Unit Tests

Use Python and the pytest package to test our previously created functions.
Run the tests through VSCode (in the Testing Menu) or the shell.
It primarily tests whether the function works as expected, assert actual == expected, rather than verifying the entire process's success.
For better file organization, we may create a tests folder as a Python package (__init__.py file) and put the unit tests (files and code) inside it.

Integration Tests

Tests the whole flow and process, as we compare the actual and expected responses.
Could use the DeepDiff Python library to compare 2 dictionaries (JSON documents).
Run the tests through the shell.
In the case of comparing decimal numbers, we could limit it to only a specific number of digits (e.g., 1 digit after the decimal point).
For better file organization, we may create an integration-tests folder and put the integration tests (files and code) inside it.
We may automate the integration test by creating a .sh file to:
1. Change to a specific directory
2. Run the bash command
3. Set and export the environment variables
4. Build the Docker image (could do so with a specific identifier using DATE in bash)
5. Run the docker-compose file
6. Run the integration test file
7. Close (Down/Terminate) the up services in the docker-compose file
8. Return with a non-zero response in case of test fail (As it may return 0 even if the test fails, as the script finishes/is executed successfully)

We’ll use LocalStack, a Python package, to test things locally (such as the cloud, e.g., AWS S3, AWS Kinesis, etc.) through Docker.
It could be automated by appending it to the integration test folder and scripts.

It’s a Python Enhancement Proposal that the Python community agreed on to follow some style guidelines in Python implementation.
There’s also a Python package (Now called pycodestyle) to be used as a linter.