public class Utf8Filter {public static void main(String[] args) throws Exception {System.out.println(filterMt4BytesUtf8(new String(new byte[] { (byte) 0x41 }, "utf-8")));System.out.println(filterMt4BytesUtf8(new String(new byte[] { (byte) 0xf3, (byte) 0xb7,(byte) 0xa2, (byte) 0xbe }, "utf-8")));System.out.println(filterMt4BytesUtf8(new String(new byte[] { (byte) 0x41, (byte) 0xf3,(byte) 0xb7, (byte) 0xa2, (byte) 0xbe }, "utf-8")));System.out.println(filterMt4BytesUtf8(new String(new byte[] { (byte) 0xf3, (byte) 0xb7,(byte) 0xa2, (byte) 0xbe, (byte) 0x41 }, "utf-8")));System.out.println(filterMt4BytesUtf8(new String(new byte[] { (byte) 0x41, (byte) 0xf3,(byte) 0xb7, (byte) 0xa2, (byte) 0xbe, (byte) 0x41 }, "utf-8")));System.out.println(filterMt4BytesUtf8(new String(new byte[] { (byte) 0xf3, (byte) 0xb7,(byte) 0xa2, (byte) 0xbe, (byte) 0xf3, (byte) 0xb7, (byte) 0xa2, (byte) 0xbe,(byte) 0x41 }, "utf-8")));System.out.println(filterMt4BytesUtf8(new String(new byte[] { (byte) 0x41, (byte) 0xf3,(byte) 0xb7, (byte) 0xa2, (byte) 0xbe, (byte) 0xf3, (byte) 0xb7, (byte) 0xa2,(byte) 0xbe }, "utf-8")));System.out.println(filterMt4BytesUtf8(new String(new byte[] { (byte) 0x41, (byte) 0xf3,(byte) 0xb7, (byte) 0xa2, (byte) 0xbe, (byte) 0xf3, (byte) 0xb7, (byte) 0xa2,(byte) 0xbe, (byte) 0x41 }, "utf-8")));}private static String filterMt4BytesUtf8(String input) {if (StringUtils.isBlank(input))return input;ByteArrayOutputStream is = new ByteArrayOutputStream();byte[] bytes = input.getBytes();outter: for (int i = 0, length = bytes.length; i < length; i++) {byte b = bytes[i];while (((b & 0xF8) == 0xF0) || ((b & 0xFc) == 0xF8) || ((b & 0xFe) == 0xFc)) {// found a byte of 4, 5, 6 UTF-8 bytes for a character// eat the following bytes of this characterwhile (++i < length && ((b = bytes[i]) & 0xC0) == 0x80);// insert a replacement character

is.write((byte) 0xEF);is.write((byte) 0xBF);is.write((byte) 0xBD);if (i >= length) {// we reach the end of byte arraybreak outter;}}// found byte of 1, 2, 3 UTF-8 bytes for a character

is.write(b);}try {return new String(is.toByteArray(), "utf-8");} catch (UnsupportedEncodingException e) {logger.error("filter more than 4 bytes utf-8 character failed!", e);}return input;}}