1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
use crate::Error;
#[cfg(feature = "alloc")]
use alloc::vec::Vec;

use core::convert::TryInto;
use core::fmt::Debug;

use unsigned_varint::encode as varint_encode;

#[cfg(feature = "std")]
use std::io;

#[cfg(not(feature = "std"))]
use core2::io;

/// A Multihash instance that only supports the basic functionality and no hashing.
///
/// With this Multihash implementation you can operate on Multihashes in a generic way, but
/// no hasher implementation is associated with the code.
///
/// # Example
///
/// ```
/// use multihash::Multihash;
///
/// const Sha3_256: u64 = 0x16;
/// let digest_bytes = [
///     0x16, 0x20, 0x64, 0x4b, 0xcc, 0x7e, 0x56, 0x43, 0x73, 0x04, 0x09, 0x99, 0xaa, 0xc8, 0x9e,
///     0x76, 0x22, 0xf3, 0xca, 0x71, 0xfb, 0xa1, 0xd9, 0x72, 0xfd, 0x94, 0xa3, 0x1c, 0x3b, 0xfb,
///     0xf2, 0x4e, 0x39, 0x38,
/// ];
/// let mh = Multihash::<32>::from_bytes(&digest_bytes).unwrap();
/// assert_eq!(mh.code(), Sha3_256);
/// assert_eq!(mh.size(), 32);
/// assert_eq!(mh.digest(), &digest_bytes[2..]);
/// ```
#[derive(Clone, Copy, Debug, Eq, Ord, PartialOrd)]
pub struct Multihash<const S: usize> {
    /// The code of the Multihash.
    code: u64,
    /// The actual size of the digest in bytes (not the allocated size).
    size: u8,
    /// The digest.
    digest: [u8; S],
}

impl<const S: usize> Default for Multihash<S> {
    fn default() -> Self {
        Self {
            code: 0,
            size: 0,
            digest: [0; S],
        }
    }
}

impl<const S: usize> Multihash<S> {
    /// Wraps the digest in a multihash.
    pub const fn wrap(code: u64, input_digest: &[u8]) -> Result<Self, Error> {
        if input_digest.len() > S {
            return Err(Error::invalid_size(input_digest.len() as _));
        }
        let size = input_digest.len();
        let mut digest = [0; S];
        let mut i = 0;
        while i < size {
            digest[i] = input_digest[i];
            i += 1;
        }
        Ok(Self {
            code,
            size: size as u8,
            digest,
        })
    }

    /// Returns the code of the multihash.
    pub const fn code(&self) -> u64 {
        self.code
    }

    /// Returns the size of the digest.
    pub const fn size(&self) -> u8 {
        self.size
    }

    /// Returns the digest.
    pub fn digest(&self) -> &[u8] {
        &self.digest[..self.size as usize]
    }

    /// Reads a multihash from a byte stream.
    pub fn read<R: io::Read>(r: R) -> Result<Self, Error>
    where
        Self: Sized,
    {
        let (code, size, digest) = read_multihash(r)?;
        Ok(Self { code, size, digest })
    }

    /// Parses a multihash from a bytes.
    ///
    /// You need to make sure the passed in bytes have the correct length. The digest length
    /// needs to match the `size` value of the multihash.
    pub fn from_bytes(mut bytes: &[u8]) -> Result<Self, Error>
    where
        Self: Sized,
    {
        let result = Self::read(&mut bytes)?;
        // There were more bytes supplied than read
        if !bytes.is_empty() {
            return Err(Error::invalid_size(bytes.len().try_into().expect(
                "Currently the maximum size is 255, therefore always fits into usize",
            )));
        }

        Ok(result)
    }

    /// Writes a multihash to a byte stream, returning the written size.
    pub fn write<W: io::Write>(&self, w: W) -> Result<usize, Error> {
        write_multihash(w, self.code(), self.size(), self.digest())
    }

    /// Returns the length in bytes needed to encode this multihash into bytes.
    pub fn encoded_len(&self) -> usize {
        let mut code_buf = varint_encode::u64_buffer();
        let code = varint_encode::u64(self.code, &mut code_buf);

        let mut size_buf = varint_encode::u8_buffer();
        let size = varint_encode::u8(self.size, &mut size_buf);

        code.len() + size.len() + usize::from(self.size)
    }

    #[cfg(feature = "alloc")]
    /// Returns the bytes of a multihash.
    pub fn to_bytes(&self) -> Vec<u8> {
        let mut bytes = Vec::with_capacity(self.size().into());
        let written = self
            .write(&mut bytes)
            .expect("writing to a vec should never fail");
        debug_assert_eq!(written, bytes.len());
        bytes
    }

    /// Truncates the multihash to the given size. It's up to the caller to ensure that the new size
    /// is secure (cryptographically) to use.
    ///
    /// If the new size is larger than the current size, this method does nothing.
    pub fn truncate(&self, size: u8) -> Self {
        let mut mh = *self;
        mh.size = mh.size.min(size);
        mh
    }

    /// Resizes the backing multihash buffer.
    ///
    /// This function fails if the hash digest is larger than the target size.
    pub fn resize<const R: usize>(&self) -> Result<Multihash<R>, Error> {
        let size = self.size as usize;
        if size > R {
            return Err(Error::invalid_size(self.size as u64));
        }
        let mut mh = Multihash {
            code: self.code,
            size: self.size,
            digest: [0; R],
        };
        mh.digest[..size].copy_from_slice(&self.digest[..size]);
        Ok(mh)
    }

    /// Decomposes struct, useful when needing a `Sized` array or moving all the data into another type
    ///
    /// It is recommended to use `digest()` `code()` and `size()` for most cases.
    pub fn into_inner(self) -> (u64, [u8; S], u8) {
        let Self { code, digest, size } = self;
        (code, digest, size)
    }
}

// Don't hash the whole allocated space, but just the actual digest
#[allow(clippy::derived_hash_with_manual_eq)]
impl<const S: usize> core::hash::Hash for Multihash<S> {
    fn hash<T: core::hash::Hasher>(&self, state: &mut T) {
        self.code.hash(state);
        self.digest().hash(state);
    }
}

#[cfg(feature = "alloc")]
impl<const S: usize> From<Multihash<S>> for Vec<u8> {
    fn from(multihash: Multihash<S>) -> Self {
        multihash.to_bytes()
    }
}

impl<const A: usize, const B: usize> PartialEq<Multihash<B>> for Multihash<A> {
    fn eq(&self, other: &Multihash<B>) -> bool {
        // NOTE: there's no need to explicitly check the sizes, that's implicit in the digest.
        self.code == other.code && self.digest() == other.digest()
    }
}

#[cfg(feature = "scale-codec")]
impl<const S: usize> parity_scale_codec::Encode for Multihash<S> {
    fn encode_to<EncOut: parity_scale_codec::Output + ?Sized>(&self, dest: &mut EncOut) {
        self.code.encode_to(dest);
        self.size.encode_to(dest);
        // **NOTE** We write the digest directly to dest, since we have known the size of digest.
        //
        // We do not choose to encode &[u8] directly, because it will add extra bytes (the compact length of digest).
        // For a valid multihash, the length of digest must equal to `size`.
        // Therefore, we can only read raw bytes whose length is equal to `size` when decoding.
        dest.write(self.digest());
    }
}

#[cfg(feature = "scale-codec")]
impl<const S: usize> parity_scale_codec::EncodeLike for Multihash<S> {}

#[cfg(feature = "scale-codec")]
impl<const S: usize> parity_scale_codec::Decode for Multihash<S> {
    fn decode<DecIn: parity_scale_codec::Input>(
        input: &mut DecIn,
    ) -> Result<Self, parity_scale_codec::Error> {
        let mut mh = Multihash {
            code: parity_scale_codec::Decode::decode(input)?,
            size: parity_scale_codec::Decode::decode(input)?,
            digest: [0; S],
        };
        if mh.size as usize > S {
            return Err(parity_scale_codec::Error::from("invalid size"));
        }
        // For a valid multihash, the length of digest must equal to the size.
        input.read(&mut mh.digest[..mh.size as usize])?;
        Ok(mh)
    }
}

/// Writes the multihash to a byte stream.
fn write_multihash<W>(mut w: W, code: u64, size: u8, digest: &[u8]) -> Result<usize, Error>
where
    W: io::Write,
{
    let mut code_buf = varint_encode::u64_buffer();
    let code = varint_encode::u64(code, &mut code_buf);

    let mut size_buf = varint_encode::u8_buffer();
    let size = varint_encode::u8(size, &mut size_buf);

    let written = code.len() + size.len() + digest.len();

    w.write_all(code)
        .map_err(crate::error::io_to_multihash_error)?;
    w.write_all(size)
        .map_err(crate::error::io_to_multihash_error)?;
    w.write_all(digest)
        .map_err(crate::error::io_to_multihash_error)?;

    Ok(written)
}

/// Reads a multihash from a byte stream that contains a full multihash (code, size and the digest)
///
/// Returns the code, size and the digest. The size is the actual size and not the
/// maximum/allocated size of the digest.
///
/// Currently the maximum size for a digest is 255 bytes.
fn read_multihash<R, const S: usize>(mut r: R) -> Result<(u64, u8, [u8; S]), Error>
where
    R: io::Read,
{
    let code = read_u64(&mut r)?;
    let size = read_u64(&mut r)?;

    if size > S as u64 || size > u8::MAX as u64 {
        return Err(Error::invalid_size(size));
    }

    let mut digest = [0; S];
    r.read_exact(&mut digest[..size as usize])
        .map_err(crate::error::io_to_multihash_error)?;
    Ok((code, size as u8, digest))
}

#[cfg(feature = "std")]
pub(crate) fn read_u64<R: io::Read>(r: R) -> Result<u64, Error> {
    unsigned_varint::io::read_u64(r).map_err(crate::error::unsigned_varint_to_multihash_error)
}

/// Reads 64 bits from a byte array into a u64
/// Adapted from unsigned-varint's generated read_u64 function at
/// https://github.com/paritytech/unsigned-varint/blob/master/src/io.rs
#[cfg(not(feature = "std"))]
pub(crate) fn read_u64<R: io::Read>(mut r: R) -> Result<u64, Error> {
    use unsigned_varint::decode;
    let mut b = varint_encode::u64_buffer();
    for i in 0..b.len() {
        let n = r
            .read(&mut (b[i..i + 1]))
            .map_err(crate::error::io_to_multihash_error)?;
        if n == 0 {
            return Err(Error::insufficient_varint_bytes());
        } else if decode::is_last(b[i]) {
            return decode::u64(&b[..=i])
                .map(|decoded| decoded.0)
                .map_err(crate::error::unsigned_varint_decode_to_multihash_error);
        }
    }
    Err(Error::varint_overflow())
}

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

    #[test]
    #[cfg(feature = "scale-codec")]
    fn test_scale() {
        use parity_scale_codec::{Decode, Encode};

        let mh1 = Multihash::<32>::wrap(0, b"hello world").unwrap();
        // println!("mh1: code = {}, size = {}, digest = {:?}", mh1.code(), mh1.size(), mh1.digest());
        let mh1_bytes = mh1.encode();
        // println!("Multihash<32>: {}", hex::encode(&mh1_bytes));
        let mh2: Multihash<32> = Decode::decode(&mut &mh1_bytes[..]).unwrap();
        assert_eq!(mh1, mh2);

        let mh3 = Multihash::<64>::wrap(0, b"hello world").unwrap();
        // println!("mh3: code = {}, size = {}, digest = {:?}", mh3.code(), mh3.size(), mh3.digest());
        let mh3_bytes = mh3.encode();
        // println!("Multihash<64>: {}", hex::encode(&mh3_bytes));
        let mh4: Multihash<64> = Decode::decode(&mut &mh3_bytes[..]).unwrap();
        assert_eq!(mh3, mh4);

        assert_eq!(mh1_bytes, mh3_bytes);
    }

    #[test]
    fn test_eq_sizes() {
        let mh1 = Multihash::<32>::default();
        let mh2 = Multihash::<64>::default();
        assert_eq!(mh1, mh2);
    }

    #[test]
    fn decode_non_minimal_error() {
        // This is a non-minimal varint.
        let data = [241, 0, 0, 0, 0, 0, 128, 132, 132, 132, 58];
        let result = read_u64(&data[..]);
        assert!(result.is_err());
    }
}