q_compress 0.3.0

Good data compression for numerical sequences
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

pub use bit_reader::BitReader;
pub use compressor::Compressor;
pub use decompressor::Decompressor;
pub use types::boolean::BoolCompressor;
pub use types::boolean::BoolDecompressor;
pub use types::float32::F32Compressor;
pub use types::float32::F32Decompressor;
pub use types::float64::F64Compressor;
pub use types::float64::F64Decompressor;
pub use types::signed32::I32Compressor;
pub use types::signed32::I32Decompressor;
pub use types::signed64::I64Compressor;
pub use types::signed64::I64Decompressor;
pub use types::timestamp_ns::TimestampNs;
pub use types::timestamp_ns::TimestampNsCompressor;
pub use types::timestamp_ns::TimestampNsDecompressor;
pub use types::unsigned32::U32Compressor;
pub use types::unsigned32::U32Decompressor;
pub use types::unsigned64::U64Compressor;
pub use types::unsigned64::U64Decompressor;

pub use constants::MAX_ENTRIES;

mod bits;
mod constants;
mod huffman;
mod prefix;
mod utils;
pub mod bit_reader;
pub mod compressor;
pub mod decompressor;
pub mod errors;
pub mod types;

#[cfg(test)]
mod tests {
  use crate::types::NumberLike;
  use crate::{Compressor, BitReader, Decompressor, TimestampNs};

  #[test]
  fn test_edge_cases() {
    assert_recovers(vec![true, true, false, true], 0);
    assert_recovers(vec![false, false, false], 0);
    assert_recovers(vec![false], 0);
    assert_recovers(vec![u64::MIN, u64::MAX], 0);
    assert_recovers(vec![f64::MIN, f64::MAX], 0);
    assert_recovers(vec![1.2_f32], 0);
    assert_recovers(vec![1.2_f32], 1);
    assert_recovers(vec![1.2_f32], 2);
    assert_recovers(Vec::<u32>::new(), 6);
    assert_recovers(Vec::<u32>::new(), 0);
  }

  #[test]
  fn test_moderate_data() {
    let mut v = Vec::new();
    for i in -50000..50000 {
      v.push(i);
    }
    assert_recovers(v, 5);
  }

  #[test]
  fn test_boolean_codec() {
    assert_recovers(vec![true, true, false, true, false], 1);
  }

  #[test]
  fn test_sparse() {
    let mut v = Vec::new();
    for _ in 0..10000 {
      v.push(true);
    }
    v.push(false);
    v.push(false);
    v.push(true);
    assert_recovers(v, 1);
  }

  #[test]
  fn test_u32_codec() {
    assert_recovers(vec![0_u32, u32::MAX, 3, 4, 5], 1);
  }

  #[test]
  fn test_u64_codec() {
    assert_recovers(vec![0_u64, u64::MAX, 3, 4, 5], 1);
  }

  #[test]
  fn test_i32_codec() {
    assert_recovers(vec![0_i32, -1, i32::MAX, i32::MIN, 7], 1);
  }

  #[test]
  fn test_i64_codec() {
    assert_recovers(vec![0_i64, -1, i64::MAX, i64::MIN, 7], 1);
  }

  #[test]
  fn test_f32_codec() {
    assert_recovers(vec![f32::MAX, f32::MIN, f32::NAN, f32::NEG_INFINITY, f32::INFINITY, 0.0, 77.7], 1);
  }

  #[test]
  fn test_f64_codec() {
    assert_recovers(vec![f64::MAX, f64::MIN, f64::NAN, f64::NEG_INFINITY, f64::INFINITY, 0.0, 77.7], 1);
  }

  #[test]
  fn test_timestamp_ns_codec() {
    assert_recovers(
      vec![
        TimestampNs::from_secs_and_nanos(i64::MIN, 0),
        TimestampNs::from_secs_and_nanos(i64::MAX, 999_999_999),
        TimestampNs::from_secs_and_nanos(i64::MIN, 999_999_999),
        TimestampNs::from_secs_and_nanos(0, 123_456_789),
        TimestampNs::from_secs_and_nanos(-1, 123_456_789),
      ],
      1
    );
  }

  fn assert_recovers<T: NumberLike>(vals: Vec<T>, max_depth: u32) {
    let compressor = Compressor::train(vals.clone(), max_depth)
      .expect("training error");
    let compressed = compressor.compress(&vals)
      .expect("compression error");
    let mut bit_reader = BitReader::from(compressed);
    let decompressor = Decompressor::<T>::from_reader(&mut bit_reader)
      .expect("header error");
    let decompressed = decompressor.decompress(&mut bit_reader);
    // can't do assert_eq on the whole vector because floating points don't compare exactly
    assert_eq!(decompressed.len(), vals.len());
    for i in 0..decompressed.len() {
      assert!(decompressed[i].num_eq(&vals[i]));
    }
  }
}