Various programmatic hashes. / published by https://github.com/dacr/code-examples-manager #b38b4236-7059-4f97-8c82-3172c4016f3a/1a322ce2a552a75bdc6807eb90dea1cc61e94e4e

hashes.sc

// summary : Various programmatic hashes.
// keywords : scala, hashes, md5sum, sha1, murmurHash3, @testable
// publish : gist
// authors : David Crosson
// license : Apache NON-AI License Version 2.0 (https://raw.githubusercontent.com/non-ai-licenses/non-ai-licenses/main/NON-AI-APACHE2)
// id : b38b4236-7059-4f97-8c82-3172c4016f3a
// created-on : 2020-05-31T19:54:52Z
// managed-by : https://github.com/dacr/code-examples-manager
// run-with : scala-cli $file

// ---------------------
//> using scala "3.4.2"
//> using dep "org.scalatest::scalatest:3.2.10"
// ---------------------
import org.scalatest.*
import flatspec.*
import matchers.*

import java.io.*
import java.nio.file.Path

object Hashes {
  def md5sum(that: String): String =
    md5sum(new ByteArrayInputStream(that.getBytes())) // TODO : Warning manage charsets...

  def md5sum(input: InputStream): String = {
    val bis = new BufferedInputStream(input)
    val buf = new Array[Byte](1024)
    val md5 = java.security.MessageDigest.getInstance("MD5")
    LazyList.continually(bis.read(buf)).takeWhile(_ != -1).foreach(md5.update(buf, 0, _))
    md5.digest().map(0xFF & _).map { "%02x".format(_) }.foldLeft("") { _ + _ }
  }

  def murmurHash3(that:String):Int =
    scala.util.hashing.MurmurHash3.stringHash(that)


  def sha1(that: String): String =
    // Inspired from https://alvinalexander.com/source-code/scala-method-create-md5-hash-of-string
    import java.security.MessageDigest
    import java.math.BigInteger
    val md = MessageDigest.getInstance("SHA-1")
    val digest = md.digest(that.getBytes)
    val bigInt = new BigInteger(1, digest)
    val hashedString = bigInt.toString(16)
    hashedString


  def fileDigest(path: Path, algo:String="SHA-256"): String =
    import java.math.BigInteger
    import java.security.{MessageDigest, DigestInputStream}
    import java.io.FileInputStream
    val buffer = new Array[Byte](8192)
    val md5    = MessageDigest.getInstance(algo)
    val dis    = new DigestInputStream(new FileInputStream(path.toFile), md5)
    try { while (dis.read(buffer) != -1) {} }
    finally { dis.close() }
    md5.digest.map("%02x".format(_)).mkString

}

import Hashes._

object HashesTest extends AnyFlatSpec with should.Matchers {
  override def suiteName: String = "HashesTest"

  val example = "Please hash me !"

  "murmurHash3" should "return the right hash" in {
    murmurHash3(example) shouldBe 370020140
  }

  "md5sum" should "return the right hash" in {
    md5sum(example) shouldBe "55340948b4b044ad6d1632908c86e765"
  }

  "sha1" should "return the right hash" in {
    sha1(example) shouldBe "4031d74d6a72919da236a388bdf3b966126b80f2"
  }
}

HashesTest.execute()