Skip to content

Latest commit

 

History

History
137 lines (88 loc) · 4.82 KB

README.md

File metadata and controls

137 lines (88 loc) · 4.82 KB
Raw

project-backend-template

Template to build new projects. This repository is the point to start a new backend project with, probably, everything you need.

1. Dependencies

Python

This template use python 3.9, so all dependencies have to be enabled for this version.

Postgres

We have moved to one of the most recent version, 14.5. so we will use that version from now.

Redis

We need redis to communicate with workers to run async tasks or cache.

requirements

There are two files of requirements:

  • requirements-dev.txt: contain dependencies that they will use in development environment only.
  • requirements-prod.txt: minimum number of dependencies to run in the production environment.

When you need to install both files, you can execute: pip install -r requirements-prod.txt -r requirements-dev.txt.

.env file

To execute the project, you have to create a file with the name .env with the following values:

# you can create a key here: https://miniwebtool.com/es/django-secret-key-generator/
SECRET_KEY=put_your_secret_key_here

DEBUG=True

ALLOWED_HOSTS=127.0.0.1,localhost

DB_NAME=
DB_USER=
DB_PASS=
DB_HOST=
DB_PORT=

REDIS_HOST=
REDIS_PORT=
REDIS_DB=

# needed in dev context only
CORS_ALLOWED_ORIGINS=127.0.0.1,localhost

2. Pycharm settings

You should have a reference problem in the code because the django project is in the backend folder, but the IDE is making source references from the root of the project, so you need to modify it as follows:

  • Right-click the backend folder in your project
  • Select "Mark Directory As"
  • Select "Sources Root"

3. Project structure

  • backend: web project
    • backend: core django configuration folder
    • user: django app to manage users
    • rest_api: django app to publish resources
    • files: folder to publish or save files related to project
    • static: folder to concentrate static files
  • docker: docker files to build images and run docker-compose
  • docs: documentation. It is a complement of README.md (images, other markdown pages, etc.)

4. Docker compose

Before executing the command docker-compose ..., you must create a new file in the root directory with the name .env.frontend. Its content should be:

GITHUB_TOKEN=put_your_github_token_here
# development value
VUE_APP_BASE_URL=http://localhost/backend

The variable GITHUB_TOKEN permits the build process to download the code from the frontend project. On the other hand, the variable VUE_APP_BASE_URL defines the domain name where API lives.

To switch from prod to dev and vice versa, you should always rebuild with the corresponding command.

Commands to run docker compose in prod environment

Build images: docker-compose -p backend -f docker\docker-compose.yml --env-file .env.frontend build

Execute docker-compose: docker-compose -p backend -f docker\docker-compose.yml up

Stop docker-compose: docker-compose -p backend -f docker\docker-compose.yml down

Commands to run docker compose in dev environment

Build images: docker-compose -p backend -f docker\docker-compose.yml -f docker\docker-compose-dev.yml --env-file .env.frontend build

Execute docker-compose: docker-compose -p backend -f docker\docker-compose.yml -f docker\docker-compose-dev.yml up

Stop docker-compose: docker-compose -p backend -f docker\docker-compose.yml -f docker\docker-compose-dev.yml down

Once docker-compose is up and running in dev mode, you should be able to access the backend at localhost:8081 and the front end at localhost:8080

5. Characteristic of the template

User Management System

There is an app to manage users in the system. Users should be created in django admin and accessed through login view to generate an auth token to communicate with the API. Additionally, there is another endpoint to verify the token. The goal is to verify authorization on each view change in a UI.

Async tasks

RQ will be the library to implement this

RQ Monitoring (Only available with docker-compose)

QR monitoring lets you see in a UI status of queues, jobs and tasks. More details see: https://github.com/Parallels/rq-dashboard#installing-with-docker.

OPEN API 2.0 with Swagger UI

Swagger UI uses the OPEN API 2.0 specification to display the endpoints of your API with the information necessary to call them, examples, and the possibility to call them right from the UI. You can see this on localhost:8080/api/schema/swagger-ui. To customize endpoints data you can check how to do so here.

6. Use this template

To use this project, you don't have to change it. Instead, you must create a new repo and choose this project as a template.

Once you have replicated the structure, you should search the word "# TODO" to see all places where you should make some modification.