// AES-128 forward cipher (FIPS-197). QUIC packet protection needs only the encrypt
// direction: AES-128-GCM builds on it through CTR mode, and header protection
// (RFC 9001 §5.4.3) samples a single ECB block. The inverse cipher is not needed and
// is left out rather than carried dead.

///|
/// The AES S-box (FIPS-197 §5.1.1): the GF(2^8) multiplicative inverse followed by
/// the fixed affine transform.
let aes_sbox : Array[Int] = [
  0x63, 0x7c, 0x77, 0x7b, 0xf2, 0x6b, 0x6f, 0xc5, 0x30, 0x01, 0x67, 0x2b, 0xfe, 0xd7,
  0xab, 0x76, 0xca, 0x82, 0xc9, 0x7d, 0xfa, 0x59, 0x47, 0xf0, 0xad, 0xd4, 0xa2, 0xaf,
  0x9c, 0xa4, 0x72, 0xc0, 0xb7, 0xfd, 0x93, 0x26, 0x36, 0x3f, 0xf7, 0xcc, 0x34, 0xa5,
  0xe5, 0xf1, 0x71, 0xd8, 0x31, 0x15, 0x04, 0xc7, 0x23, 0xc3, 0x18, 0x96, 0x05, 0x9a,
  0x07, 0x12, 0x80, 0xe2, 0xeb, 0x27, 0xb2, 0x75, 0x09, 0x83, 0x2c, 0x1a, 0x1b, 0x6e,
  0x5a, 0xa0, 0x52, 0x3b, 0xd6, 0xb3, 0x29, 0xe3, 0x2f, 0x84, 0x53, 0xd1, 0x00, 0xed,
  0x20, 0xfc, 0xb1, 0x5b, 0x6a, 0xcb, 0xbe, 0x39, 0x4a, 0x4c, 0x58, 0xcf, 0xd0, 0xef,
  0xaa, 0xfb, 0x43, 0x4d, 0x33, 0x85, 0x45, 0xf9, 0x02, 0x7f, 0x50, 0x3c, 0x9f, 0xa8,
  0x51, 0xa3, 0x40, 0x8f, 0x92, 0x9d, 0x38, 0xf5, 0xbc, 0xb6, 0xda, 0x21, 0x10, 0xff,
  0xf3, 0xd2, 0xcd, 0x0c, 0x13, 0xec, 0x5f, 0x97, 0x44, 0x17, 0xc4, 0xa7, 0x7e, 0x3d,
  0x64, 0x5d, 0x19, 0x73, 0x60, 0x81, 0x4f, 0xdc, 0x22, 0x2a, 0x90, 0x88, 0x46, 0xee,
  0xb8, 0x14, 0xde, 0x5e, 0x0b, 0xdb, 0xe0, 0x32, 0x3a, 0x0a, 0x49, 0x06, 0x24, 0x5c,
  0xc2, 0xd3, 0xac, 0x62, 0x91, 0x95, 0xe4, 0x79, 0xe7, 0xc8, 0x37, 0x6d, 0x8d, 0xd5,
  0x4e, 0xa9, 0x6c, 0x56, 0xf4, 0xea, 0x65, 0x7a, 0xae, 0x08, 0xba, 0x78, 0x25, 0x2e,
  0x1c, 0xa6, 0xb4, 0xc6, 0xe8, 0xdd, 0x74, 0x1f, 0x4b, 0xbd, 0x8b, 0x8a, 0x70, 0x3e,
  0xb5, 0x66, 0x48, 0x03, 0xf6, 0x0e, 0x61, 0x35, 0x57, 0xb9, 0x86, 0xc1, 0x1d, 0x9e,
  0xe1, 0xf8, 0x98, 0x11, 0x69, 0xd9, 0x8e, 0x94, 0x9b, 0x1e, 0x87, 0xe9, 0xce, 0x55,
  0x28, 0xdf, 0x8c, 0xa1, 0x89, 0x0d, 0xbf, 0xe6, 0x42, 0x68, 0x41, 0x99, 0x2d, 0x0f,
  0xb0, 0x54, 0xbb, 0x16,
]

///|
/// The AES round constants (FIPS-197 §5.2), the high byte of each Rcon word.
let aes_rcon : Array[Int] = [
  0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x1b, 0x36,
]

///|
/// Multiply by x (0x02) in GF(2^8) with the AES reduction polynomial (FIPS-197 §4.2).
fn aes_xtime(b : Int) -> Int {
  let s = (b << 1) & 0xff
  if (b & 0x80) != 0 {
    s ^ 0x1b
  } else {
    s
  }
}

///|
/// The 11 AES-128 round keys as 176 flattened bytes (FIPS-197 §5.2 KeyExpansion).
/// Round `r` occupies bytes `16*r ..< 16*r+16`.
pub fn aes128_key_schedule(key : Bytes) -> Array[Int] {
  let w = Array::make(176, 0)
  for i = 0; i < 16; i = i + 1 {
    w[i] = key[i].to_int()
  }
  for i = 16; i < 176; i = i + 4 {
    let mut t0 = w[i - 4]
    let mut t1 = w[i - 3]
    let mut t2 = w[i - 2]
    let mut t3 = w[i - 1]
    if i % 16 == 0 {
      // RotWord, then SubWord, then XOR the round constant into the first byte.
      let r0 = aes_sbox[t1]
      let r1 = aes_sbox[t2]
      let r2 = aes_sbox[t3]
      let r3 = aes_sbox[t0]
      t0 = r0 ^ aes_rcon[i / 16 - 1]
      t1 = r1
      t2 = r2
      t3 = r3
    }
    w[i] = w[i - 16] ^ t0
    w[i + 1] = w[i - 15] ^ t1
    w[i + 2] = w[i - 14] ^ t2
    w[i + 3] = w[i - 13] ^ t3
  }
  w
}

///|
/// Encrypt one 16-byte block under the expanded key (FIPS-197 §5.1 Cipher). The state
/// is column-major: byte `k` is row `k % 4`, column `k / 4`.
pub fn aes128_encrypt_block(schedule : Array[Int], block : Bytes) -> Bytes {
  let s = Array::make(16, 0)
  for i = 0; i < 16; i = i + 1 {
    s[i] = block[i].to_int() ^ schedule[i]
  }
  for round = 1; round <= 10; round = round + 1 {
    // SubBytes.
    for i = 0; i < 16; i = i + 1 {
      s[i] = aes_sbox[s[i]]
    }
    // ShiftRows: row r rotates left by r.
    let t = Array::make(16, 0)
    for r = 0; r < 4; r = r + 1 {
      for c = 0; c < 4; c = c + 1 {
        t[r + 4 * c] = s[r + 4 * ((c + r) % 4)]
      }
    }
    for i = 0; i < 16; i = i + 1 {
      s[i] = t[i]
    }
    // MixColumns, skipped on the final round.
    if round != 10 {
      for c = 0; c < 4; c = c + 1 {
        let a0 = s[4 * c]
        let a1 = s[4 * c + 1]
        let a2 = s[4 * c + 2]
        let a3 = s[4 * c + 3]
        s[4 * c] = aes_xtime(a0) ^ (aes_xtime(a1) ^ a1) ^ a2 ^ a3
        s[4 * c + 1] = a0 ^ aes_xtime(a1) ^ (aes_xtime(a2) ^ a2) ^ a3
        s[4 * c + 2] = a0 ^ a1 ^ aes_xtime(a2) ^ (aes_xtime(a3) ^ a3)
        s[4 * c + 3] = aes_xtime(a0) ^ a0 ^ a1 ^ a2 ^ aes_xtime(a3)
      }
    }
    // AddRoundKey.
    for i = 0; i < 16; i = i + 1 {
      s[i] = s[i] ^ schedule[16 * round + i]
    }
  }
  let out = Buffer()
  for i = 0; i < 16; i = i + 1 {
    out.write_byte(s[i].to_byte())
  }
  out.to_bytes()
}