SocketChat

A lightweight real‑time group chat built with Spring Boot, WebSockets (STOMP over SockJS), and MySQL. I built this to explore simple realtime messaging with a clean, minimal stack and a single static front‑end page.

---

Screenshots

From Screenshots/ in this repo

---

What it does

• Users can “sign in” with a username/password (simple status response: created/found/wrong).
• Join any group by name; if it doesn’t exist, it’s created.
• Receive past messages for the joined group (history backfill) and live messages in realtime.
• Single static page UI with neon-ish theme.

---

Tech stack

• Backend: Spring Boot 3.5.x (Java 17)
  • Web, WebSocket (STOMP), Data JPA
• DB: MySQL (Connector/J)
• Build: Maven
• Frontend: Static HTML + SockJS + STOMP.js

Key deps are in pom.xml.

---

Project layout

• src/main/java/com/suguru/geto/socket/chat/
  • SocketChatApplication.java — Spring Boot entry point
  • config/WebSocketConfig.java — STOMP endpoint /stomp, broker /topic and /queue, prefixes /app and /user
  • controller/UserController.java — POST /api/users (create/login)
  • controller/ChatController.java — STOMP handlers
    • /app/chat.joinGroup — join group + backfill history to /user/queue/group/{name}
    • /app/chat.sendMessage — broadcast to /topic/group/{name}
  • service/ChatService.java — user/group lifecycle, persist messages
  • model/ — JPA entities: User, ChatGroup, ChatMessage
  • repository/ — Spring Data JPA repos
  • payload/ — UserRequest, ChatMessageDto
• src/main/resources/
  • application.properties — app name, datasource, JPA
  • static/index.html — single‑page UI
• Screenshots/ — PNGs used in README

---

Configuration

src/main/resources/application.properties (excerpt):

spring.application.name=SocketChat
spring.datasource.url=jdbc:mysql://localhost:3306/chat_app?createDatabaseIfNotExist=true&useSSL=false&serverTimezone=UTC
spring.datasource.username=root
spring.datasource.password=${DB_Pass}
spring.datasource.driver-class-name=com.mysql.cj.jdbc.Driver
spring.jpa.hibernate.ddl-auto=update
spring.jpa.show-sql=true

• Set DB_Pass in your environment (recommended) or replace with a literal password for local testing.
  • PowerShell: $env:DB_Pass="yourpass"
  • Bash: export DB_Pass=yourpass
• MySQL will auto‑create chat_app schema if missing (via JDBC param).

CORS for STOMP endpoint allows http://localhost:8080 by default; change in WebSocketConfig if needed.

---

Run locally

1. Requirements

• Java 17+
• Maven 3.9+
• MySQL running locally

2. Create DB (optional if using createDatabaseIfNotExist)

CREATE DATABASE IF NOT EXISTS chat_app;

3. Build & run

mvn clean package
mvn spring-boot:run
# or
java -jar target/SocketChat-0.0.1-SNAPSHOT.jar

4. Open UI

• Navigate to http://localhost:8080/ and use the page.

---

How the realtime flow works

• Client connects to SockJS endpoint /stomp and upgrades to STOMP.
• On joining a group (/app/chat.joinGroup):
  • Server ensures the group exists.
  • Past messages are sent only to the requester’s user queue: /user/queue/group/{group}.
• On sending a message (/app/chat.sendMessage):
  • Server persists the message (with timestamp) and publishes to /topic/group/{group} for everyone in that group.

---

REST and STOMP contract

• REST

  • POST /api/users → body { username, password } → response: "created" | "found" | "wrong"

• STOMP (client → server)

  • /app/chat.joinGroup — payload: string group name
  • /app/chat.sendMessage — payload JSON:

    { "sender": "alice", "content": "hi", "groupName": "general" }

• STOMP (server → client)

  • /user/queue/group/{group} — historical messages (private)
  • /topic/group/{group} — new messages (public topic)

---

Notes & caveats

• Passwords are stored in plain text (demo simplicity). For anything real:
  • Hash with BCrypt/Argon2, add validation, sessions/JWT, CSRF, etc.
• No DB migrations (Flyway/Liquibase) — schema managed by JPA ddl-auto=update.
• Single basic UI page; no build tooling.
• Allowed origins are restricted for STOMP; adjust for your front‑end host if different.

---

Roadmap (nice‑to‑have)

• Proper auth (hashing, tokens), user sessions
• Message pagination and lazy history load
• Typing indicators, presence, read receipts
• Basic moderation (ban/kick), private DMs
• Docker Compose for app + MySQL
• Migrations with Flyway
• A few integration tests

---

License

MIT for the sample code unless noted otherwise.