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
//! A minimal implementation of Adler32 for Rust.
//!
//! This provides the simple method adler32(), that exhausts a Read and
//! computes the Adler32 hash, as well as the RollingAdler32 struct, that can
//! build a hash byte-by-byte, allowing to 'forget' past bytes in a rolling
//! fashion.
//!
//! The adler32 code has been translated (as accurately as I could manage) from
//! the zlib implementation.

#[cfg(test)]
extern crate rand;

use std::io;

// adler32 algorithm and implementation taken from zlib; http://www.zlib.net/
// It was translated into Rust as accurately as I could manage
// The (slow) reference was taken from Wikipedia; https://en.wikipedia.org/

/* zlib.h -- interface of the 'zlib' general purpose compression library
  version 1.2.8, April 28th, 2013

  Copyright (C) 1995-2013 Jean-loup Gailly and Mark Adler

  This software is provided 'as-is', without any express or implied
  warranty.  In no event will the authors be held liable for any damages
  arising from the use of this software.

  Permission is granted to anyone to use this software for any purpose,
  including commercial applications, and to alter it and redistribute it
  freely, subject to the following restrictions:

  1. The origin of this software must not be misrepresented; you must not
     claim that you wrote the original software. If you use this software
     in a product, an acknowledgment in the product documentation would be
     appreciated but is not required.
  2. Altered source versions must be plainly marked as such, and must not be
     misrepresented as being the original software.
  3. This notice may not be removed or altered from any source distribution.

  Jean-loup Gailly        Mark Adler
  jloup@gzip.org          madler@alumni.caltech.edu

*/

// largest prime smaller than 65536
const BASE: u32 = 65521;

// NMAX is the largest n such that 255n(n+1)/2 + (n+1)(BASE-1) <= 2^32-1
const NMAX: usize = 5552;

#[inline(always)]
fn do1(adler: &mut u32, sum2: &mut u32, buf: &[u8]) {
    *adler += buf[0] as u32;
    *sum2 += *adler;
}

#[inline(always)]
fn do2(adler: &mut u32, sum2: &mut u32, buf: &[u8]) {
    do1(adler, sum2, &buf[0..1]);
    do1(adler, sum2, &buf[1..2]);
}

#[inline(always)]
fn do4(adler: &mut u32, sum2: &mut u32, buf: &[u8]) {
    do2(adler, sum2, &buf[0..2]);
    do2(adler, sum2, &buf[2..4]);
}

#[inline(always)]
fn do8(adler: &mut u32, sum2: &mut u32, buf: &[u8]) {
    do4(adler, sum2, &buf[0..4]);
    do4(adler, sum2, &buf[4..8]);
}

#[inline(always)]
fn do16(adler: &mut u32, sum2: &mut u32, buf: &[u8]) {
    do8(adler, sum2, &buf[0..8]);
    do8(adler, sum2, &buf[8..16]);
}

/// A rolling version of the Adler32 hash, which can 'forget' past bytes.
///
/// Calling remove() will update the hash to the value it would have if that
/// past byte had never been fed to the algorithm. This allows you to get the
/// hash of a rolling window very efficiently.
pub struct RollingAdler32 {
    a: u32,
    b: u32,
}

impl RollingAdler32 {
    /// Creates an empty Adler32 context (with hash 1).
    pub fn new() -> RollingAdler32 {
        Self::from_value(1)
    }

    /// Creates an Adler32 context with the given initial value.
    pub fn from_value(adler32: u32) -> RollingAdler32 {
        let a = adler32 & 0xFFFF;
        let b = adler32 >> 16;
        RollingAdler32 { a: a, b: b }
    }

    /// Convenience function initializing a context from the hash of a buffer.
    pub fn from_buffer(buffer: &[u8]) -> RollingAdler32 {
        let mut hash = RollingAdler32::new();
        hash.update_buffer(buffer);
        hash
    }

    /// Returns the current hash.
    pub fn hash(&self) -> u32 {
        (self.b << 16) | self.a
    }

    /// Removes the given `byte` that was fed to the algorithm `size` bytes ago.
    pub fn remove(&mut self, size: usize, byte: u8) {
        let byte = byte as u32;
        self.a = (self.a + BASE - byte) % BASE;
        self.b = ((self.b + BASE - 1)
                      .wrapping_add(BASE.wrapping_sub(size as u32)
                                        .wrapping_mul(byte))) % BASE;
    }

    /// Feeds a new `byte` to the algorithm to update the hash.
    pub fn update(&mut self, byte: u8) {
        let byte = byte as u32;
        self.a = (self.a + byte) % BASE;
        self.b = (self.b + self.a) % BASE;
    }

    /// Feeds a vector of bytes to the algorithm to update the hash.
    pub fn update_buffer(&mut self, buffer: &[u8]) {
        let len = buffer.len();

        // in case user likes doing a byte at a time, keep it fast
        if len == 1 {
            self.update(buffer[0]);
            return;
        }

        // in case short lengths are provided, keep it somewhat fast
        if len < 16 {
            for pos in 0..len {
                self.a += buffer[pos] as u32;
                self.b += self.a;
            }
            if self.a >= BASE {
                self.a -= BASE;
            }
            self.b %= BASE;
            return;
        }

        let mut pos = 0;

        // do length NMAX blocks -- requires just one modulo operation;
        while pos + NMAX <= len {
            let end = pos + NMAX;
            while pos < end {
                // 16 sums unrolled
                do16(&mut self.a, &mut self.b, &buffer[pos..pos + 16]);
                pos += 16;
            }
            self.a %= BASE;
            self.b %= BASE;
        }

        // do remaining bytes (less than NMAX, still just one modulo)
        if pos < len { // avoid modulos if none remaining
            while len - pos >= 16 {
                do16(&mut self.a, &mut self.b, &buffer[pos..pos + 16]);
                pos += 16;
            }
            while len - pos > 0 {
                self.a += buffer[pos] as u32;
                self.b += self.a;
                pos += 1;
            }
            self.a %= BASE;
            self.b %= BASE;
        }
    }
}

/// Consume a Read object and returns the Adler32 hash.
pub fn adler32<R: io::Read>(mut reader: R) -> io::Result<u32> {
    let mut hash = RollingAdler32::new();
    let mut buffer = [0u8; NMAX];
    let mut read = try!(reader.read(&mut buffer));
    while read > 0 {
        hash.update_buffer(&buffer[..read]);
        read = try!(reader.read(&mut buffer));
    }
    Ok(hash.hash())
}

#[cfg(test)]
mod test {
    use rand;
    use rand::Rng;
    use std::io;

    use super::{BASE, adler32, RollingAdler32};

    fn adler32_slow<R: io::Read>(reader: R) -> io::Result<u32> {
        let mut a: u32 = 1;
        let mut b: u32 = 0;

        for byte in reader.bytes() {
            let byte = try!(byte) as u32;
            a = (a + byte) % BASE;
            b = (b + a) % BASE;
        }

        Ok((b << 16) | a)
    }

    #[test]
    fn testvectors() {
        fn do_test(v: u32, bytes: &[u8]) {
            let mut hash = RollingAdler32::new();
            hash.update_buffer(&bytes);
            assert_eq!(hash.hash(), v);

            let r = io::Cursor::new(bytes);
            assert_eq!(adler32(r).unwrap(), v);
        }
        do_test(0x00000001, b"");
        do_test(0x00620062, b"a");
        do_test(0x024d0127, b"abc");
        do_test(0x29750586, b"message digest");
        do_test(0x90860b20, b"abcdefghijklmnopqrstuvwxyz");
        do_test(0x8adb150c, b"ABCDEFGHIJKLMNOPQRSTUVWXYZ\
                              abcdefghijklmnopqrstuvwxyz\
                              0123456789");
        do_test(0x97b61069, b"1234567890123456789012345678901234567890\
                              1234567890123456789012345678901234567890");
        do_test(0xD6251498, &[255; 64000]);
    }

    #[test]
    fn compare() {
        let mut rng = rand::thread_rng();
        let mut data = vec![0u8; 5589];
        for size in [0, 1, 3, 4, 5, 31, 32, 33, 67,
                     5550, 5552, 5553, 5568, 5584, 5589].iter().cloned() {
            rng.fill_bytes(&mut data[..size]);
            let r1 = io::Cursor::new(&data[..size]);
            let r2 = r1.clone();
            if adler32_slow(r1).unwrap() != adler32(r2).unwrap() {
                panic!("Comparison failed, size={}", size);
            }
        }
    }

    #[test]
    fn rolling() {
        assert_eq!(RollingAdler32::from_value(0x01020304).hash(), 0x01020304);

        fn do_test(a: &[u8], b: &[u8]) {
            let mut total = Vec::with_capacity(a.len() + b.len());
            total.extend(a);
            total.extend(b);
            let mut h = RollingAdler32::from_buffer(&total[..(b.len())]);
            for i in 0..(a.len()) {
                h.remove(b.len(), a[i]);
                h.update(total[b.len() + i]);
            }
            assert_eq!(h.hash(), adler32(b).unwrap());
        }
        do_test(b"a", b"b");
        do_test(b"", b"this a test");
        do_test(b"th", b"is a test");
        do_test(b"this a ", b"test");
    }

    #[test]
    fn long_window_remove() {
        let mut hash = RollingAdler32::new();
        let w = 65536;
        assert!(w as u32 > BASE);

        let mut bytes = vec![0; w*3];
        for (i, b) in bytes.iter_mut().enumerate() {
            *b = i as u8;
        }

        for (i, b) in bytes.iter().enumerate() {
            if i >= w {
                hash.remove(w, bytes[i - w]);
            }
            hash.update(*b);
            if i > 0 && i % w == 0 {
                assert_eq!(hash.hash(), 0x433a8772);
            }
        }
        assert_eq!(hash.hash(), 0xbbba8772);
    }
}