How micrusa/proyecto-aiss works
Project type and purpose
• This is a microservices-based Java application (built with Spring Boot and Maven) whose goal is to mine source-control data from Bitbucket and GitHub, normalize (“ETL”) it, and persist it in a central Git data store.
• The root folder contains three independent microservice modules:
• BitbucketMiner – pulls data from the Bitbucket API, transforms it, then pushes it onward
• GithubMiner – same pattern for the GitHub API
• GitMiner – exposes CRUD endpoints to receive normalized commits, issues, comments, projects and users, and persists them via JPA
File structure
• Top-level directories = microservices. Each is a standalone Spring Boot app with its own pom.xml, src/main/java entry point, application.properties, and unit tests.
• Common layers within each:
• controller – Spring MVC REST endpoints
• service – business logic, ETL orchestration (and in Bitbucket/Github cases, an injected GitMinerService REST client)
• model – data transfer objects, mapping external schema to internal schema
• exception – custom exceptions and handlers
• In GitMiner: an additional repository package under src/main/java for Spring Data JPA interfaces, indicating a persistence layer backed by a relational database.
Insights from README
• Confirms that each folder is a microservice.
• Provides Postman collection (pruebas_postman.json) to test each microservice’s endpoints.
Main components
a. API clients / users (e.g., Postman or other orchestration tool)
b. BitbucketMiner microservice
c. GithubMiner microservice
d. GitMiner microservice (persistence)
e. External third-party APIs: Bitbucket REST API, GitHub REST API
f. Relational Database (MySQL, Postgres, etc.) used by GitMiner
Relationships and data flows
• Users → call BitbucketMiner / GithubMiner REST endpoints
• Miner service controllers → call external REST APIs → receive raw JSON
• Transformer (ETL) components → map external JSON to internal “GitMiner” models
• Service layer → invokes GitMinerService (a REST client) to post normalized data to GitMiner
• GitMiner controllers → accept data → service → repository interfaces → persist to DB
• Optionally, users can also query GitMiner directly for stored data
Architectural patterns and design principles
• Microservice separation of concerns (one source per service + one persistence service)
• Layered architecture within each service (Controller → Service → Repository/External API)
• Use of DTOs for external‐to‐internal model transformation
• Centralized persistence via REST between services
• Spring Boot for auto-configuration and embedded server per service
• Maven for build and dependency management