0N/A

0N/A

0N/Aimport java.io.InputStreamReader;

0N/Aimport java.io.OutputStreamWriter;

0N/Aimport java.io.ByteArrayInputStream;

0N/Aimport java.io.ByteArrayOutputStream;

0N/Aimport java.nio.charset.Charset;

0N/Aimport java.nio.charset.UnsupportedCharsetException;

0N/A

0N/A

0N/Apublic class Encodings {

0N/A

0N/A

0N/A static boolean equals(byte[] a, byte[] b) {

0N/A if (a.length != b.length) return false;

0N/A for (int i = 0; i < a.length; i++)

0N/A if (a[i] != b[i]) return false;

0N/A return true;

0N/A }

0N/A

0N/A

0N/A static void go(String enc, String str, final byte[] bytes, boolean bidir)

0N/A throws Exception

0N/A {

0N/A final Charset charset = Charset.forName(enc);

0N/A

0N/A /* String(byte[] bs, String enc) */

0N/A if (!(new String(bytes, enc).equals(str)))

0N/A throw new Exception(enc + ": String constructor failed");

0N/A

0N/A /* String(byte[] bs, Charset charset) */

0N/A if (!(new String(bytes, charset).equals(str)))

0N/A throw new Exception(charset + ": String constructor failed");

0N/A

0N/A /* String(byte[] bs, int off, int len, Charset charset) */

0N/A String start = str.substring(0, 2);

0N/A String end = str.substring(2);

0N/A if (enc.equals("UTF-16BE") || enc.equals("UTF-16LE")) {

0N/A if (!(new String(bytes, 0, 4, charset).equals(start)))

0N/A throw new Exception(charset + ": String constructor failed");

0N/A if (!(new String(bytes, 4, bytes.length - 4, charset).equals(end)))

0N/A throw new Exception(charset + ": String constructor failed");

0N/A } else if (enc.equals("UTF-16")) {

0N/A if (!(new String(bytes, 0, 6, charset).equals(start)))

0N/A throw new Exception(charset + ": String constructor failed");

0N/A } else {

0N/A if (!(new String(bytes, 0, 2, charset).equals(start)))

0N/A throw new Exception(charset + ": String constructor failed");

0N/A if (!(new String(bytes, 2, bytes.length - 2, charset).equals(end)))

0N/A throw new Exception(charset + ": String constructor failed");

0N/A }

0N/A

0N/A /* InputStreamReader */

0N/A ByteArrayInputStream bi = new ByteArrayInputStream(bytes);

0N/A InputStreamReader r = new InputStreamReader(bi, enc);

0N/A String inEnc = r.getEncoding();

0N/A int n = str.length();

0N/A char[] cs = new char[n];

0N/A for (int i = 0; i < n;) {

0N/A int m;

0N/A if ((m = r.read(cs, i, n - i)) < 0)

0N/A throw new Exception(enc + ": EOF on InputStreamReader");

0N/A i += m;

0N/A }

0N/A if (!(new String(cs).equals(str)))

0N/A throw new Exception(enc + ": InputStreamReader failed");

0N/A

0N/A if (!bidir) {

0N/A System.err.println(enc + " --> " + inEnc);

0N/A return;

0N/A }

0N/A

0N/A /* String.getBytes(String enc) */

0N/A byte[] bs = str.getBytes(enc);

0N/A if (!equals(bs, bytes))

0N/A throw new Exception(enc + ": String.getBytes failed");

0N/A

0N/A /* String.getBytes(Charset charset) */

0N/A bs = str.getBytes(charset);

0N/A if (!equals(bs, bytes))

0N/A throw new Exception(charset + ": String.getBytes failed");

0N/A

0N/A // Calls to String.getBytes(Charset) shouldn't automatically

0N/A // use the cached thread-local encoder.

0N/A if (charset.name().equals("UTF-16BE")) {

0N/A String s = new String(bytes, charset);

0N/A // Replace the thread-local encoder with this one.

0N/A byte[] bb = s.getBytes(Charset.forName("UTF-16LE"));

0N/A if (bytes.length != bb.length) {

0N/A // Incidental test.

0N/A throw new RuntimeException("unequal length: "

0N/A + bytes.length + " != "

0N/A + bb.length);

0N/A } else {

0N/A boolean diff = false;

0N/A // Expect different byte[] between UTF-16LE and UTF-16BE

0N/A // even though encoder was previously cached by last call

0N/A // to getBytes().

0N/A for (int i = 0; i < bytes.length; i++) {

0N/A if (bytes[i] != bb[i])

0N/A diff = true;

0N/A }

0N/A if (!diff)

0N/A throw new RuntimeException("byte arrays equal");

0N/A }

0N/A }

0N/A

0N/A /* OutputStreamWriter */

0N/A ByteArrayOutputStream bo = new ByteArrayOutputStream();

0N/A OutputStreamWriter w = new OutputStreamWriter(bo, enc);

0N/A String outEnc = w.getEncoding();

0N/A w.write(str);

0N/A w.close();

0N/A bs = bo.toByteArray();

0N/A if (!equals(bs, bytes))

0N/A throw new Exception(enc + ": OutputStreamWriter failed");

0N/A

0N/A System.err.println(enc + " --> " + inEnc + " / " + outEnc);

0N/A }

0N/A

0N/A

0N/A static void go(String enc, String str, byte[] bytes) throws Exception {

0N/A go(enc, str, bytes, true);

0N/A }

0N/A

0N/A

0N/A public static void main(String[] args) throws Exception {

0N/A

0N/A go("US-ASCII", "abc", new byte[] { 'a', 'b', 'c' });

0N/A go("us-ascii", "abc", new byte[] { 'a', 'b', 'c' });

0N/A go("ISO646-US", "abc", new byte[] { 'a', 'b', 'c' });

0N/A go("ISO-8859-1", "ab\u00c7", new byte[] { 'a', 'b', (byte)'\u00c7' });

0N/A go("UTF-8", "ab\u1e09",

0N/A new byte[] { 'a', 'b',

0N/A (byte)(0xe0 | (0x0f & (0x1e09 >> 12))),

0N/A (byte)(0x80 | (0x3f & (0x1e09 >> 6))),

0N/A (byte)(0x80 | (0x3f & 0x1e09)) });

0N/A go("UTF-16BE", "ab\u1e09",

0N/A new byte[] { 0, 'a', 0, 'b', 0x1e, 0x09 });

0N/A go("UTF-16LE", "ab\u1e09",

0N/A new byte[] { 'a', 0, 'b', 0, 0x09, 0x1e });

0N/A

0N/A /* UTF-16 accepts both byte orders on input but always uses big-endian

0N/A go("UTF-16", "ab\u1e09",

0N/A new byte[] { (byte)0xfe, (byte)0xff, 0, 'a', 0, 'b', 0x1e, 0x09 });

0N/A go("UTF-16", "ab\u1e09",

0N/A new byte[] { (byte)0xff, (byte)0xfe, 'a', 0, 'b', 0, 0x09, 0x1e },

0N/A false);

0N/A }

0N/A

0N/A}