hello all,
the attached patch fixes the above PR.
2006-08-06 Raif S. Naffah <[EMAIL PROTECTED]>
PR Classpath/27372
* java/math/BigInteger.java: Updated copyright year.
(init): Consume as little bytes as possible.
a new Mauve test gnu/testlet/java/math/BigInteger/TestOfPR27372.java has been
added to highlight the problem and validate the fix. a new method
(testCtorIntRandom) has been added to the Mauve test
gnu/testlet/java/math/BigInteger/ctor.java to ensure no regression was
introduced.
still, because BigInteger is a sensitive class, i would appreciate it if
another pair of eyes can go over the patch to ensure no mistakes have been
introduced.
TIA + cheers;
rsn
Index: BigInteger.java
===================================================================
RCS file: /cvsroot/classpath/classpath/java/math/BigInteger.java,v
retrieving revision 1.28
diff -u -r1.28 BigInteger.java
--- BigInteger.java 27 May 2006 02:09:30 -0000 1.28
+++ BigInteger.java 6 Aug 2006 07:30:15 -0000
@@ -1,5 +1,5 @@
/* java.math.BigInteger -- Arbitary precision integers
- Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003, 2005 Free Software Foundation, Inc.
+ Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003, 2005, 2006 Free Software Foundation, Inc.
This file is part of GNU Classpath.
@@ -197,8 +197,20 @@
private void init(int numBits, Random rnd)
{
int highbits = numBits & 31;
+ // minimum number of bytes to store the above number of bits
+ int highBitByteCount = (highbits + 7) / 8;
+ // number of bits to discard from the last byte
+ int discardedBitCount = highbits % 8;
+ if (discardedBitCount != 0)
+ discardedBitCount = 8 - discardedBitCount;
+ byte[] highBitBytes = new byte[highBitByteCount];
if (highbits > 0)
- highbits = rnd.nextInt() >>> (32 - highbits);
+ {
+ rnd.nextBytes(highBitBytes);
+ highbits = (highBitBytes[highBitByteCount - 1] & 0xFF) >>> discardedBitCount;
+ for (int i = highBitByteCount - 2; i >= 0; i--)
+ highbits = (highbits << 8) | (highBitBytes[i] & 0xFF);
+ }
int nwords = numBits / 32;
while (highbits == 0 && nwords > 0)
