arcode 0.2.4

An Arithmetic Coder (lossless entropy encoder)
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

//! This module contains the main code for the encoder. It also
//! contains an simple implementation of a binary encoder.

use std::io::{Error, Write};

use bitbit::BitWriter;

use crate::{Model, Range};

pub struct ArithmeticEncoder {
    _precision: u64,
    pending_bit_count: u32,
    range: Range,
}

impl ArithmeticEncoder {
    /// # Arguments
    /// `precision` is the [bit precision](https://en.wikipedia.org/wiki/Arithmetic_coding#Precision_and_renormalization)
    /// that the encoder should use. If the
    /// precision is too low than symbols will not be able to be differentiated.
    pub fn new(precision: u64) -> Self {
        Self {
            _precision: precision,
            pending_bit_count: 0,
            range: Range::new(precision),
        }
    }

    pub fn encode<T: Write>(
        &mut self,
        symbol: u32,
        source_model: &Model,
        output: &mut BitWriter<T>,
    ) -> Result<(), Error> {
        let low_high = self.range.calculate_range(symbol, source_model);
        self.range.update_range(low_high);

        while self.range.in_bottom_half() || self.range.in_upper_half() {
            if self.range.in_bottom_half() {
                self.range.scale_bottom_half();
                self.emit(false, output)?;
            } else if self.range.in_upper_half() {
                self.range.scale_upper_half();
                self.emit(true, output)?;
            }
        }

        while self.range.in_middle_half() {
            self.pending_bit_count += 1;
            self.range.scale_middle_half();
        }

        Ok(())
    }

    fn emit<T: Write>(&mut self, bit: bool, output: &mut BitWriter<T>) -> Result<(), Error> {
        output.write_bit(bit)?;

        while self.pending_bit_count > 0 {
            output.write_bit(!bit)?;
            self.pending_bit_count -= 1;
        }

        Ok(())
    }

    pub fn finish_encode<T: Write>(&mut self, output: &mut BitWriter<T>) -> Result<(), Error> {
        self.pending_bit_count += 1;

        if self.range.in_bottom_quarter() {
            self.emit(false, output)?;
        } else {
            self.emit(true, output)?;
        }

        Ok(())
    }
}

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

    use bitbit::BitWriter;

    use super::ArithmeticEncoder;
    use crate::{EOFKind, Model};

    #[test]
    fn e2e() {
        let mut encoder = ArithmeticEncoder::new(30);
        let mut source_model = Model::builder().num_symbols(10).eof(EOFKind::End).build();
        let mut output = Cursor::new(vec![]);
        let mut out_writer = BitWriter::new(&mut output);
        let to_encode: [u32; 5] = [7, 2, 2, 2, 7];
        for x in &to_encode {
            encoder.encode(*x, &source_model, &mut out_writer).unwrap();
            source_model.update_symbol(*x);
        }
        encoder
            .encode(source_model.eof(), &source_model, &mut out_writer)
            .unwrap();
        encoder.finish_encode(&mut out_writer).unwrap();
        out_writer.pad_to_byte().unwrap();
        assert_eq!(output.get_ref(), &[184, 96, 208]);
    }
}