Md5Chunking.java

import java.io.File;
import java.io.FileInputStream;
import java.io.IOException;
import java.security.MessageDigest;
import java.security.NoSuchAlgorithmException;

public class Md5Chunking {

    private static MessageDigest md;
    private static final int BLOCK_SIZE = 63;

    public static void main(String[] args) throws Exception {
        // Init
        md = MessageDigest.getInstance("MD5");

        // Read the entire file to memory for simplicity
        byte[] input = readInput("ubuntu-20.04.3-live-server-amd64.iso");

        long startTime = System.nanoTime();

        // Start at the position where we have enough bytes for BLOCK_SIZE
        int currentPos = BLOCK_SIZE - 1;
        int currentStartPos = 0;
        while (currentPos < input.length) {
            if (shouldSplit(input, currentPos)) {
                System.out.println(currentPos + "," + (currentPos - currentStartPos + 1));
                currentStartPos = currentPos + 1;
                currentPos += BLOCK_SIZE;
            } else {
                currentPos++;
            }
        }
        // Last chunk
        currentPos = input.length - 1;
        System.out.println(currentPos + "," + (currentPos - currentStartPos + 1));

        long endTime = System.nanoTime();
        System.out.println("Finished in " + (endTime - startTime) / 1000000 + " ms");
    }

    private static byte[] readInput(String filename) throws IOException {
        File file = new File(filename);
        try (FileInputStream fis = new FileInputStream(file)) {
            long len = file.length();
            byte[] content = new byte[(int) len];
            if (len != fis.read(content)) {
                throw new IOException("Unexpected bytes read");
            }
            return content;
        }
    }

    private static boolean shouldSplit(byte[] b, int pos) throws NoSuchAlgorithmException {
        md.reset();
        md.update(b, pos - (BLOCK_SIZE - 1), BLOCK_SIZE);
        byte[] result = md.digest();

        // We want first 23 bits to be 0 (first 2 bytes and 7 bits of the third byte)
        return result[0] == 0 && result[1] == 0 && (result[2] & 0b1111111) == 0;
    }

}