zstd-complete 0.3.0

A pure Rust Zstandard compressor and decompressor with bounded streaming I/O
Documentation 
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

//! Match-length counter shared by the C-port match finders.

#![forbid(unsafe_code)]

use super::unaligned::{read16, read32, read64};

pub(super) fn count_match(
    src: &[u8],
    mut pos: usize,
    mut match_pos: usize,
    match_limit: usize,
) -> usize {
    let start = pos;

    while pos + 8 <= match_limit && match_pos + 8 <= src.len() {
        let diff = read64(src, pos) ^ read64(src, match_pos);
        if diff != 0 {
            return pos - start + common_prefix_bytes(diff);
        }
        pos += 8;
        match_pos += 8;
    }

    if pos + 4 <= match_limit
        && match_pos + 4 <= src.len()
        && read32(src, pos) == read32(src, match_pos)
    {
        pos += 4;
        match_pos += 4;
    }

    if pos + 2 <= match_limit
        && match_pos + 2 <= src.len()
        && read16(src, pos) == read16(src, match_pos)
    {
        pos += 2;
        match_pos += 2;
    }

    if pos < match_limit && match_pos < src.len() && src[pos] == src[match_pos] {
        pos += 1;
    }

    pos - start
}

#[inline(always)]
pub(super) fn count_match_no_dict(
    src: &[u8],
    mut pos: usize,
    mut match_pos: usize,
    match_limit: usize,
) -> usize {
    debug_assert!(match_pos <= pos);
    debug_assert!(match_limit <= src.len());

    let start = pos;
    let loop_limit = match_limit.saturating_sub(7);

    if pos < loop_limit {
        let diff = read64(src, pos) ^ read64(src, match_pos);
        if diff != 0 {
            return common_prefix_bytes(diff);
        }
        pos += 8;
        match_pos += 8;

        while pos < loop_limit {
            let diff = read64(src, pos) ^ read64(src, match_pos);
            if diff != 0 {
                return pos - start + common_prefix_bytes(diff);
            }
            pos += 8;
            match_pos += 8;
        }
    }

    if pos + 4 <= match_limit && read32(src, pos) == read32(src, match_pos) {
        pos += 4;
        match_pos += 4;
    }

    if pos + 2 <= match_limit && read16(src, pos) == read16(src, match_pos) {
        pos += 2;
        match_pos += 2;
    }

    if pos < match_limit && src[pos] == src[match_pos] {
        pos += 1;
    }

    pos - start
}

#[inline(always)]
pub(super) fn count_match_behind(
    src: &[u8],
    mut pos: usize,
    mut match_pos: usize,
    match_limit: usize,
) -> usize {
    debug_assert!(match_pos <= pos);
    debug_assert!(match_limit <= src.len());

    let start = pos;

    while pos + 8 <= match_limit {
        let diff = read64(src, pos) ^ read64(src, match_pos);
        if diff != 0 {
            return pos - start + common_prefix_bytes(diff);
        }
        pos += 8;
        match_pos += 8;
    }

    if pos + 4 <= match_limit && read32(src, pos) == read32(src, match_pos) {
        pos += 4;
        match_pos += 4;
    }

    if pos + 2 <= match_limit && read16(src, pos) == read16(src, match_pos) {
        pos += 2;
        match_pos += 2;
    }

    if pos < match_limit && src[pos] == src[match_pos] {
        pos += 1;
    }

    pos - start
}

#[inline(always)]
fn common_prefix_bytes(diff: u64) -> usize {
    (diff.trailing_zeros() >> 3) as usize
}

#[cfg(test)]
mod tests {
    use super::{count_match, count_match_behind, count_match_no_dict};

    #[test]
    fn counts_full_match_to_limit() {
        let data = b"abcdefghabcdefgh_tail";
        assert_eq!(count_match(data, 0, 8, 8), 8);
    }

    #[test]
    fn stops_inside_first_word() {
        let data = b"abcxefghabcdzzzz";
        assert_eq!(count_match(data, 0, 8, 8), 3);
    }

    #[test]
    fn no_dict_counter_matches_general_counter() {
        let data = b"0123456789abcdef0123456789abcx";
        assert_eq!(
            count_match_no_dict(data, 16, 0, data.len()),
            count_match(data, 16, 0, data.len())
        );
    }

    #[test]
    fn counts_tail_after_word_chunks() {
        let data = b"abcdefghijkXabcdefghijkY";
        assert_eq!(count_match(data, 0, 12, 12), 11);
    }

    #[test]
    fn respects_match_limit() {
        let data = b"abcdefghijklmnop";
        assert_eq!(count_match(data, 0, 0, 5), 5);
    }

    #[test]
    fn behind_counter_counts_full_match_to_limit() {
        let data = b"abcdefghabcdefgh_tail";
        assert_eq!(count_match_behind(data, 8, 0, 16), 8);
    }
}