Deployed to Render, my microservice can communicate with any client using json via its POST, PATCH, DELETE endpoints.
- 2 example clients that interact with the RestAPI
- Web GUI Flask App to perform CRUD operations
- Jupyter Notebook to display statistics about the database
Render does take a minute to spin up because I use the free service
- Logging of CRUD activities in .log file
- Testing
- 23 parametrized unit tests in Pytest check multiple data entry possibilities including SQL injections
- CI/ CD
- A GitHub workflow runs tests in Pytest on every pull request
- Dynamically generated lists are used for data entry to help ensure data integrity
- Several Layers of data validation:
- HTML validation of user entered data
- Validation checks within the RestAPI that runs when data is received as json
- The validation in the RestAPI is checked with tests that run automatically on pull request, to ensure there is always working code in the RestAPI
- RestAPI uses parameterized queries to ensure user input is always treated as data and not as executable code
A web sequencing diagram to describe how this API interacts.
A diagram illustrating the microservice's role in separating the front end's semantic layer from the API's business logic. The green dashed box represents the database and API, which are shown together but would be decoupled in practice.
