This API exposes a RESTful interface to query the treetracker data, capture, planter and organization and others.
This repository was created from Greenstand's template for microservice projects. This means it comes with many development tools that we use for development and deployment. As a contributor to this repository, you should learn and use these tools. They are outlined below.
- Conventional Commits
- husky
- prettier / lint
- github actions
- Jest
- TypeScript
Open terminal and navigate to a folder to install this project:
git clone https://github.com/Greenstand/treetracker-repository-name.git
Install all necessary dependencies:
npm ci
Run the server with database settings:
DATABASE_URL=[...] npm run server
Please join our slack channel to get help with setting up the database.
-
Every endpoint should have a e2e test to cover the main use cases.
-
For edge cases, we can use unit tests to test the edge cases, don't need to use e2e test to cover all cases, e2e just cover main workflow.
-
Class name should be capitalized.
-
Do not write SQL directly in
router
andmodel
files, there is a function calleddelegateRepository
can help to simplify some simple cases; -
Please use
loglevel
to replaceconsole.log
, and always use appropriate log level to log.
This project use multiple layer structure to build the whole system. The architecture follows some principles of Domain Driven Design. And inspired by this article: https://medium.com/spotlight-on-javascript/domain-driven-design-for-javascript-developers-9fc3f681931a
We compose/build the model instance in a functional style, then execute the business.
- Protocol layer
I think we can also call it the Application Layer, a term in the DDD, the entry of this project.
This microservice offers RESTFul API interface based on HTTP protocol. We use Express to handle all HTTP requests.
The Express-routers work like the controller role in MVC, they receive the requests and parameters from client, and translate it and dispatch tasks to appropriate business models. Then receive the result from them, translate to the 'view', the JSON response, to client.
- Model layer
The business model, most of the business logic is here. We are considering put most of the business logic in the model layer. So it should be the thickest layer, including all the business logic, and build up the relationship between models(business).
-
Infrastructure layer
-
Repository layer
Repository is responsible for communicating with the real database.
All the SQL statements should be here.
-
Others
For example, the Message Queue.
-
We use OpenAPI 3.0 to document the API.
You can copy the yaml file and import to swagger-ui to see the API.
To run the unit tests:
npm run test-unit
All the end to end tests are located under folder __tests__/e2e
, the test will run against the dev database.
To run the integration test:
Run tests:
npm run test-e2e
If errors:
DATABASE_URL is undefined
or
{ "code": 500,"message": "Unknown error (self signed certificate in certificate chain)" }
Follows those steps:
1- Go to the .env file, copy the DATABASE_URL with its value.
2- Add it with NODE_TLS_REJECT_UNAUTHORIZED='0' , npm run test-e2e, and run the tests.
For example:
DATABASE_URL=[the link provided] NODE_TLS_REJECT_UNAUTHORIZED='0' npm run test-e2e