velesdb-core 1.9.2

High-performance vector database engine written in 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
160
161
162

//! Tests for full-text (BM25) and hybrid (vector + text) search methods.

#![cfg(all(test, feature = "persistence"))]

use crate::collection::types::Collection;
use crate::filter::{Condition, Filter};
use crate::test_fixtures::fixtures::{make_point_with_payload, setup_collection};

/// Helper: create a collection with points that have text-rich payloads.
///
/// The BM25 index is populated automatically during `upsert` when payloads
/// contain string fields (Collection::upsert indexes all string values).
fn setup_text_collection() -> (tempfile::TempDir, Collection) {
    let points = vec![
        make_point_with_payload(
            1,
            vec![1.0, 0.0, 0.0, 0.0],
            serde_json::json!({
                "title": "rust programming language",
                "category": "tech"
            }),
        ),
        make_point_with_payload(
            2,
            vec![0.0, 1.0, 0.0, 0.0],
            serde_json::json!({
                "title": "python programming tutorial",
                "category": "tech"
            }),
        ),
        make_point_with_payload(
            3,
            vec![0.0, 0.0, 1.0, 0.0],
            serde_json::json!({
                "title": "football world cup results",
                "category": "sports"
            }),
        ),
        make_point_with_payload(
            4,
            vec![0.5, 0.5, 0.0, 0.0],
            serde_json::json!({
                "title": "rust web framework comparison",
                "category": "tech"
            }),
        ),
    ];

    let (dir, col) = setup_collection(4);
    col.upsert(points).expect("test: upsert");
    (dir, col)
}

// -----------------------------------------------------------------------
// Basic BM25 text search
// -----------------------------------------------------------------------

#[test]
fn test_text_search_finds_matching_documents() {
    let (_dir, col) = setup_text_collection();

    let results = col.text_search("rust", 10).expect("text search");

    // Points 1 and 4 both contain "rust" in their title.
    let ids: Vec<u64> = results.iter().map(|r| r.point.id).collect();
    assert!(ids.contains(&1), "should find point 1 (rust programming)");
    assert!(ids.contains(&4), "should find point 4 (rust web framework)");
    assert!(!ids.contains(&3), "should not find point 3 (football)");
}

#[test]
fn test_text_search_no_match_returns_empty() {
    let (_dir, col) = setup_text_collection();

    let results = col
        .text_search("quantum_entanglement_xyz", 10)
        .expect("text search");
    assert!(results.is_empty(), "nonsense query should return no hits");
}

#[test]
fn test_text_search_respects_k_limit() {
    let (_dir, col) = setup_text_collection();

    let results = col.text_search("programming", 1).expect("text search");
    assert!(results.len() <= 1, "should respect k=1 limit");
}

// -----------------------------------------------------------------------
// BM25 text search with metadata filter
// -----------------------------------------------------------------------

#[test]
fn test_text_search_with_filter_narrows_results() {
    let (_dir, col) = setup_text_collection();

    let filter = Filter::new(Condition::eq("category", "sports"));
    let results = col
        .text_search_with_filter("programming", 10, &filter)
        .expect("filtered text search");

    // "programming" appears in tech docs only; sports filter should exclude them.
    assert!(
        results.is_empty(),
        "no sports docs contain 'programming', got {} results",
        results.len()
    );
}

#[test]
fn test_text_search_with_filter_includes_matching_category() {
    let (_dir, col) = setup_text_collection();

    let filter = Filter::new(Condition::eq("category", "tech"));
    let results = col
        .text_search_with_filter("rust", 10, &filter)
        .expect("filtered text search");

    // Both rust docs (id=1, id=4) are category=tech.
    for r in &results {
        let cat = r
            .point
            .payload
            .as_ref()
            .and_then(|p| p.get("category"))
            .and_then(|v| v.as_str());
        assert_eq!(cat, Some("tech"), "all filtered results must be tech");
    }
    assert!(!results.is_empty(), "should find tech rust docs");
}

// -----------------------------------------------------------------------
// Hybrid search (vector + text)
// -----------------------------------------------------------------------

#[test]
fn test_hybrid_search_combines_vector_and_text() {
    let (_dir, col) = setup_text_collection();

    // Vector query close to point 1 ([1,0,0,0]) + text query "rust"
    let vector_query = vec![0.9, 0.1, 0.0, 0.0];
    let results = col
        .hybrid_search(&vector_query, "rust", 5, None)
        .expect("hybrid search");

    // Point 1 should rank high (close vector + "rust" in title).
    assert!(!results.is_empty(), "hybrid search should return results");
    let ids: Vec<u64> = results.iter().map(|r| r.point.id).collect();
    assert!(
        ids.contains(&1),
        "point 1 should appear (vector + text match)"
    );
}

#[test]
fn test_hybrid_search_dimension_mismatch_error() {
    let (_dir, col) = setup_text_collection();

    let bad_vec = vec![1.0, 0.0]; // dim=2, expected=4
    let result = col.hybrid_search(&bad_vec, "rust", 5, None);
    assert!(result.is_err(), "wrong dimension should error");
}