[JGIT PATCH 02/14] Fix NB.decodeUInt16 to correctly handle the high byte

Our decodeUInt16 method was buggy and always cleared the high byte
of the pair.  This meant we always lost the upper 8 bits when we
read in a 16 bit unsigned integer, possibly causing us to misread
the data associated with that pair.

Signed-off-by: Shawn O. Pearce <redacted>
---
 org.spearce.jgit/src/org/spearce/jgit/util/NB.java |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/org.spearce.jgit/src/org/spearce/jgit/util/NB.java b/org.spearce.jgit/src/org/spearce/jgit/util/NB.java
index c6176f8..fa13354 100644
--- a/org.spearce.jgit/src/org/spearce/jgit/util/NB.java
+++ b/org.spearce.jgit/src/org/spearce/jgit/util/NB.java

@@ -102,7 +102,7 @@ public static int compareUInt32(final int a, final int b) {
 	 * @return unsigned integer value that matches the 16 bits read.
 	 */
 	public static int decodeUInt16(final byte[] intbuf, final int offset) {
-		int r = (intbuf[offset] << 8) & 0xff;
+		int r = (intbuf[offset] & 0xff) << 8;
 		return r | (intbuf[offset + 1] & 0xff);
 	}
 

-- 
1.6.0.87.g2858d