Rust Personal Knowledge Search Engine (PKSE)

Project Overview

The Rust Personal Knowledge Search Engine (PKSE) is a local, offline tool that allows you to search through your personal notes stored as .md or .txt files. It combines a CLI tool with an optional HTTP server to provide fast, searchable access to your documents.

PKSE is designed as a learning project for Rust, emphasizing ownership, borrowing, error handling, modular design, and idiomatic Rust practices.

---

Features

• Folder Watching: Automatically watches the ./notes directory and updates the index when files change.
• Document Loader: Reads and processes .md and .txt files into structured Document objects.
• Tokenizer: Normalizes and tokenizes text into words for indexing.
• Inverted Index: Maps words to the documents they appear in, enabling fast search.
• CLI Search Tool: Allows you to search your notes directly from the terminal.
• Optional HTTP Server: Provides a /search endpoint via Axum for programmatic access.
• Rust Engineering Practices: Modular design, proper error handling, tests, and documentation.

---

Learning Goals

This project is not just about building a search engine, it’s about learning Rust deeply through practical implementation:

• Understanding ownership, borrowing, and lifetimes
• File I/O and error handling (Result, Option)
• Working with HashMap and other collections
• Modular project structure and separation of concerns
• Async programming and event-driven design
• Writing tests and documentation

---

Project Structure

pkse/
├── src/
│   ├── main.rs      # Entry point
│   ├── index.rs     # Inverted index logic
│   ├── watcher.rs   # Folder watcher logic
│   └── search.rs    # CLI and HTTP search functions
│   └── ingestion.rs # Everything related to loading files
└── Cargo.toml       # Cargo configuration
└── README.md        # Repo readme/info file

---

How to Use

CLI Search

cargo run -- search "your query"

HTTP Server (Optional)

1. Run the server

cargo run -- server

2. Access search endpoint

GET http://localhost:3000/search?q=your+query

---

Getting Started

1. Clone the repository
2. Create a notes/ folder and add .md or .txt files
3. Build and run with cargo run
4. Start searching with the CLI or optional HTTP server

---

Contribution and Learning

This project is meant for learning Rust hands-on. You are encouraged to:

• Read Rust documentation and examples
• Experiment with the code
• Only ask AI for guidance, not full solutions
• Add tests and improve modularity

---

License

This project is open for personal learning purposes.