A minimalist MySQL-compatible server written in Go.
gosql supports basic SQL query parsing, execution, and the MySQL client/server protocol, enabling interaction with MySQL clients such as mysql CLI or GUI tools.

• MySQL client/server protocol implementation (basic)
• SQL query parsing for simple commands (SELECT, INSERT, CREATE TABLE, SHOW TABLES, etc.)
• In-memory and file-based storage backends
• Configurable via a simple settings/server.conf file
• Multi-client support with concurrency
• Basic result set encoding and error handling

gosql/
├── cmd/
│ └── gosql/ # main.go entrypoint
├── config/ # config parsing package
├── executor/ # SQL execution engine
├── parser/ # SQL parser
├── protocol/ # MySQL protocol implementation
├── storage/ # Storage backends (memory, file)
└── settings/ # Configuration files

• Go 1.20+ installed
• MySQL client or compatible GUI tool for testing

git clone https://github.com/yourusername/gosql.git
cd gosql
go build -o gosql ./cmd/gosql

Create or modify settings/server.conf:

[server]
port = 3306
data_path = data

• port: TCP port where gosql listens for connections (default 3306)
• data_path: Directory path for data storage files

Run the server:

./gosql --defaults-file=settings/server.conf

Connect with MySQL client:

mysql -h 127.0.0.1 -P 3306 -u root

Execute SQL queries such as:

CREATE TABLE users (id INT, name VARCHAR(64));
INSERT INTO users VALUES (1, 'Alice'), (2, 'Bob');
SELECT * FROM users;

• Parsing: The parser package implements a simplistic SQL parser that converts SQL text to an AST.

• Execution: The executor runs SQL commands against the storage backends.

• Storage: Supports both in-memory and file-based key-value storage.

• Protocol: Handles MySQL protocol packets, including query reading and result set writing.

• Only supports a subset of MySQL SQL syntax

• No authentication yet — connections are trusted

• Limited data types (mostly strings and integers)

• No transaction or concurrency control beyond basic Go concurrency

• Add authentication and user management
• Support more SQL syntax and data types
• Implement prepared statements
• Improve file storage durability and indexing
• Add comprehensive tests and benchmarks

Contributions are welcome! Feel free to submit issues or pull requests.

MIT License © 2025 Kenelite

For questions or support, please open an issue or contact kenelite.sg@gmail.com.