uqa-storage 0.1.6

Document store, inverted index, IVF/HNSW vectors, B-tree, R*Tree, catalog
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

//
// Unified Query Algebra
//
// Copyright (c) 2023-2026 Cognica, Inc.
//

//! Construction and public vector-index contract for `SQLite` IVF.

use std::sync::Arc;

use uqa_core::{DocId, PostingList};

use super::writing::drop_metadata;
use super::SQLiteIVFIndex;
use crate::sqlite::vector_index::SQLiteVectorIndex;
use crate::sqlite::{ManagedConnection, Result as SQLiteResult};
use crate::vector_index::{IVFIndexParams, VectorIndex};
use crate::StorageBackendResult;

impl SQLiteIVFIndex {
    pub fn new(
        conn: ManagedConnection,
        table: impl Into<String>,
        field: impl Into<String>,
        dimensions: u32,
    ) -> Self {
        Self::from_params(conn, table, field, dimensions, IVFIndexParams::default())
    }

    pub fn with_params(
        conn: ManagedConnection,
        table: impl Into<String>,
        field: impl Into<String>,
        dimensions: u32,
        nlist: usize,
        nprobe: usize,
        train_threshold: usize,
    ) -> Self {
        Self::from_params(
            conn,
            table,
            field,
            dimensions,
            IVFIndexParams {
                nlist: nlist.max(1),
                nprobe: nprobe.max(1),
                train_threshold: train_threshold.max(1),
            },
        )
    }

    fn from_params(
        conn: ManagedConnection,
        table: impl Into<String>,
        field: impl Into<String>,
        dimensions: u32,
        params: IVFIndexParams,
    ) -> Self {
        Self {
            persistent: SQLiteVectorIndex::new(conn, table, field, dimensions),
            params,
        }
    }

    pub fn open_existing(
        conn: ManagedConnection,
        table: impl Into<String>,
        field: impl Into<String>,
        dimensions: u32,
        nlist: usize,
        nprobe: usize,
        train_threshold: usize,
    ) -> Self {
        Self::with_params(
            conn,
            table,
            field,
            dimensions,
            nlist,
            nprobe,
            train_threshold,
        )
    }

    pub fn drop_metadata(conn: &ManagedConnection, table: &str, field: &str) -> SQLiteResult<()> {
        drop_metadata(conn, table, field)
    }
}

impl VectorIndex for SQLiteIVFIndex {
    fn dimensions(&self) -> u32 {
        self.persistent.dimensions
    }

    fn index_kind(&self) -> &'static str {
        "sqlite-ivf"
    }

    fn add(&mut self, doc_id: DocId, vector: Vec<f32>) -> StorageBackendResult<()> {
        self.replace_document(doc_id, vec![vector])
    }

    fn add_many(&mut self, doc_id: DocId, vectors: Vec<Vec<f32>>) -> StorageBackendResult<()> {
        self.replace_document(doc_id, vectors)
    }

    fn delete(&mut self, doc_id: DocId) -> StorageBackendResult<()> {
        self.delete_document(doc_id)
    }

    fn clear(&mut self) -> StorageBackendResult<()> {
        self.clear_index()
    }

    fn search_knn(&self, query: &[f32], k: usize) -> StorageBackendResult<PostingList> {
        self.search_top_k(query, k)
    }

    fn search_threshold(&self, query: &[f32], threshold: f32) -> StorageBackendResult<PostingList> {
        self.persistent.search_threshold(query, threshold)
    }

    fn count(&self) -> StorageBackendResult<usize> {
        self.persistent.count()
    }

    fn initialize(&mut self) -> StorageBackendResult<()> {
        self.initialize_metadata()
    }

    fn snapshot(&self) -> StorageBackendResult<Arc<dyn VectorIndex>> {
        Ok(Arc::new(self.clone()))
    }
}