libcore/io/Base64.java

/*
 * Copyright (C) 2015 The Android Open Source Project
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License
 */

package libcore.io;

import java.io.ByteArrayOutputStream;
import java.nio.charset.StandardCharsets;

/**
 * Perform encoding and decoding of Base64 byte arrays as described in
 * http://www.ietf.org/rfc/rfc2045.txt
 */
public final class Base64 {
    private static final byte[] BASE_64_ALPHABET = initializeBase64Alphabet();

    private static byte[] initializeBase64Alphabet() {
        return "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"
                    .getBytes(StandardCharsets.US_ASCII);
    }

    // Bit masks for the 4 output 6-bit values from 3 input bytes.
    private static final int FIRST_OUTPUT_BYTE_MASK = 0x3f << 18;
    private static final int SECOND_OUTPUT_BYTE_MASK = 0x3f << 12;
    private static final int THIRD_OUTPUT_BYTE_MASK = 0x3f << 6;
    private static final int FOURTH_OUTPUT_BYTE_MASK = 0x3f;

    private Base64() {}

    public static String encode(byte[] in) {
        int len = in.length;
        int outputLen = computeEncodingOutputLen(len);
        byte[] output = new byte[outputLen];

        int outputIndex = 0;
        for (int i = 0; i < len; i += 3) {
            // Only a "triplet" if there are there are at least three remaining bytes
            // in the input...
            // Mask with 0xff to avoid signed extension.
            int byteTripletAsInt = in[i] & 0xff;
            if (i + 1 < len) {
                // Add second byte to the triplet.
                byteTripletAsInt <<= 8;
                byteTripletAsInt |= in[i + 1] & 0xff;
                if (i + 2 < len) {
                    byteTripletAsInt <<= 8;
                    byteTripletAsInt |= in[i + 2] & 0xff;
                } else {
                    // Insert 2 zero bits as to make output 18 bits long.
                    byteTripletAsInt <<= 2;
                }
            } else {
                // Insert 4 zero bits as to make output 12 bits long.
                byteTripletAsInt <<= 4;
            }

            if (i + 2 < len) {
                // The int may have up to 24 non-zero bits.
                output[outputIndex++] = BASE_64_ALPHABET[
                        (byteTripletAsInt & FIRST_OUTPUT_BYTE_MASK) >>> 18];
            }
            if (i + 1 < len) {
                // The int may have up to 18 non-zero bits.
                output[outputIndex++] = BASE_64_ALPHABET[
                        (byteTripletAsInt & SECOND_OUTPUT_BYTE_MASK) >>> 12];
            }
            output[outputIndex++] = BASE_64_ALPHABET[
                    (byteTripletAsInt & THIRD_OUTPUT_BYTE_MASK) >>> 6];
            output[outputIndex++] = BASE_64_ALPHABET[
                    byteTripletAsInt & FOURTH_OUTPUT_BYTE_MASK];
        }

        int inLengthMod3 = len % 3;
        // Add padding as per the spec.
        if (inLengthMod3 > 0) {
            output[outputIndex++] = '=';
            if (inLengthMod3 == 1) {
                output[outputIndex++] = '=';
            }
        }

        return new String(output, StandardCharsets.US_ASCII);
    }

    private static int computeEncodingOutputLen(int inLength) {
        int inLengthMod3 = inLength % 3;
        int outputLen = (inLength / 3) * 4;
        if (inLengthMod3 == 2) {
            // Need 3 6-bit characters as to express the last 16 bits, plus 1 padding.
            outputLen += 4;
        } else if (inLengthMod3 == 1) {
            // Need 2 6-bit characters as to express the last 8 bits, plus 2 padding.
            outputLen += 4;
        }
        return outputLen;
    }

    public static byte[] decode(byte[] in) {
        return decode(in, in.length);
    }

    /** Decodes the input from position 0 (inclusive) to len (exclusive). */
    public static byte[] decode(byte[] in, int len) {
        final int inLength = Math.min(in.length, len);
        // Overestimating 3 bytes per each 4 blocks of input (plus a possibly incomplete one).
        ByteArrayOutputStream output = new ByteArrayOutputStream((inLength / 4) * 3 + 3);
        // Position in the input. Use an array so we can pass it to {@code getNextByte}.
        int[] pos = new int[1];

        try {
            while (pos[0] < inLength) {
                int byteTripletAsInt = 0;

                // j is the index in a 4-tuple of 6-bit characters where are trying to read from the
                // input.
                for (int j = 0; j < 4; j++) {
                    byte c = getNextByte(in, pos, inLength);
                    if (c == END_OF_INPUT || c == PAD_AS_BYTE) {
                        // Padding or end of file...
                        switch (j) {
                            case 0:
                            case 1:
                                return (c == END_OF_INPUT) ? output.toByteArray() : null;
                            case 2:
                                // The input is over with two 6-bit characters: a single byte padded
                                // with 4 extra 0's.

                                if (c == END_OF_INPUT) {
                                    // Do not consider the block, since padding is not present.
                                    return checkNoTrailingAndReturn(output, in, pos[0], inLength);
                                }
                                // We are at a pad character, consume and look for the second one.
                                pos[0]++;
                                c = getNextByte(in, pos, inLength);
                                if (c == END_OF_INPUT) {
                                    // Do not consider the block, since padding is not present.
                                    return checkNoTrailingAndReturn(output, in, pos[0], inLength);
                                }
                                if (c == PAD_AS_BYTE) {
                                    byteTripletAsInt >>= 4;
                                    output.write(byteTripletAsInt);
                                    return checkNoTrailingAndReturn(output, in, pos[0], inLength);
                                }
                                // Something other than pad and non-alphabet characters, illegal.
                                return null;


                            case 3:
                                // The input is over with three 6-bit characters: two bytes padded
                                // with 2 extra 0's.
                                if (c == PAD_AS_BYTE) {
                                    // Consider the block only if padding is present.
                                    byteTripletAsInt >>= 2;
                                    output.write(byteTripletAsInt >> 8);
                                    output.write(byteTripletAsInt & 0xff);
                                }
                                return checkNoTrailingAndReturn(output, in, pos[0], inLength);
                        }
                    } else {
                        byteTripletAsInt <<= 6;
                        byteTripletAsInt += (c & 0xff);
                        pos[0]++;
                    }
                }
                // We have four 6-bit characters: output the corresponding 3 bytes
                output.write(byteTripletAsInt >> 16);
                output.write((byteTripletAsInt >> 8) & 0xff);
                output.write(byteTripletAsInt & 0xff);
            }
            return checkNoTrailingAndReturn(output, in, pos[0], inLength);
        } catch (InvalidBase64ByteException e) {
            return null;
        }
    }

    /**
     * On decoding, an illegal character always return null.
     *
     * Using this exception to avoid "if" checks every time.
     */

    private static class InvalidBase64ByteException extends Exception { }

    /**
     * Obtain the numeric value corresponding to the next relevant byte in the input.
     *
     * Calculates the numeric value (6-bit, 0 <= x <= 63) of the next Base64 encoded byte in
     * {@code in} at or after {@code pos[0]} and before {@code inLength}. Returns
     * {@link #WHITESPACE_AS_BYTE}, {@link #PAD_AS_BYTE}, {@link #END_OF_INPUT} or the 6-bit value.
     * {@code pos[0]} is updated as a side effect of this method.
     */
    private static byte getNextByte(byte[] in, int[] pos, int inLength)
            throws InvalidBase64ByteException {
        // Ignore all whitespace.
        while (pos[0] < inLength) {
            byte c = base64AlphabetToNumericalValue(in[pos[0]]);
            if (c != WHITESPACE_AS_BYTE) {
                return c;
            }
            pos[0]++;
        }
        return END_OF_INPUT;
    }

    /**
     * Check that there are no invalid trailing characters (ie, other then whitespace and padding)
     *
     * Returns {@code output} as a byte array in case of success, {@code null} in case of invalid
     * characters.
     */
    private static byte[] checkNoTrailingAndReturn(
            ByteArrayOutputStream output, byte[] in, int i, int inLength)
                    throws InvalidBase64ByteException{
        while (i < inLength) {
            byte c = base64AlphabetToNumericalValue(in[i]);
            if (c != WHITESPACE_AS_BYTE && c != PAD_AS_BYTE) {
                return null;
            }
            i++;
        }
        return output.toByteArray();
    }

    private static final byte PAD_AS_BYTE = -1;
    private static final byte WHITESPACE_AS_BYTE = -2;
    private static final byte END_OF_INPUT = -3;
    private static byte base64AlphabetToNumericalValue(byte c) throws InvalidBase64ByteException {
        if ('A' <= c && c <= 'Z') {
            return (byte) (c - 'A');
        }
        if ('a' <= c && c <= 'z') {
            return (byte) (c - 'a' + 26);
        }
        if ('0' <= c && c <= '9') {
            return (byte) (c - '0' + 52);
        }
        if (c == '+') {
            return (byte) 62;
        }
        if (c == '/') {
            return (byte) 63;
        }
        if (c == '=') {
            return PAD_AS_BYTE;
        }
        if (c == ' ' || c == '\t' || c == '\r' || c == '\n') {
            return WHITESPACE_AS_BYTE;
        }
        throw new InvalidBase64ByteException();
    }
}