Given an IMDB list check availability on Netflix, Amazon Prime & Co.
Try it out: https://imdb-checker.herokuapp.com/
Feel free to host it on your own Heroku instance, see here.
A web-client (browser) sends an IMDb url to the server which returns the availability result after it is processed.
In order to decouple time consuming processing we use Celery. A worker picks up a task from the AMQP broker and pushes the result to the result store (Redis).
In the mean time, the frontend polls the server, whether the result is ready.
In case of SUCCESS state, client can query /get_state to retrieve the results.
The logic (which runs in Celery worker) queries JustWatch API for availability information and OMDb for media description and posters.
See Procfile for running the server on a free Heroku instance. Also refer to Heroku docs regarding more details on deployment. Note, that you have to set some environment variables, see parameters in logic/config.py.
Celery requires a broker queue to collect the requested IMDB urls and process them.
The address to this queue is set via CELERY_BROKER_URL environment variable.
If not set, Celery will use amqp://localhost.
In order to store processed results Celery requires a result store. The address to the result store is set via CELERY_RESULT_URL environment variable.
If CELERY_RESULT_URL is not set (for local development) a sqlite database is used.
For retrieving media information and posters OMDb API-key OMDB_API_KEY has to be set.
First install the requirements:
pip install -r requirements.txt
choco install rabbitmq
Set environment variables (if not set):
source .env OR call .env.bat
Start celery worker (with --pool=solo on Windows):
celery -A celery_tasks worker --loglevel=INFO --pool=solo
Start server in a new terminal:
uvicorn app:app --reload
In order to run unittests for app's logic:
pytest .\tests\