CODEWITHSUNDEEP
javaencryptioncharacter-encodingbase64
yamori
yamori

Reputation: 1243

What is the target charset when I encrypt a string with Cipher and Base64.Encoder?

I'm trying to accommodate encrypted tokens in a DSL I'm designing (i.e. I need a char to be used as delimiter). The encoder.encodeToString(...) docs says it uses the ISO-8859-1 charset. But when I encrypt a sampling of texts, it looks like it's not using all of the ISO-8859-1 charset, instead upper/lower-case and some symbols, and not certain punctuation and accented chars. What am I missing about this encodeToString() call and what is the final char domain?

//import java.util.Base64;
//import javax.crypto.Cipher;
//import javax.crypto.SecretKey;
static Cipher cipher;
public static String decrypt(String encryptedText, SecretKey secretKey) throws Exception {
    Base64.Decoder decoder = Base64.getDecoder();
    byte[] encryptedTextByte = decoder.decode(encryptedText);
    cipher.init(Cipher.DECRYPT_MODE, secretKey);
    byte[] decryptedByte = cipher.doFinal(encryptedTextByte);
    String decryptedText = new String(decryptedByte);
    return decryptedText;
}

Upvotes: 3

Views: 2909

Answers (4)

hagrawal7777
hagrawal7777

Reputation: 14668

Read the document line carefully:

This method first encodes all input bytes into a base64 encoded byte array and then constructs a new String by using the encoded byte array and the ISO-8859-1 charset.

This means 2 things:

  • Encoding scheme used is Base64, that's the reason you can not getting expected output because ISO-8859-1 is not used as encoding scheme.
  • ISO-8859-1 is only used as charset reference for constructing String.

Base64 class give your encoders and decoders for the Base64 encoding scheme, so if you are using its encoders and decoders then output will be encoded/decoded using Base64 encoding scheme.

This class consists exclusively of static methods for obtaining encoders and decoders for the Base64 encoding scheme.


So, what you need is to properly encode your byte array using some encoding scheme - UTF-8 or ISO-8859-1.

I would personally recommend using "UTF-8" because it is widely used + it encodes all ASCII and some Latin character in 1 byte, Unicode other BMP character using 2 byte and Unicode supplementary characters using 4 bytes. So, it is not space consuming for all ASCII and some Latin character + ability to encode all Unicode characters.

There are many ways you can find to encode your String using desired encoding scheme, an example below:

    byte[] byteArr = new byte[3];
    String decodeText = new String(byteArr);
    Charset.forName("UTF-8").encode(decodeText);

Upvotes: 0

Raniz
Raniz

Reputation: 11113

Base64 is named just so because it uses 64 characters from the ASCII table. The encoding used should not matter as long as it is compatible with ASCII.

If you want to use more than 64 characters you will have to use a different encoding.

Upvotes: 1

gavard.e
gavard.e

Reputation: 142

Encryption returns bytes, no matter the encryption is.

I personally use the following function to convert from decrypted bytes to string:

    public static String getStringFromBytes(byte[] data) {
    StringBuilder sb = new StringBuilder();
    if (data != null && data.length > 0) {
        for (byte b : data) {
            sb.append((char) (b));
        }
    }
    return sb.toString();
}

Upvotes: 0

Joop Eggen
Joop Eggen

Reputation: 109593

String has a constructor with charset; otherwise the default OS charset is taken.

new String(decryptedByte, StandardCharsets.ISO_8859_1);

As there often is a mix-up of Latin-1 (ISO-8859-1) with Windows Latin-1 (Windows-1252) you might try "Windows-1252" too.

Upvotes: 4

Related Questions