autosnippet
Version:
Extract code patterns into a knowledge base for AI coding assistants
76 lines (75 loc) • 2.34 kB
TypeScript
/**
* JsonVectorAdapter — 基于 JSON 文件的向量存储实现
* 适用于中小规模（<10K 文档），无外部依赖
* 支持余弦相似度搜索、混合搜索（向量 70% + 关键词 30%）
*/
import { VectorStore } from './VectorStore.js';
export declare class JsonVectorAdapter extends VectorStore {
#private;
constructor(projectRoot: string, options?: {
contextDir?: string;
indexPath?: string;
});
init(): Promise<void>;
/**
* 同步初始化 — 供 ServiceContainer 懒加载工厂使用
* （#load 本身就是同步的 readFileSync，无需 await）
*/
initSync(): void;
upsert(item: {
id: string;
content?: string;
vector?: number[];
metadata?: Record<string, unknown>;
}): Promise<void>;
batchUpsert(items: Array<{
id: string;
content?: string;
vector?: number[];
metadata?: Record<string, unknown>;
}>): Promise<void>;
remove(id: string): Promise<void>;
getById(id: string): Promise<any>;
/** 向量相似度搜索（余弦相似度） */
searchVector(queryVector: number[], options?: {
topK?: number;
filter?: Record<string, unknown> | null;
minScore?: number;
}): Promise<{
item: any;
score: number;
}[]>;
/** 混合搜索：向量 70% + 关键词 30% */
hybridSearch(queryVector: number[], queryText: string, options?: {
topK?: number;
filter?: Record<string, unknown> | null;
}): Promise<{
item: any;
score: number;
vectorScore: number;
keywordScore: number;
}[]>;
/**
* query() — SearchEngine 使用的向量搜索别名
* 接口: query(vector, topK) → Array<{ id, similarity, metadata }>
*/
query(queryVector: number[], topK?: number): Promise<{
id: any;
similarity: number;
score: number;
content: any;
metadata: any;
}[]>;
searchByFilter(filter: Record<string, unknown>): Promise<{
[key: string]: unknown;
metadata?: Record<string, unknown>;
}[]>;
listIds(): Promise<any[]>;
clear(): Promise<void>;
getStats(): Promise<{
count: number;
indexSize: number;
indexPath: string;
hasVectors: number;
}>;
}