Designing a Full‑Lifecycle API Development Platform: From Object Modeling to Service Orchestration

Previously, when discussing API gateways, we mentioned rapid development platforms that embed fast API creation capabilities into the gateway, highlighting that API lifecycle management includes development, runtime, and operations phases.

API Full‑Lifecycle Management

The platform is divided into four subsystems: an API development platform, an API gateway engine, an API monitoring & operations platform, and an API lifecycle governance portal that integrates the three underlying systems.

Traditional ESB adapters for protocol conversion should be offloaded to the rapid development platform, which exposes standard HTTP APIs and registers them with the gateway.

Object‑Model‑Driven Design

Core to the development platform is object‑model‑driven design, allowing objects to be defined as multi‑level tree structures that map to multiple database tables, decoupling interfaces from underlying storage and supporting multi‑database scenarios.

Objects generate standard CRUD endpoints (POST, GET, DELETE, etc.) automatically, and can export contracts in RAML, YAML, WADL, or Swagger formats, as well as client SDKs for various languages.

Object‑to‑database mapping handles table creation, field mapping, and primary‑foreign key relationships, enabling immediate API availability.

API Interface Publishing

Developers can select which operations to expose (e.g., query only, import only) and configure data items for each endpoint, supporting partial object updates.

Testing tools allow online simulation and validation of published APIs, with test cases and data persisted for reuse.

Automatic documentation generation can integrate with open‑source tools like Swagger.

Common Object Operations

Typical generated endpoints include create, update by primary key, query, and delete, with open‑source tools such as npm install -g xmysql enabling MySQL tables to be exposed as REST APIs.

npm install -g xmysql
xmysql -h localhost -u mysqlUsername -p mysqlPassword -d databaseName
http://localhost:3000

These generated APIs lack built‑in permission controls and are intended for testing or as reference implementations.

Rule Processing

The platform supports input data validation (type, length, range), data‑item rule handling (mapping, enrichment, truncation), and custom scripting for low‑code extensions.

Message headers (e.g., user, token) and output fields (error codes, messages) are predefined to ensure consistent security and error handling.

Service Composition & Orchestration

Beyond individual APIs, the platform enables visual service composition: combining multiple atomic services (queries, imports) into new composite services, supporting parallel or sequential execution, data merging, and result enrichment.

Examples include aggregating contract information from several sources into a single response, or merging order header and line‑item services into a unified order API.

Source Code Export

For complex business rules, developers can export fully compilable source code from the platform, allowing custom extensions while preserving generated scaffolding.

//BeforeDo();
//ProcessAPI();
//AfterDo();

This hook structure permits pre‑processing (validation) and post‑processing (output transformation) around the core API logic.

Microservice Packaging

Multiple objects or APIs can be bundled into a microservice application, packaged as a JAR for independent deployment or hosted directly on the platform, which should integrate with a microservice runtime.

Overall, the platform aims to provide a low‑code, object‑driven environment for rapid API creation, comprehensive lifecycle management, rule handling, service orchestration, and seamless transition to production‑grade microservices.