Build a real-time group chat system with WebSocket connections and WeChat QR code login using this Java backend.
Learn how to combine multi-level caching with Redis and Caffeine for high-performance messaging in a Java app.
Use as a template for a full-stack Java e-commerce backend with shopping cart, orders, and payment features.
Requires MySQL, Redis, Docker, and Nginx, documentation and README are primarily written in Chinese.
MallChat is a Java backend system that combines real-time group chat with an online shopping platform. The project describes its goal as meeting the development standards used at larger internet companies, and it treats both the messaging and e-commerce features as equally important. The repository is the backend service, a separate frontend project called MallChatWeb handles the browser interface. The real-time chat component is built on Netty, a Java library for handling persistent network connections. It supports WeChat QR code login, a member list, message history, multiple message types, and interactive reactions like likes and replies. The system uses WebSocket connections between the backend and the browser client. On the shopping side, the README lists planned features including a shopping cart, orders, payment handling, product recommendations, search, user acquisition tools, push notifications, logistics tracking, and customer service. These are described as in progress. The backend technology stack includes Spring Boot for the web layer, MyBatis Plus for database access with less manual SQL writing, Redis and Caffeine for caching at different levels, Nginx for load balancing and WebSocket protocol upgrades, Docker for running the application in containers, and JWT for user authentication. Swagger provides an auto-generated API documentation interface. A live demo of the chat system was available at mallchat.cn. The project includes an accompanying documentation resource described as over 100,000 characters of guides covering setup, core features, and performance testing. Configuration is handled through a YAML file. The project is licensed under Apache 2.0. The README is written primarily in Chinese.
← zongzibinbin on gitmyhub — every repo by this author, as a profile.
Verify against the repo before relying on details.