DEVOPS.md

References:

• Build Real-World Microservices with Python

DevOps Structure

1. Create Python virtual Environment python3 -m venv ~/.venv
2. Create empty files: Makefile, requirements.txt, Dockerfile, shared_code/__init__.py
3. Populate Makefile
4. Improve test coverage

3. Populate Makefile

We create a makefile:

Advantage:

• we can incrementally check every step in the local cli before runnin the Github workflow
  • eg: make test
• we simplify the steps in the workflow

install:
	#install commands
	pip install --upgrade pip &&\
	pip install -r requirements.txt
format:
	#format code
	black *.py shared_code/*.py
lint:
	#pylint with no refactor or convention msg's
	pylint --errors-only --disable=no-self-argument --extension-pkg-whitelist='pydantic' *.py shared_code/*.py
test:
	#test
	python -m pytest -vv --cov=shared_code test_api.py
build:
    #build container
deploy:
	#deploy
all: install format lint test deploy

4. Improve test coverage

Goal: We want to execute, cover, all lines of code to check execution.

4.1 run pytest

Step 1:: We have following in our makefile

test:
	#test
	python -m pytest -vv --cov=shared_code test_api.py

Step 2: We can first run locally

make test

4.2 Find uncovered code

Execute in cli: coverage html

This will create a new foled htmlcovand open index.html in browser (eg via liveserver)

Click: on the files in the browser to reveal the uncoverd lines

4.3 Cover 100%

Eliminate unnecessary code or add some code for testing purpuses.

Eg adding tests: Here we added some code in the library weatherforcast.py to cover or run every single line of code.

Appendix:

Build a your docker image

# Build your image with name = `solar-forecast-api`
docker build -t solar-forecast-api .

# Run a container it locally
docker run -p 8080:8080 --name -e OPENWEATHERMAP_API_KEY='<yourAPIkey>' cont_SFA solar-forecast-api