object-hash 0.1.9

//! SHA-1, used to turn the canonical form into a short digest.
//!
//! Written out rather than pulled in, because the whole point of this crate is
//! to have no dependencies, and SHA-1 is short. It is not a security primitive
//! here and is not used as one: the digest names a value, it does not protect
//! it. If a stronger digest is wanted, hash the canonical bytes with any other
//! function.

/// The SHA-1 digest of `data`.
pub(crate) fn digest(data: &[u8]) -> [u8; 20] {
    let mut state: [u32; 5] = [
        0x6745_2301,
        0xefcd_ab89,
        0x98ba_dcfe,
        0x1032_5476,
        0xc3d2_e1f0,
    ];
    let bit_length = (data.len() as u64).wrapping_mul(8);

    let mut message = data.to_vec();
    message.push(0x80);
    while message.len() % 64 != 56 {
        message.push(0);
    }
    message.extend_from_slice(&bit_length.to_be_bytes());

    for block in message.as_chunks::<64>().0 {
        let mut words = [0u32; 80];
        for (index, word) in words.iter_mut().take(16).enumerate() {
            let start = index * 4;
            *word = u32::from_be_bytes([
                block[start],
                block[start + 1],
                block[start + 2],
                block[start + 3],
            ]);
        }
        for index in 16..80 {
            words[index] =
                (words[index - 3] ^ words[index - 8] ^ words[index - 14] ^ words[index - 16])
                    .rotate_left(1);
        }

        let [mut a, mut b, mut c, mut d, mut e] = state;
        for (index, word) in words.iter().enumerate() {
            let (f, k) = match index {
                0..=19 => ((b & c) | ((!b) & d), 0x5a82_7999),
                20..=39 => (b ^ c ^ d, 0x6ed9_eba1),
                40..=59 => ((b & c) | (b & d) | (c & d), 0x8f1b_bcdc),
                _ => (b ^ c ^ d, 0xca62_c1d6),
            };
            let temp = a
                .rotate_left(5)
                .wrapping_add(f)
                .wrapping_add(e)
                .wrapping_add(k)
                .wrapping_add(*word);
            e = d;
            d = c;
            c = b.rotate_left(30);
            b = a;
            a = temp;
        }

        state[0] = state[0].wrapping_add(a);
        state[1] = state[1].wrapping_add(b);
        state[2] = state[2].wrapping_add(c);
        state[3] = state[3].wrapping_add(d);
        state[4] = state[4].wrapping_add(e);
    }

    let mut out = [0u8; 20];
    for (index, value) in state.iter().enumerate() {
        out[index * 4..index * 4 + 4].copy_from_slice(&value.to_be_bytes());
    }
    out
}

#[cfg(test)]
mod tests {
    use super::*;

    /// The digest as lowercase hexadecimal, for the published vectors.
    fn hex(data: &[u8]) -> String {
        let mut out = String::new();
        for byte in digest(data) {
            out.push(char::from_digit(u32::from(byte >> 4), 16).expect("nibble"));
            out.push(char::from_digit(u32::from(byte & 0x0f), 16).expect("nibble"));
        }
        out
    }

    #[test]
    fn the_published_test_vectors_match() {
        // The three vectors from FIPS 180-1 and the well-known extras.
        assert_eq!(hex(b""), "da39a3ee5e6b4b0d3255bfef95601890afd80709");
        assert_eq!(hex(b"abc"), "a9993e364706816aba3e25717850c26c9cd0d89d");
        assert_eq!(
            hex(b"abcdbcdecdefdefgefghfghighijhijkijkljklmklmnlmnomnopnopq"),
            "84983e441c3bd26ebaae4aa1f95129e5e54670f1"
        );
        assert_eq!(
            hex(b"The quick brown fox jumps over the lazy dog"),
            "2fd4e1c67a2d28fced849ee1bb76e7391b93eb12"
        );
    }

    #[test]
    fn a_longer_message_spanning_many_blocks_matches() {
        // One million 'a', the FIPS long vector.
        let data = vec![b'a'; 1_000_000];
        assert_eq!(hex(&data), "34aa973cd4c4daa4f61eeb2bdbad27316534016f");
    }
}