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
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
//! Functions for decoding Base58 encoded strings.

use core::fmt;

#[cfg(feature = "alloc")]
use alloc::vec::Vec;

use crate::Check;
#[cfg(any(feature = "check", feature = "cb58"))]
use crate::CHECKSUM_LEN;

use crate::Alphabet;

/// A builder for setting up the alphabet and output of a base58 decode.
///
/// See the documentation for [`bs58::decode`](crate::decode()) for a more
/// high level view of how to use this.
#[allow(missing_debug_implementations)]
pub struct DecodeBuilder<'a, I: AsRef<[u8]>> {
    input: I,
    alpha: &'a Alphabet,
    check: Check,
}

/// A specialized [`Result`](core::result::Result) type for [`bs58::decode`](module@crate::decode)
pub type Result<T> = core::result::Result<T, Error>;

/// Errors that could occur when decoding a Base58 encoded string.
#[derive(Copy, Clone, Debug, Eq, PartialEq)]
#[non_exhaustive]
pub enum Error {
    /// The output buffer was too small to contain the entire input.
    BufferTooSmall,

    /// The input contained a character that was not part of the current Base58
    /// alphabet.
    InvalidCharacter {
        /// The unexpected character.
        character: char,
        /// The (byte) index in the input string the character was at.
        index: usize,
    },

    /// The input contained a multi-byte (or non-utf8) character which is
    /// unsupported by this Base58 decoder.
    NonAsciiCharacter {
        /// The (byte) index in the input string the start of the character was
        /// at.
        index: usize,
    },

    #[cfg(any(feature = "check", feature = "cb58"))]
    /// The checksum did not match the payload bytes
    InvalidChecksum {
        ///The given checksum
        checksum: [u8; CHECKSUM_LEN],
        ///The checksum calculated for the payload
        expected_checksum: [u8; CHECKSUM_LEN],
    },

    #[cfg(any(feature = "check", feature = "cb58"))]
    /// The version did not match the payload bytes
    InvalidVersion {
        ///The given version
        ver: u8,
        ///The expected version
        expected_ver: u8,
    },

    #[cfg(any(feature = "check", feature = "cb58"))]
    ///Not enough bytes to have both a checksum and a payload (less than to CHECKSUM_LEN)
    NoChecksum,
}

/// Represents a buffer that can be decoded into. See [`DecodeBuilder::onto`] and the provided
/// implementations for more details.
pub trait DecodeTarget {
    /// Decodes into this buffer, provides the maximum length for implementations that wish to
    /// preallocate space, along with a function that will write bytes into the buffer and return
    /// the length written to it.
    fn decode_with(
        &mut self,
        max_len: usize,
        f: impl for<'a> FnOnce(&'a mut [u8]) -> Result<usize>,
    ) -> Result<usize>;
}

impl<T: DecodeTarget + ?Sized> DecodeTarget for &mut T {
    fn decode_with(
        &mut self,
        max_len: usize,
        f: impl for<'a> FnOnce(&'a mut [u8]) -> Result<usize>,
    ) -> Result<usize> {
        T::decode_with(self, max_len, f)
    }
}

#[cfg(feature = "alloc")]
impl DecodeTarget for Vec<u8> {
    fn decode_with(
        &mut self,
        max_len: usize,
        f: impl for<'a> FnOnce(&'a mut [u8]) -> Result<usize>,
    ) -> Result<usize> {
        let original = self.len();
        self.resize(original + max_len, 0);
        let len = f(&mut self[original..])?;
        self.truncate(original + len);
        Ok(len)
    }
}

#[cfg(feature = "smallvec")]
impl<A: smallvec::Array<Item = u8>> DecodeTarget for smallvec::SmallVec<A> {
    /// Decodes data into a [`smallvec::SmallVec`].
    fn decode_with(
        &mut self,
        max_len: usize,
        f: impl for<'a> FnOnce(&'a mut [u8]) -> Result<usize>,
    ) -> Result<usize> {
        let original = self.len();
        self.resize(original + max_len, 0);
        let len = f(&mut self[original..])?;
        self.truncate(original + len);
        Ok(len)
    }
}

#[cfg(feature = "tinyvec")]
impl<A: tinyvec::Array<Item = u8>> DecodeTarget for tinyvec::ArrayVec<A> {
    fn decode_with(
        &mut self,
        max_len: usize,
        f: impl for<'a> FnOnce(&'a mut [u8]) -> Result<usize>,
    ) -> Result<usize> {
        let _ = max_len;
        let original = self.len();
        let len = f(self.grab_spare_slice_mut())?;
        self.set_len(original + len);
        Ok(len)
    }
}

#[cfg(feature = "tinyvec")]
impl DecodeTarget for tinyvec::SliceVec<'_, u8> {
    fn decode_with(
        &mut self,
        max_len: usize,
        f: impl for<'a> FnOnce(&'a mut [u8]) -> Result<usize>,
    ) -> Result<usize> {
        let _ = max_len;
        let original = self.len();
        let len = f(self.grab_spare_slice_mut())?;
        self.set_len(original + len);
        Ok(len)
    }
}

#[cfg(all(feature = "tinyvec", feature = "alloc"))]
impl<A: tinyvec::Array<Item = u8>> DecodeTarget for tinyvec::TinyVec<A> {
    fn decode_with(
        &mut self,
        max_len: usize,
        f: impl for<'a> FnOnce(&'a mut [u8]) -> Result<usize>,
    ) -> Result<usize> {
        let original = self.len();
        self.resize(original + max_len, 0);
        let len = f(&mut self[original..])?;
        self.truncate(original + len);
        Ok(len)
    }
}

impl DecodeTarget for [u8] {
    fn decode_with(
        &mut self,
        max_len: usize,
        f: impl for<'a> FnOnce(&'a mut [u8]) -> Result<usize>,
    ) -> Result<usize> {
        let _ = max_len;
        f(&mut *self)
    }
}

impl<const N: usize> DecodeTarget for [u8; N] {
    fn decode_with(
        &mut self,
        max_len: usize,
        f: impl for<'a> FnOnce(&'a mut [u8]) -> Result<usize>,
    ) -> Result<usize> {
        let _ = max_len;
        f(&mut *self)
    }
}

impl<'a, I: AsRef<[u8]>> DecodeBuilder<'a, I> {
    /// Setup decoder for the given string using the given alphabet.
    /// Preferably use [`bs58::decode`](crate::decode()) instead of this directly.
    pub const fn new(input: I, alpha: &'a Alphabet) -> DecodeBuilder<'a, I> {
        DecodeBuilder {
            input,
            alpha,
            check: Check::Disabled,
        }
    }

    /// Setup decoder for the given string using default prepared alphabet.
    pub(crate) const fn from_input(input: I) -> DecodeBuilder<'static, I> {
        DecodeBuilder {
            input,
            alpha: Alphabet::DEFAULT,
            check: Check::Disabled,
        }
    }

    /// Change the alphabet that will be used for decoding.
    ///
    /// # Examples
    ///
    /// ```rust
    /// assert_eq!(
    ///     vec![0x60, 0x65, 0xe7, 0x9b, 0xba, 0x2f, 0x78],
    ///     bs58::decode("he11owor1d")
    ///         .with_alphabet(bs58::Alphabet::RIPPLE)
    ///         .into_vec()?);
    /// # Ok::<(), bs58::decode::Error>(())
    /// ```
    pub const fn with_alphabet(mut self, alpha: &'a Alphabet) -> DecodeBuilder<'a, I> {
        self.alpha = alpha;
        self
    }

    /// Expect and check checksum using the [Base58Check][] algorithm when
    /// decoding.
    ///
    /// Optional parameter for version byte. If provided, the version byte will
    /// be used in verification.
    ///
    /// [Base58Check]: https://en.bitcoin.it/wiki/Base58Check_encoding
    ///
    /// # Examples
    ///
    /// ```rust
    /// assert_eq!(
    ///     vec![0x2d, 0x31],
    ///     bs58::decode("PWEu9GGN")
    ///         .with_check(None)
    ///         .into_vec()?);
    /// # Ok::<(), bs58::decode::Error>(())
    /// ```
    #[cfg(feature = "check")]
    pub fn with_check(self, expected_ver: Option<u8>) -> DecodeBuilder<'a, I> {
        let check = Check::Enabled(expected_ver);
        DecodeBuilder { check, ..self }
    }

    /// Expect and check checksum using the [CB58][] algorithm when
    /// decoding.
    ///
    /// Optional parameter for version byte. If provided, the version byte will
    /// be used in verification.
    ///
    /// [CB58]: https://support.avax.network/en/articles/4587395-what-is-cb58
    ///
    /// # Examples
    ///
    /// ```rust
    /// assert_eq!(
    ///     vec![0x2d, 0x31],
    ///     bs58::decode("PWHVMzdR")
    ///         .as_cb58(None)
    ///         .into_vec()?);
    /// # Ok::<(), bs58::decode::Error>(())
    /// ```
    #[cfg(feature = "cb58")]
    pub fn as_cb58(self, expected_ver: Option<u8>) -> DecodeBuilder<'a, I> {
        let check = Check::CB58(expected_ver);
        DecodeBuilder { check, ..self }
    }
    /// Decode into a new vector of bytes.
    ///
    /// See the documentation for [`bs58::decode`](crate::decode()) for an
    /// explanation of the errors that may occur.
    ///
    /// # Examples
    ///
    /// ```rust
    /// assert_eq!(
    ///     vec![0x04, 0x30, 0x5e, 0x2b, 0x24, 0x73, 0xf0, 0x58],
    ///     bs58::decode("he11owor1d").into_vec()?);
    /// # Ok::<(), bs58::decode::Error>(())
    /// ```
    ///
    #[cfg(feature = "alloc")]
    pub fn into_vec(self) -> Result<Vec<u8>> {
        let mut output = Vec::new();
        self.onto(&mut output)?;
        Ok(output)
    }

    /// Decode into the given buffer.
    ///
    /// Returns the length written into the buffer.
    ///
    /// If the buffer is resizeable it will be extended and the new data will be written to the end
    /// of it.
    ///
    /// If the buffer is not resizeable bytes will be written from the beginning and bytes after
    /// the final encoded byte will not be touched.
    ///
    /// See the documentation for [`bs58::decode`](crate::decode()) for an
    /// explanation of the errors that may occur.
    ///
    /// # Examples
    ///
    /// ## `Vec<u8>`
    ///
    /// ```rust
    /// let mut output = b"hello ".to_vec();
    /// assert_eq!(5, bs58::decode("EUYUqQf").onto(&mut output)?);
    /// assert_eq!(b"hello world", output.as_slice());
    /// # Ok::<(), bs58::decode::Error>(())
    /// ```
    ///
    /// ## `&mut [u8]`
    ///
    /// ```rust
    /// let mut output = b"hello ".to_owned();
    /// assert_eq!(5, bs58::decode("EUYUqQf").onto(&mut output)?);
    /// assert_eq!(b"world ", output.as_ref());
    /// # Ok::<(), bs58::decode::Error>(())
    /// ```
    pub fn onto(self, mut output: impl DecodeTarget) -> Result<usize> {
        let max_decoded_len = self.input.as_ref().len();
        match self.check {
            Check::Disabled => output.decode_with(max_decoded_len, |output| {
                decode_into(self.input.as_ref(), output, self.alpha)
            }),
            #[cfg(feature = "check")]
            Check::Enabled(expected_ver) => output.decode_with(max_decoded_len, |output| {
                decode_check_into(self.input.as_ref(), output, self.alpha, expected_ver)
            }),
            #[cfg(feature = "cb58")]
            Check::CB58(expected_ver) => output.decode_with(max_decoded_len, |output| {
                decode_cb58_into(self.input.as_ref(), output, self.alpha, expected_ver)
            }),
        }
    }
}

/// For `const` compatibility we are restricted to using a concrete input and output type, as
/// `const` trait implementations and `&mut` are unstable. These methods will eventually be
/// deprecated once the primary interfaces can be converted into `const fn` directly.
impl<'a, 'b> DecodeBuilder<'a, &'b [u8]> {
    /// Decode into a new array.
    ///
    /// Returns the decoded array as bytes.
    ///
    /// See the documentation for [`bs58::decode`](crate::decode())
    /// for an explanation of the errors that may occur.
    ///
    /// # Examples
    ///
    /// ```rust
    /// const _: () = {
    ///     let Ok(output) = bs58::decode(b"EUYUqQf".as_slice()).into_array_const::<5>() else {
    ///         panic!()
    ///     };
    ///     assert!(matches!(&output, b"world"));
    /// };
    /// ```
    pub const fn into_array_const<const N: usize>(self) -> Result<[u8; N]> {
        assert!(
            matches!(self.check, Check::Disabled),
            "checksums in const aren't supported (why are you using this API at runtime)",
        );
        decode_into_const(self.input, self.alpha)
    }

    /// [`Self::into_array_const`] but the result will be unwrapped, turning any error into a panic
    /// message via [`Error::unwrap_const`], as a simple `into_array_const().unwrap()` isn't
    /// possible yet.
    ///
    /// # Examples
    ///
    /// ```rust
    /// const _: () = {
    ///     let output: [u8; 5] = bs58::decode(b"EUYUqQf".as_slice()).into_array_const_unwrap();
    ///     assert!(matches!(&output, b"world"));
    /// };
    /// ```
    ///
    /// ```rust
    /// const _: () = {
    ///     assert!(matches!(
    ///         bs58::decode(b"he11owor1d".as_slice())
    ///             .with_alphabet(bs58::Alphabet::RIPPLE)
    ///             .into_array_const_unwrap(),
    ///         [0x60, 0x65, 0xe7, 0x9b, 0xba, 0x2f, 0x78],
    ///     ));
    /// };
    /// ```
    pub const fn into_array_const_unwrap<const N: usize>(self) -> [u8; N] {
        match self.into_array_const() {
            Ok(result) => result,
            Err(err) => err.unwrap_const(),
        }
    }
}

fn decode_into(input: &[u8], output: &mut [u8], alpha: &Alphabet) -> Result<usize> {
    let mut index = 0;
    let zero = alpha.encode[0];

    for (i, c) in input.iter().enumerate() {
        if *c > 127 {
            return Err(Error::NonAsciiCharacter { index: i });
        }

        let mut val = alpha.decode[*c as usize] as usize;
        if val == 0xFF {
            return Err(Error::InvalidCharacter {
                character: *c as char,
                index: i,
            });
        }

        for byte in &mut output[..index] {
            val += (*byte as usize) * 58;
            *byte = (val & 0xFF) as u8;
            val >>= 8;
        }

        while val > 0 {
            let byte = output.get_mut(index).ok_or(Error::BufferTooSmall)?;
            *byte = (val & 0xFF) as u8;
            index += 1;
            val >>= 8
        }
    }

    for _ in input.iter().take_while(|c| **c == zero) {
        let byte = output.get_mut(index).ok_or(Error::BufferTooSmall)?;
        *byte = 0;
        index += 1;
    }

    output[..index].reverse();
    Ok(index)
}

#[cfg(feature = "check")]
fn decode_check_into(
    input: &[u8],
    output: &mut [u8],
    alpha: &Alphabet,
    expected_ver: Option<u8>,
) -> Result<usize> {
    use sha2::{Digest, Sha256};

    let decoded_len = decode_into(input, output, alpha)?;
    if decoded_len < CHECKSUM_LEN {
        return Err(Error::NoChecksum);
    }
    let checksum_index = decoded_len - CHECKSUM_LEN;

    let expected_checksum = &output[checksum_index..decoded_len];

    let first_hash = Sha256::digest(&output[0..checksum_index]);
    let second_hash = Sha256::digest(first_hash);
    let (checksum, _) = second_hash.split_at(CHECKSUM_LEN);

    if checksum == expected_checksum {
        if let Some(ver) = expected_ver {
            if output[0] == ver {
                Ok(checksum_index)
            } else {
                Err(Error::InvalidVersion {
                    ver: output[0],
                    expected_ver: ver,
                })
            }
        } else {
            Ok(checksum_index)
        }
    } else {
        let mut a: [u8; CHECKSUM_LEN] = Default::default();
        a.copy_from_slice(checksum);
        let mut b: [u8; CHECKSUM_LEN] = Default::default();
        b.copy_from_slice(expected_checksum);
        Err(Error::InvalidChecksum {
            checksum: a,
            expected_checksum: b,
        })
    }
}

#[cfg(feature = "cb58")]
fn decode_cb58_into(
    input: &[u8],
    output: &mut [u8],
    alpha: &Alphabet,
    expected_ver: Option<u8>,
) -> Result<usize> {
    use sha2::{Digest, Sha256};

    let decoded_len = decode_into(input, output, alpha)?;
    if decoded_len < CHECKSUM_LEN {
        return Err(Error::NoChecksum);
    }
    let checksum_index = decoded_len - CHECKSUM_LEN;

    let expected_checksum = &output[checksum_index..decoded_len];

    let hash = Sha256::digest(&output[0..checksum_index]);
    let (_, checksum) = hash.split_at(hash.len() - CHECKSUM_LEN);

    if checksum == expected_checksum {
        if let Some(ver) = expected_ver {
            if output[0] == ver {
                Ok(checksum_index)
            } else {
                Err(Error::InvalidVersion {
                    ver: output[0],
                    expected_ver: ver,
                })
            }
        } else {
            Ok(checksum_index)
        }
    } else {
        let mut a: [u8; CHECKSUM_LEN] = Default::default();
        a.copy_from_slice(checksum);
        let mut b: [u8; CHECKSUM_LEN] = Default::default();
        b.copy_from_slice(expected_checksum);
        Err(Error::InvalidChecksum {
            checksum: a,
            expected_checksum: b,
        })
    }
}

const fn decode_into_const<const N: usize>(input: &[u8], alpha: &Alphabet) -> Result<[u8; N]> {
    let mut output = [0u8; N];
    let mut index = 0;
    let zero = alpha.encode[0];

    let mut i = 0;
    while i < input.len() {
        let c = input[i];
        if c > 127 {
            return Err(Error::NonAsciiCharacter { index: i });
        }

        let mut val = alpha.decode[c as usize] as usize;
        if val == 0xFF {
            return Err(Error::InvalidCharacter {
                character: c as char,
                index: i,
            });
        }

        let mut j = 0;
        while j < index {
            let byte = output[j];
            val += (byte as usize) * 58;
            output[j] = (val & 0xFF) as u8;
            val >>= 8;
            j += 1;
        }

        while val > 0 {
            if index >= output.len() {
                return Err(Error::BufferTooSmall);
            }
            output[index] = (val & 0xFF) as u8;
            index += 1;
            val >>= 8
        }
        i += 1;
    }

    let mut i = 0;
    while i < input.len() && input[i] == zero {
        if index >= output.len() {
            return Err(Error::BufferTooSmall);
        }
        output[index] = 0;
        index += 1;
        i += 1;
    }

    // reverse
    let mut i = 0;
    let n = index / 2;
    while i < n {
        let x = output[i];
        output[i] = output[index - 1 - i];
        output[index - 1 - i] = x;
        i += 1;
    }

    Ok(output)
}

#[cfg(feature = "std")]
impl std::error::Error for Error {}

impl fmt::Display for Error {
    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
        match *self {
            Error::BufferTooSmall => write!(
                f,
                "buffer provided to decode base58 encoded string into was too small"
            ),
            Error::InvalidCharacter { character, index } => write!(
                f,
                "provided string contained invalid character {:?} at byte {}",
                character, index
            ),
            Error::NonAsciiCharacter { index } => write!(
                f,
                "provided string contained non-ascii character starting at byte {}",
                index
            ),
            #[cfg(any(feature = "check", feature = "cb58"))]
            Error::InvalidChecksum {
                checksum,
                expected_checksum,
            } => write!(
                f,
                "invalid checksum, calculated checksum: '{:?}', expected checksum: {:?}",
                checksum, expected_checksum
            ),
            #[cfg(any(feature = "check", feature = "cb58"))]
            Error::InvalidVersion { ver, expected_ver } => write!(
                f,
                "invalid version, payload version: '{:?}', expected version: {:?}",
                ver, expected_ver
            ),
            #[cfg(any(feature = "check", feature = "cb58"))]
            Error::NoChecksum => write!(f, "provided string is too small to contain a checksum"),
        }
    }
}

impl Error {
    /// Panic with an error message based on this error. This cannot include any of the dynamic
    /// content because formatting in `const` is not yet possible.
    pub const fn unwrap_const(self) -> ! {
        match self {
            Error::BufferTooSmall => {
                panic!("buffer provided to decode base58 encoded string into was too small")
            }
            Error::InvalidCharacter { .. } => panic!("provided string contained invalid character"),
            Error::NonAsciiCharacter { .. } => {
                panic!("provided string contained non-ascii character")
            }
            #[cfg(any(feature = "check", feature = "cb58"))]
            Error::InvalidChecksum { .. } => panic!("invalid checksum"),
            #[cfg(any(feature = "check", feature = "cb58"))]
            Error::InvalidVersion { .. } => panic!("invalid version"),
            #[cfg(any(feature = "check", feature = "cb58"))]
            Error::NoChecksum => panic!("provided string is too small to contain a checksum"),
        }
    }
}