在Java中将Unicode值转换为字符串

Question

I am trying to extract currencies in my texts and I am getting currencies from db which contains special currency symbols as well. For example for the pound, I have unicode of pound "\£" in the db along with other identifiers such as "gbp" as well.

I am trying to get the corresponding symbol from the unicode and compare with my text in a loop as suggested in here .

But when I evaluate my code, the result is like in the image here:

private Optional<Currency> extractTokenWise(Iterable<String> tokens){
    try {
        for (String aToken : tokens) {
            for (String currency : currencies.keySet()) {
                for (String arep : currencies.get(currency)) {
                    if(arep.startsWith("\\")){ //special character for currency written in unicode representation                  
                        byte[] charset = arep.getBytes("UTF-8");
                        arep = new String(charset, "UTF-8");
                    }
                    if (aToken.equals(arep)) {
                        return Optional.of(Currency.findProperEnum(currency));
                    }
                }
            }
        }
    }catch (UnsupportedEncodingException e) {
        e.printStackTrace();
    }
    return Optional.empty();
}

It is interesting that when arep is equal to "\£" , it does not work but when I specifically give String value of "\£" , It produces the result I want. What am I missing here?

Answer 1

As mentioned in comments something like this should work:

if (arep.startsWith("\\u")) {
        arep = Character.toString((char) Integer.parseInt(arep.substring(2), 16));
}

Answer 2

I think you mix up unicode escape sequences in java code with strings containing such escape sequences.

String poundSign = "\£"; assigns poundSign a string containing the single character £ . This string has a length of 1 character. In memory and in the class file it will occupy 2 bytes.

It looks like arep contains the string \£ as assigned by String unicodeEscapeForPoundSign = "\\\£"; -- that's what your first if statement tests for. It contains the unicode escape sequence as used in java code, but not the character this escape sequence represents . It contains the 6 characters '\\', 'u', '0', '0', 'A', and '3' (as your IDE shows). arep.getBytes("UTF-8"); returns an array of just these characters and new String(charset, "UTF-8"); converts the array back to the string \£ and not the string £

The solution depends on what you get from your database . Assuming you have a mapping from the db-value to a Currency object or the ISO currency code, you won't need your first if statement, just make sure arep contains the correct string:

• String arep = "\£" (single pound character)
• String arep = "\\\£" (pound character java unicode escape string)