Skip to content

README.md

Latest commit

 

History

History
89 lines (61 loc) · 2.43 KB

README.md

File metadata and controls

89 lines (61 loc) · 2.43 KB
Raw

Description

This is the backend of the app. It has two components:

  • The Celery worker, which runs image resizing tasks
  • The Flask server, which serves the API to upload new images

Configuration

Configuration is loaded from the config.py file. By default, this file reads environment variables to configure the application:

  • CELERY_BROKER_URL: URL of the message broker
  • S3_ENDPOINT_URL: Endpoint of the S3 service
  • AWS_ACCESS_KEY_ID: Access key used for authenticating S3 operations
  • AWS_SECRET_ACCESS_KEY: Secret key used for authenticating S3 operations
  • S3_BUCKET_NAME: Bucket to use for storing images

Try out the API

The easiest way to try the API locally is to use Poetry to install dependencies, Redis as message broker and Minio for S3-compatible storage.

# Install Poetry
curl -LsSf https://astral.sh/uv/install.sh | sh

# Install dependencies
uv sync

# Run a Redis server via Docker
docker run -d -p 6379:6379 --rm --name redis redis:7.4

# Run a Minio instance via Docker
docker run -d -p 9000:9000 -p 9001:9001 --rm --name minio quay.io/minio/minio server /data --console-address ":9001"

# Set a bunch of environment variables
export CELERY_BROKER_URL=redis://localhost:6379/0
export S3_ENDPOINT_URL=http://localhost:9000
export AWS_ACCESS_KEY_ID=minioadmin
export AWS_SECRET_ACCESS_KEY=minioadmin
export S3_BUCKET_NAME=images

# Run the dev. web server
uv run image-api

# Run the celery worker
uv run celery --app image_api.worker.app worker

# Upload an image
curl -X POST \
  http://localhost:8080/image \
  --form file=@/path/to/an/image.png

Run in production

Installation (both web and worker)

Requires Python >=3.9.

uv sync --no-dev --locked

Run the web server

The application exposes a WSGI-compatible API under the image_api.web:app module. Any WSGI-compatible server can be used, like gunicorn or waitress.

Example with gunicorn:

# Install gunicorn
uv pip install gunicorn

# Run gunicorn with 4 threads on port 8080
uv run --no-dev gunicorn --workers 4 --bind 0.0.0.0:8080 image_api.web:app

See https://flask.palletsprojects.com/en/2.1.x/deploying/wsgi-standalone/

Run the worker

uv run --no-dev celery --app image_api.worker.app worker

See https://docs.celeryq.dev/en/stable/userguide/workers.html