UTF8から文字コード変換するときにSJISの範囲外の文字を任意の文字に置き換えるサンプル

CharTest.java

String str = "a吉野屋𠮷野\r\n屋ア緣ｲ?";
//        byte[] bytes = str.getBytes("Windows-31j");
char alt = '□';

final String charset = "Windows-31j";

String filteredStr = filterLowSurrogate(str);
String translatedStr = new String(str.getBytes(charset), Charset.forName(charset));
StringBuilder sb = new StringBuilder(filteredStr.length());
for (int i = 0; i < filteredStr.length(); i++) {
    char c1 = filteredStr.charAt(i);
    char c2 = translatedStr.charAt(i);

    if (c1 == c2) {
        sb.append(c1);
    } else {
        sb.append(alt);
    }

    String x1 = Integer.toHexString(c1).toUpperCase();
    String x2 = Integer.toHexString(c2).toUpperCase();

    System.out.println(x1 + ":" + x2);
}

String str2 = sb.toString();
System.out.println("---");
System.out.println(str);
System.out.println("---");
System.out.println(filteredStr);
System.out.println("---");
System.out.println(str2);
System.out.println("---");
System.out.println(new String(str2.getBytes(charset), Charset.forName(charset)));