Index: lucene/analysis/common/src/test/org/apache/lucene/analysis/charfilter/TestMappingCharFilter.java
===================================================================
--- lucene/analysis/common/src/test/org/apache/lucene/analysis/charfilter/TestMappingCharFilter.java	(revision 1373984)
+++ lucene/analysis/common/src/test/org/apache/lucene/analysis/charfilter/TestMappingCharFilter.java	(working copy)
@@ -33,6 +33,7 @@
 import org.apache.lucene.analysis.MockTokenizer;
 import org.apache.lucene.analysis.TokenStream;
 import org.apache.lucene.analysis.Tokenizer;
+import org.apache.lucene.util.UnicodeUtil;
 import org.apache.lucene.util._TestUtil;
 
 public class TestMappingCharFilter extends BaseTokenStreamTestCase {
@@ -55,6 +56,8 @@
 
     builder.add( "empty", "" );
 
+    builder.add(UnicodeUtil.newString(new int[] {0x1D122}, 0, 1), "fclef");
+
     normMap = builder.build();
   }
 
@@ -128,6 +131,12 @@
     assertTokenStreamContents(ts, new String[0], new int[]{}, new int[]{}, 5);
   }
 
+  public void testNonBMP() throws Exception {
+    CharFilter cs = new MappingCharFilter( normMap, new StringReader( UnicodeUtil.newString(new int[] {0x1D122}, 0, 1) ) );
+    TokenStream ts = new MockTokenizer(cs, MockTokenizer.WHITESPACE, false);
+    assertTokenStreamContents(ts, new String[]{"fclef"}, new int[]{0}, new int[]{2}, 2);
+  }
+
   //
   //                1111111111222
   //      01234567890123456789012
Index: lucene/analysis/common/src/java/org/apache/lucene/analysis/charfilter/NormalizeCharMap.java
===================================================================
--- lucene/analysis/common/src/java/org/apache/lucene/analysis/charfilter/NormalizeCharMap.java	(revision 1373984)
+++ lucene/analysis/common/src/java/org/apache/lucene/analysis/charfilter/NormalizeCharMap.java	(working copy)
@@ -111,9 +111,8 @@
         final org.apache.lucene.util.fst.Builder<CharsRef> builder = new org.apache.lucene.util.fst.Builder<CharsRef>(FST.INPUT_TYPE.BYTE2, outputs);
         final IntsRef scratch = new IntsRef();
         for(Map.Entry<String,String> ent : pendingPairs.entrySet()) {
-          builder.add(Util.toUTF32(ent.getKey(), scratch),
+          builder.add(Util.toUTF16(ent.getKey(), scratch),
                       new CharsRef(ent.getValue()));
-      
         }
         map = builder.finish();
         pendingPairs.clear();
Index: lucene/core/src/java/org/apache/lucene/util/fst/Util.java
===================================================================
--- lucene/core/src/java/org/apache/lucene/util/fst/Util.java	(revision 1373984)
+++ lucene/core/src/java/org/apache/lucene/util/fst/Util.java	(working copy)
@@ -767,6 +767,21 @@
     }
   }
 
+  /** Just maps each UTF16 unit (char) to the ints in an
+   *  IntsRef. */
+  public static IntsRef toUTF16(CharSequence s, IntsRef scratch) {
+    final int charLimit = s.length();
+    scratch.grow(charLimit);
+    int idx = 0;
+    while(idx < charLimit) {
+      scratch.ints[idx] = (int) s.charAt(idx);
+      idx++;
+    }
+    scratch.offset = 0;
+    scratch.length = idx;
+    return scratch;
+  }
+
   /** Decodes the Unicode codepoints from the provided
    *  CharSequence and places them in the provided scratch
    *  IntsRef, which must not be null, returning it. */