il2cpp-dump-analyzer-mcp/dist/embeddings/vector-store.d.ts

Agentic RAG system for analyzing IL2CPP dump.cs files from Unity games

import { Document } from '@langchain/core/documents';
import { CodeChunk } from './chunker';
/**
 * Vector store for IL2CPP code chunks
 */
export declare class IL2CPPVectorStore {
    private vectorStore;
    private embeddings;
    private useSupabase;
    private documentHashes;
    private hashesFilePath;
    constructor(model?: string);
    /**
     * Load document hashes from file
     */
    private loadDocumentHashes;
    /**
     * Save document hashes to file
     */
    private saveDocumentHashes;
    /**
     * Generate a unique hash for a document based on its content and metadata
     * @param document Document to generate hash for
     * @returns SHA-256 hash of the document content and metadata
     */
    private generateDocumentHash;
    /**
     * Add documents to the vector store
     * @param documents Array of documents to add
     */
    addDocuments(documents: Document[]): Promise<void>;
    /**
     * Add code chunks to the vector store
     * @param chunks Array of code chunks to add
     */
    addCodeChunks(chunks: CodeChunk[]): Promise<void>;
    /**
     * Search for similar documents based on a query string
     * @param query Query string
     * @param k Number of results to return
     * @returns Array of documents with similarity scores
     */
    similaritySearch(query: string, k?: number): Promise<Document[]>;
    /**
     * Search for similar documents based on a query string with scores
     * @param query Query string
     * @param k Number of results to return
     * @returns Array of documents with similarity scores
     */
    similaritySearchWithScore(query: string, k?: number): Promise<[Document, number][]>;
    /**
     * Search for similar documents based on a vector
     * @param embedding Embedding vector
     * @param k Number of results to return
     * @returns Array of documents with similarity scores
     */
    similaritySearchVectorWithScore(embedding: number[], k?: number): Promise<[Document, number][]>;
    /**
     * Filter search results based on metadata
     * @param query Query string
     * @param filter Metadata filter
     * @param k Number of results to return
     * @returns Array of documents matching the filter
     */
    searchWithFilter(query: string, filter: Record<string, any>, k?: number): Promise<Document[]>;
    /**
     * Get the total number of documents in the vector store
     * @returns Number of documents
     */
    getDocumentCount(): Promise<number>;
    /**
     * Delete all documents from the vector store
     */
    deleteAll(): Promise<void>;
    /**
     * Check if the vector store is using Supabase
     * @returns True if using Supabase, false otherwise
     */
    isUsingSupabase(): boolean;
}