueOf(2) ).compareTo( BigInteger.valueOf( 0 )) 
== 0)
+       if( p.mod( BigInteger.valueOf(2) ).compareTo( BigInteger.valueOf( 0 ))
+           == 0)
        p.add( r );
 
       for(;;)
@@ -135,7 +141,7 @@
              p = p.subtract( BigInteger.valueOf( 1 ) );
 
              //Step 5.2 - goto to Step 2
-             break steptwo;
+               continue steptwo;
            }
 
          //Step 6
@@ -158,7 +164,5 @@
          //Step 9
        }
     }
-    //Should never reach here but makes the compiler happy
-    return BigInteger.valueOf(0);
   }
 }

 > Any comments on this?  I didn't look up the IEEE spec that this
 > references, for all I know the comments are the only part that are
 > wrong.

Your patch looks right to me.

Andrew.