kb-mcp-server

#!/usr/bin/env python3 """ Script to demonstrate proper document processing and indexing with txtai Application. """ import yaml import os from txtai.app import Application def main(): # Path to the document doc_path = "../knowledgebase/data_science.pdf" # Load the configuration config_path = "./kb.yml" print(f"Using configuration from: {config_path}") # Create application app = Application(config_path) # Clear the existing index by deleting the index directory index_path = app.config.get("path", ".txtai") if os.path.exists(index_path): print(f"Clearing existing index at: {index_path}") print(f"Processing document: {doc_path}") # Extract text segments using the textractor pipeline segments = list(app.pipelines["textractor"](doc_path)) print(f"Extracted {len(segments)} segments") # Print a sample of the first segment to verify content if segments: print("\nSample of first segment:") print(segments[0][:200] + "...\n") # Create documents with proper structure documents = [] for i, text in enumerate(segments): doc_id = f"data_science_{i}" documents.append({ "id": doc_id, "text": text, "metadata": { "source": doc_path, "index": i } }) # Add documents to the index print("Adding documents to index...") app.add(documents) # Build the index print("Building index...") app.index() # Run a test search query = "What tools are used for data visualization?" print(f"\nSearching for: '{query}'") results = app.search(query, limit=5) # Print results print("\nSearch results:") for result in results: print(f"Score: {result['score']:.4f}") print(f"Text: {result['text'][:200]}...") print("-" * 80) if __name__ == "__main__": main()