bearing 0.1.0-alpha.5

A Rust port of Apache Lucene
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

// SPDX-License-Identifier: Apache-2.0

//! Range encoding for multi-dimensional numeric ranges.
//!
//! Encodes min/max pairs for 1-4 dimensions as sortable byte arrays.
//! Used by `IntRange`, `LongRange`, `FloatRange`, and `DoubleRange` fields.

use std::fmt;

use crate::encoding::sortable_bytes::{from_double, from_float, from_int, from_long};

/// Encodes a range as sortable bytes: `[min1..minN, max1..maxN]`.
///
/// Each min/max value is converted to sortable bytes via `encode_fn`, then
/// packed into a single byte array with mins first, then maxs.
fn encode_range<T: PartialOrd + fmt::Display + Copy>(
    mins: &[T],
    maxs: &[T],
    elem_size: usize,
    encode_fn: fn(T) -> Vec<u8>,
) -> Vec<u8> {
    validate_range_dims(mins.len(), maxs.len());
    for i in 0..mins.len() {
        assert!(
            mins[i] <= maxs[i],
            "min[{i}] ({}) > max[{i}] ({})",
            mins[i],
            maxs[i]
        );
    }
    let dims = mins.len();
    let mut bytes = vec![0u8; dims * 2 * elem_size];
    for (i, &min) in mins.iter().enumerate() {
        bytes[i * elem_size..(i + 1) * elem_size].copy_from_slice(&encode_fn(min));
    }
    for (i, &max) in maxs.iter().enumerate() {
        let offset = (dims + i) * elem_size;
        bytes[offset..offset + elem_size].copy_from_slice(&encode_fn(max));
    }
    bytes
}

/// Encodes an integer range as sortable bytes: `[min1..minN, max1..maxN]`.
///
/// Each dimension's min and max are encoded as 4-byte sortable values.
/// Panics if `mins` and `maxs` differ in length, exceed 4 dimensions,
/// or any `min[i] > max[i]`.
pub fn encode_int(mins: &[i32], maxs: &[i32]) -> Vec<u8> {
    encode_range(mins, maxs, 4, |v| from_int(v).to_vec())
}

/// Encodes a long range as sortable bytes: `[min1..minN, max1..maxN]`.
///
/// Each dimension's min and max are encoded as 8-byte sortable values.
pub fn encode_long(mins: &[i64], maxs: &[i64]) -> Vec<u8> {
    encode_range(mins, maxs, 8, |v| from_long(v).to_vec())
}

/// Encodes a float range as sortable bytes: `[min1..minN, max1..maxN]`.
///
/// Each dimension's min and max are converted to sortable ints, then encoded
/// as 4-byte sortable values.
pub fn encode_float(mins: &[f32], maxs: &[f32]) -> Vec<u8> {
    encode_range(mins, maxs, 4, |v| from_float(v).to_vec())
}

/// Encodes a double range as sortable bytes: `[min1..minN, max1..maxN]`.
///
/// Each dimension's min and max are converted to sortable longs, then encoded
/// as 8-byte sortable values.
pub fn encode_double(mins: &[f64], maxs: &[f64]) -> Vec<u8> {
    encode_range(mins, maxs, 8, |v| from_double(v).to_vec())
}

fn validate_range_dims(min_len: usize, max_len: usize) {
    assert_eq!(min_len, max_len, "mins and maxs must have equal length");
    assert!(
        (1..=4).contains(&min_len),
        "dimensions must be 1-4, got {min_len}"
    );
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn test_encode_int_single_dim() {
        let bytes = encode_int(&[10], &[20]);
        assert_len_eq_x!(&bytes, 8); // 1 dim * 2 * 4 bytes
        assert_eq!(&bytes[0..4], &from_int(10));
        assert_eq!(&bytes[4..8], &from_int(20));
    }

    #[test]
    fn test_encode_int_multi_dim() {
        let bytes = encode_int(&[1, 2], &[10, 20]);
        assert_len_eq_x!(&bytes, 16); // 2 dims * 2 * 4 bytes
        assert_eq!(&bytes[0..4], &from_int(1));
        assert_eq!(&bytes[4..8], &from_int(2));
        assert_eq!(&bytes[8..12], &from_int(10));
        assert_eq!(&bytes[12..16], &from_int(20));
    }

    #[test]
    fn test_encode_long_single_dim() {
        let bytes = encode_long(&[100], &[200]);
        assert_len_eq_x!(&bytes, 16); // 1 dim * 2 * 8 bytes
        assert_eq!(&bytes[0..8], &from_long(100));
        assert_eq!(&bytes[8..16], &from_long(200));
    }

    #[test]
    fn test_encode_float_single_dim() {
        let bytes = encode_float(&[1.0], &[2.0]);
        assert_len_eq_x!(&bytes, 8);
        assert_eq!(&bytes[0..4], &from_float(1.0));
        assert_eq!(&bytes[4..8], &from_float(2.0));
    }

    #[test]
    fn test_encode_double_single_dim() {
        let bytes = encode_double(&[1.0], &[2.0]);
        assert_len_eq_x!(&bytes, 16);
        assert_eq!(&bytes[0..8], &from_double(1.0));
        assert_eq!(&bytes[8..16], &from_double(2.0));
    }

    #[test]
    #[should_panic(expected = "equal length")]
    fn test_encode_int_mismatched_dims() {
        encode_int(&[1, 2], &[10]);
    }

    #[test]
    #[should_panic(expected = "dimensions must be 1-4")]
    fn test_encode_int_zero_dims() {
        encode_int(&[], &[]);
    }

    #[test]
    #[should_panic(expected = "dimensions must be 1-4")]
    fn test_encode_int_five_dims() {
        encode_int(&[1, 2, 3, 4, 5], &[10, 20, 30, 40, 50]);
    }

    #[test]
    #[should_panic(expected = "min[0]")]
    fn test_encode_int_min_gt_max() {
        encode_int(&[20], &[10]);
    }

    #[test]
    fn test_encode_int_equal_min_max() {
        let bytes = encode_int(&[5], &[5]);
        assert_eq!(&bytes[0..4], &bytes[4..8]);
    }

    #[test]
    fn test_encode_long_multi_dim() {
        let bytes = encode_long(&[1, 2, 3], &[10, 20, 30]);
        assert_len_eq_x!(&bytes, 48); // 3 dims * 2 * 8 bytes
    }

    #[test]
    fn test_encode_float_multi_dim() {
        let bytes = encode_float(&[0.5, 1.5], &[2.5, 3.5]);
        assert_len_eq_x!(&bytes, 16); // 2 dims * 2 * 4 bytes
    }

    #[test]
    fn test_encode_double_four_dims() {
        let bytes = encode_double(&[1.0, 2.0, 3.0, 4.0], &[5.0, 6.0, 7.0, 8.0]);
        assert_len_eq_x!(&bytes, 64); // 4 dims * 2 * 8 bytes
    }
}