hikma-engine/dist/modules/enhanced-search-service.d.ts

Code Knowledge Graph Indexer - A sophisticated TypeScript-based indexer that transforms Git repositories into multi-dimensional knowledge stores for AI agents

/**
 * @file Enhanced search functionality specifically designed for the embedding_nodes table.
 *       Provides semantic vector search and metadata-based queries using the unified embedding storage.
 */
import { ConfigManager } from '../config';
/**
 * Enhanced search result interface for embedding_nodes table.
 */
export interface EmbeddingSearchResult {
    node: {
        id: string;
        nodeId: string;
        nodeType: string;
        filePath: string;
        sourceText: string;
        embedding?: number[];
    };
    similarity: number;
    rank: number;
}
/**
 * Search options for embedding-based search.
 */
export interface EmbeddingSearchOptions {
    limit?: number;
    nodeTypes?: string[];
    minSimilarity?: number;
    filePaths?: string[];
    includeEmbedding?: boolean;
}
/**
 * Metadata filters for embedding nodes.
 */
export interface EmbeddingMetadataFilters {
    nodeType?: string;
    filePath?: string;
    fileExtension?: string;
    sourceTextContains?: string;
}
/**
 * Enhanced search service specifically for embedding_nodes table.
 */
export declare class EnhancedSearchService {
    private embeddingService;
    private sqliteClient;
    private config;
    private logger;
    private isInitialized;
    constructor(config: ConfigManager);
    /**
     * Initializes the enhanced search service.
     */
    initialize(): Promise<void>;
    /**
     * Performs semantic search using vector embeddings on embedding_nodes table.
     */
    semanticSearch(query: string, options?: EmbeddingSearchOptions): Promise<EmbeddingSearchResult[]>;
    /**
     * Performs text-based search when vector search is not available.
     */
    textBasedSearch(query: string, options?: EmbeddingSearchOptions): Promise<EmbeddingSearchResult[]>;
    /**
     * Performs metadata-based search on embedding_nodes.
     */
    metadataSearch(filters: EmbeddingMetadataFilters, options?: EmbeddingSearchOptions): Promise<EmbeddingSearchResult[]>;
    /**
     * Performs hybrid search combining semantic and metadata filters.
     */
    hybridSearch(query: string, filters?: EmbeddingMetadataFilters, options?: EmbeddingSearchOptions): Promise<EmbeddingSearchResult[]>;
    /**
     * Gets statistics about the embedding_nodes table.
     */
    getEmbeddingStats(): Promise<{
        totalNodes: number;
        nodeTypeBreakdown: Record<string, number>;
        filePathBreakdown: Record<string, number>;
        embeddingCoverage: number;
    }>;
    /**
     * Finds similar nodes to a given node ID.
     */
    findSimilarNodes(nodeId: string, options?: EmbeddingSearchOptions): Promise<EmbeddingSearchResult[]>;
    /**
     * Performs vector search with a given embedding.
     */
    private vectorSearchWithEmbedding;
    /**
     * Verifies that the embedding_nodes table exists and has the expected structure.
     */
    private verifyEmbeddingTable;
    /**
     * Deserializes embedding from database blob format.
     */
    private deserializeEmbedding;
    /**
     * Filter out test files from search results
     */
    private filterOutTestFiles;
    /**
     * Disconnects from the database.
     */
    disconnect(): Promise<void>;
}
//# sourceMappingURL=enhanced-search-service.d.ts.map