From b55bbd25a9f4275e5cb3b5c17aef829757bf9ccf Mon Sep 17 00:00:00 2001
From: Mark Davis <mark@macchiato.com>
Date: Fri, 4 Mar 2011 23:56:24 +0000
Subject: [PATCH] ICU-8175 Added note to clarify that j-1 is safe.

X-SVN-Rev: 29546
---
 icu4j/main/classes/core/src/com/ibm/icu/lang/CharSequences.java | 1 +
 1 file changed, 1 insertion(+)

diff --git a/icu4j/main/classes/core/src/com/ibm/icu/lang/CharSequences.java b/icu4j/main/classes/core/src/com/ibm/icu/lang/CharSequences.java
index ecf6d52389..2bb2fe6cc1 100644
--- a/icu4j/main/classes/core/src/com/ibm/icu/lang/CharSequences.java
+++ b/icu4j/main/classes/core/src/com/ibm/icu/lang/CharSequences.java
@@ -275,6 +275,7 @@ public class CharSequences {
             if (cp >= 0xDC00 && cp <= 0xDFFF && i != 0 ) { // hand-code for speed
                 char last = (char) result[j-1];
                 if (last >= 0xD800 && last <= 0xDBFF) {
+                    // Note: j-1 is safe, because j can only be zero if i is zero. But i!=0 in this block.
                     result[j-1] = Character.toCodePoint(last, cp);
                     continue;
                 }