[Bug classpath/23637] New: Incorrect conversion of UTF-16 decoder

[freebeans at xqb dot biglobe dot ne dot jp]

UTF-16 encoder (gnu.java.nio.charset.UTF_16Decoder) could not decode correctly.
Example:
public class EncodingTest {
  static final String JAPANESE_HIRAGANA_LETTER_RA = "\u3089";
  public static void main(String[] args) throws Exception {
    String s1 = JAPANESE_HIRAGANA_LETTER_RA;
    byte[] data = s1.getBytes("UTF-16LE");
    String s2 = new String(data, "UTF-16LE");
    if (s1.equals(s2)) {
      System.out.println("OK");
    } else {
      System.out.println("NG");
    }
}
---
Example shows "NG".

UTF_16Decoder line 108:
---
108: char c = byteOrder == BIG_ENDIAN ? (char) ((b1 << 8) | b2)
109:                                          : (char) ((b2 << 8) | b1);
---

But I think that the following codes are correct. 
---
108: char c = byteOrder == BIG_ENDIAN ? (char) ((b1 << 8) | (b2 & 0xff))
109:                                          : (char) ((b2 << 8) | (b1 & 
0xff));
---

Thank you.

-- 
           Summary: Incorrect conversion of UTF-16 decoder
           Product: classpath
           Version: 0.17
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: classpath
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: freebeans at xqb dot biglobe dot ne dot jp
                CC: bug-classpath at gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23637