How linxfeng/flow-eda works
Project type and purpose
• This is a full‑stack, microservices‑based, event‑driven low‑code flow‑engine application.
• It provides a visual, drag‑and‑drop flow editor (Vue3 & React UIs) and a Java/Spring‑Boot backend engine to run, orchestrate, and log “flows” (business processes).
File structure
• Top‑level modules:
• flow-eda-common: shared utilities, exceptions, HTTP wrappers, Dubbo APIs/models
• flow-eda-oauth2: OAuth2 authentication microservice
• flow-eda-web: web/admin microservice providing REST APIs for flow/node management
• flow-eda-runner: core flow execution engine microservice (runs nodes, multi-threaded, event callbacks)
• flow-eda-logger: log collection microservice (consumes events, writes to DB, pushes via WebSocket)
• flow-eda-vue & flow-eda-react: two alternative single‑page frontends
• Infra/config: Dockerfile, docker-compose.yml, Nacos config, application.yaml in each service
README for insights
• Architecture: Backend‑heavy, frontends are thin clients driven by backend JSON config
• Tech stack: Spring Boot, MyBatis, MySQL, Nacos (service registry/config), Dubbo (RPC), RabbitMQ (messaging), WebSocket (real‑time push), Docker/Compose
• Microservice pattern, event‑driven (pub/sub via RabbitMQ), module decoupling
Essential details without over‑splitting
• Related modules (all Spring Boot services) in a “Backend” cluster.
• Only one MySQL instance unless multiple schemas are used.
• Omit minor utility classes and focus on runtime components and integrations.