Ghibli Movies

Ghibli movies is a simple application which, upon login, serves a page with a plain list of all movies from the Ghibli API. For each movie the people that appear in it should be listed.

Contents:

• Run using Docker
• Run in development mode
• Examples
• Tests
• Documentation:
• Prepared for Google Authentication:
• Additional Information:

Run using Docker

1. Clone repository

git clone <url>

2. Start up with docker-compose

docker-compose build --no-cache
docker-compose up -d

3. Open application and enjoy (or not :D)!

http://127.0.0.1:8000

Run in development mode

docker-compose -f docker-compose-development.yml build --no-cache
docker-compose up -d

Examples

1. First Page

2. Sign Up

3. Sign In

4. Ghibli Movies

Tests

1. Run Tests (not using selenium/geckodriver/firefox) and check coverage

docker exec -it <container-id> coverage run manage.py test
docker exec -it <container-id> coverage html
open htmlcov/index.html

Current Coverage:

2. Run Tests (including tests using selenium/geckodriver/firefox) and check coverage

   Use docker-compose-tests.yml and check docker/view_tests/Dockerfile or try to install locally geckodriver, firefox and the requirements.txt

Documentation:

The documentation was created using Sphinx.

Prepared for Google Authentication:

With the administration account add new site section and a social application (Fill those values with your OAuth details).

Additional Information:

This project uses the celery in order to run one task (get_movies) every minute. The task will use the app services to obtain all movies from Ghibli API and save that data in redis. So, the cache data is used every time that the page is load.

Since the people field on /films endpoint is broken, a list of people is called from /people endpoint and the relationship between movies and people is done. This operation is executed in services.py. Since the external API has a few data and the task is executed every minute, the operation is performed in 0.0001 seconds. However, if the API increases the movies sample size, a study must be realized to understand if we can keep this approach or if we have to pick a better one (optimization, memory, etc).

Some unit tests were applied, however with more time a more complex strategy plan should be defined. For example, we could create a mock API (e.g. developed in flask) where we can realize some tests like improve the sample of movies, create some fake data, etc.