Local FAISS MCP Server

#!/usr/bin/env python3 # -*- coding: utf-8 -*- """ Standalone test for the FAISS vector store. Tests document ingestion and querying without MCP infrastructure. """ import os import sys from pathlib import Path # Import from the package from local_faiss_mcp import FAISSVectorStore # Ensure UTF-8 encoding for output on Windows (only when run as script, not pytest) def _setup_windows_encoding(): """Setup UTF-8 encoding for Windows console output when running standalone.""" if sys.platform == "win32" and not hasattr(sys, '_called_from_test'): try: import io sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8') except (AttributeError, ValueError): # Already wrapped or pytest is capturing - skip pass def test_vector_store(): """Test the FAISSVectorStore functionality.""" # Use a test directory for the index test_dir = Path(".test_vector_store") test_dir.mkdir(exist_ok=True) index_path = test_dir / "test_faiss.index" metadata_path = test_dir / "test_metadata.json" print("=" * 60) print("FAISS Vector Store Standalone Test") print("=" * 60) try: # Initialize the vector store print("\n1. Initializing FAISS vector store...") vector_store = FAISSVectorStore( index_path=str(index_path), metadata_path=str(metadata_path) ) print(" ✓ Vector store initialized successfully") # Test document ingestion print("\n2. Testing document ingestion...") doc1 = """ FAISS (Facebook AI Similarity Search) is a library for efficient similarity search and clustering of dense vectors. It contains algorithms that search in sets of vectors of any size, up to ones that possibly do not fit in RAM. It also contains supporting code for evaluation and parameter tuning. """ doc2 = """ Vector databases are specialized databases designed to store and query high-dimensional vectors efficiently. They use techniques like approximate nearest neighbor (ANN) search to find similar vectors quickly, which is essential for applications like semantic search, recommendation systems, and RAG (Retrieval-Augmented Generation). """ doc3 = """ The Model Context Protocol (MCP) is an open protocol that standardizes how applications provide context to LLMs. It enables secure, controlled interactions between AI applications and external data sources and tools. MCP servers expose resources, tools, and prompts to LLM applications. """ # Ingest documents result1 = vector_store.ingest(doc1, source="faiss_doc.txt") print(f" ✓ Document 1 ingested: {result1['chunks_added']} chunks") result2 = vector_store.ingest(doc2, source="vector_db_doc.txt") print(f" ✓ Document 2 ingested: {result2['chunks_added']} chunks") result3 = vector_store.ingest(doc3, source="mcp_doc.txt") print(f" ✓ Document 3 ingested: {result3['chunks_added']} chunks") print(f"\n Total documents in store: {result3['total_documents']}") # Test querying print("\n3. Testing semantic search queries...") queries = [ ("What is FAISS?", 2), ("How do vector databases work?", 2), ("Tell me about MCP protocol", 2), ] for query_text, top_k in queries: print(f"\n Query: '{query_text}'") print(f" Top {top_k} results:") results = vector_store.query(query_text, top_k=top_k) for i, result in enumerate(results, 1): print(f"\n [{i}] Source: {result['source']}") print(f" Distance: {result['distance']:.4f}") print(f" Text: {result['text'][:150]}...") # Test persistence print("\n4. Testing persistence...") print(" Saving vector store...") vector_store.save() print(" ✓ Vector store saved successfully") print(f"\n Index file: {index_path} ({os.path.getsize(index_path)} bytes)") print(f" Metadata file: {metadata_path} ({os.path.getsize(metadata_path)} bytes)") # Test loading print("\n5. Testing reload from disk...") vector_store_reloaded = FAISSVectorStore( index_path=str(index_path), metadata_path=str(metadata_path) ) print(" ✓ Vector store reloaded successfully") # Verify data persisted test_query = "FAISS similarity search" results_reloaded = vector_store_reloaded.query(test_query, top_k=1) if results_reloaded: print(f"\n Verification query: '{test_query}'") print(f" Found {len(results_reloaded)} result(s)") print(" ✓ Data persisted correctly") else: print(" ✗ No results found after reload!") print("\n" + "=" * 60) print("ALL TESTS PASSED ✓") print("=" * 60) except Exception as e: print(f"\n✗ ERROR: {e}") import traceback traceback.print_exc() return False finally: # Cleanup test files print("\n6. Cleaning up test files...") try: if index_path.exists(): index_path.unlink() if metadata_path.exists(): metadata_path.unlink() if test_dir.exists(): test_dir.rmdir() print(" ✓ Cleanup complete") except Exception as e: print(f" Warning: Cleanup failed: {e}") return True if __name__ == "__main__": # Setup Windows encoding only when run as standalone script _setup_windows_encoding() print("\nStarting standalone test...\n") success = test_vector_store() sys.exit(0 if success else 1)