libribzip2 0.5.0

a bzip2 library written in pure rust
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

use self::kmeansclustering::KMeansProblem;

use super::zle::ZleSymbol;
mod kmeansclustering;

#[derive(Clone, Copy)]
#[non_exhaustive]
/// Strategy for using Huffman Tables
///
/// * [EncodingStrategy::Single] - every block of 900k uses a single huffman code table
/// * [EncodingStrategy::BlockWise] - usage of code tables in 50 byte chunks is optimized using Lloyd's algorithm with given parameters
pub enum EncodingStrategy {
    BlockWise {
        num_clusters: usize,
        num_iterations: usize,
    },
    Single,
}

pub(crate) struct SinglePropabilityMap {
    frequencies: Vec<usize>,
    symbol_count: usize,
}

impl SinglePropabilityMap {
    pub(crate) fn create(size: usize) -> Self {
        SinglePropabilityMap {
            frequencies: vec![0; size + 1],
            symbol_count: 0,
        }
    }
}

impl SymbolReporter for SinglePropabilityMap {
    fn report_symbol(&mut self, symbol: &ZleSymbol) {
        self.symbol_count += 1;
        match symbol {
            ZleSymbol::RunA => {
                self.frequencies[0] += 1;
            }
            ZleSymbol::RunB => {
                self.frequencies[1] += 1;
            }
            ZleSymbol::Number(i) => {
                self.frequencies[*i as usize + 1] += 1;
            }
        }
    }

    fn finalize(&mut self) -> ReportedSymbols {
        let table = IntoFrequencyTable {
            frequencies: self.frequencies.clone(),
        };
        ReportedSymbols {
            reported_frequencies: vec![table.clone(), table],
            selectors: vec![0; (self.symbol_count as f32 / 50.0).ceil() as usize],
        }
    }
}

pub(crate) struct BlockWisePropabilityMap {
    current_frequencies: Vec<u8>,
    pub(crate) maps: Vec<Vec<u8>>,
    pub(crate) size: usize,
    counter: usize,
    num_iterations: usize,
    num_clusters: usize,
}

impl BlockWisePropabilityMap {
    pub(crate) fn create(size: usize, num_clusters: usize, num_iterations: usize) -> Self {
        Self {
            current_frequencies: vec![0; size + 1],
            maps: vec![],
            counter: 0,
            size,
            num_clusters,
            num_iterations,
        }
    }
}

impl SymbolReporter for BlockWisePropabilityMap {
    fn report_symbol(&mut self, symbol: &ZleSymbol) {
        match symbol {
            ZleSymbol::RunA => {
                self.current_frequencies[0] += 1;
            }
            ZleSymbol::RunB => {
                self.current_frequencies[1] += 1;
            }
            ZleSymbol::Number(i) => {
                self.current_frequencies[*i as usize + 1] += 1;
            }
        }
        self.counter += 1;
        if self.counter >= 50 {
            self.maps.push(std::mem::replace(
                &mut self.current_frequencies,
                vec![0; self.size + 1],
            ));
            self.counter = 0;
        }
    }

    fn finalize(&mut self) -> ReportedSymbols {
        self.maps.push(std::mem::replace(
            &mut self.current_frequencies,
            vec![0; self.size + 1],
        ));

        let p = KMeansProblem {
            dimension: self.size + 1,
            data: &self.maps,
            num_iterations: self.num_iterations,
            num_clusters: self.num_clusters,
        };
        let tables = p.solve();

        ReportedSymbols {
            reported_frequencies: tables
                .means
                .iter()
                .map(|x| IntoFrequencyTable {
                    frequencies: x.iter().map(|x| *x as usize).collect::<Vec<_>>(),
                })
                .collect::<Vec<_>>(),
            selectors: tables.assignments,
        }
    }
}

pub(crate) trait SymbolReporter {
    fn report_symbol(&mut self, symbol: &ZleSymbol);
    fn finalize(&mut self) -> ReportedSymbols;
}

pub(crate) struct ReportedSymbols {
    pub(crate) reported_frequencies: Vec<IntoFrequencyTable>,
    pub(crate) selectors: Vec<u8>,
}

#[derive(Clone)]
pub(crate) struct IntoFrequencyTable {
    pub(crate) frequencies: Vec<usize>,
}

impl IntoFrequencyTable {
    pub(crate) fn iterate(self) -> impl Iterator<Item = (ZleSymbol, usize)> {
        self.frequencies
            .into_iter()
            .enumerate()
            .map(|(symbol, frequency)| match symbol {
                0 => (ZleSymbol::RunA, frequency),
                1 => (ZleSymbol::RunB, frequency),
                x => (ZleSymbol::Number((x - 1) as u8), frequency),
            })
    }
}