Meal Planner

Scrapes and stores meals from hellofresh and blueapron

Tech Stack

Python + Django
Versions (as of 10/11/22):
- Python version: 3.11.0
- Django version: 4.1.3
- Postgres version: 14.1
Built with Docker and docker compose
Scripts executable in the bin directory along with just
CI built for Github and Github Actions
- Reference file: .github/workflows/ci.yml

Technical Environments

TODO: Add more details about the environments when they're up

Local Development

System Dependencies

Docker
just

Get the code

Clone the repo

git clone git@github.com:vigetlabs/meal_planner.git

Set up pre-commit and hooks

Run the setup script to install pre-commit and set up the git hooks
```
bin/setup
```
At any time, you can manually execute the hooks like this:
```
bin/lint
```
If you want to skip the hooks, you can use the --no-verify flag
```
git commit --no-verify
```
Run the project-specific commands (as needed)
```
just makemigrations
just reset
# etc.
```

Configure environment variables

If the setup script did not correctly create an .env file, create one:
```
cp .env.example .env
```
Replace the variables in your .env file with the ones you got in the steps above.

Rebuild containers if necessary

If dependencies have changed, rebuild the containers with:

just rebuild

This should fix errors like, "No module named 'Some newly added package.'"

Tests and Coverage

To run the tests without coverage:
```
bin/test
```
To run a single test file, just pass in the path, for example:
```
bin/test apps/core/tests/test_admin.py
```
If you want to run the tests and open the coverage report in your browser, run:
```
bin/cov
```

Run the application

bin/server

You can now access the app at localhost:8000

Set up data

We need to create a superuser in order to access the admin.

just createsuperuser

Use the CLI from above to create your user
Go to localhost:8000/admin
Log in using that user to access the admin

bin executables

This project contains some standard executables in the bin directory. These are intended to be used in the development environment. They are not intended to be used in production.

bin/bootstrap will install the python version and pip.
bin/setup will run the bootstrap script, create the .env file from .env.example, and set up the git hooks.
bin/update will rerun the boostrap script. Used mostly by other scripts, you will rarely run this.
bin/server will run the server.
bin/lint will lint all of the files using pre-commit.
bin/test will run the pytest test suite.
bin/cov will run the pytest test suite with coverage enabled and open the coverage file.
bin/ci will be used in CI to run the test suite, check for coverage, and fail if below 100%.

'just' commands

To make development a little easier, some common commands are set up as executables. The current available options are below. To view the commands in your terminal, run just -l.

just shell will open a shell_plus instance in your terminal.
just makemigrations will call the django makemigrations command, creating migrations file from changes to models.
just migrate will call unapplied migrations.
just reset will reset the database and rerun migrations.
just rebuild will rebuild the docker containers.
just createsuperuser will open a CLI for creating a django superuser.
just bash will open a bash shell into the docker container itself.

Adding Python Dependencies

Since this project is run and built through Docker, there is a certain workflow for adding a dependency.

Open the particular requirement file
- base.txt, local.txt, production.txt depending on what your package does
  - For a package that will be used in all environments, add to base.txt. For things just used locally (debugging, testing), add to local.txt. For production-only add to production.txt
Add the package name where it makes sense in the file
- i.e. add a new line with
```
pytest-django==4.0.4
```
in the section of other django or pytest packages
Close any running backend containers
Rebuild the backend container
- This can be done in its own step
```
just rebuild
```
OPTIONAL: Bash into the container
```
just bash
```
and check that the package is installed
```
pip list --format=freeze
```
if you did not provide a package version, you should see the version number in the output. Please put this version number in the requirements file.

DanielSBrown/meal_planner