Catsnap is a tool for managing your pictures. Never again will you lose your collection to a faulty hard drive. Never again will you run into upload limits, bandwidth maximums, or service closures.

Catsnap uses Amazon S3 and a PostgreSQL database to store and organize your images. Images are hosted for public access on S3, and each image can have one or more tags associated with it. Images can also be in an album, to organize sets of related images.

Once you store an image, you can look it up by its tags or album. Easy!

If you like, you can check out my Catsnap instance to see what it looks like when running.

Catsnap uses a web service to provide you access to your images. Let's walk through the process of getting it set up.

First, you'll need to create an Amazon WebServices account, if you don't already have one. This may take a bit of time, so be patient. Amazon will require a credit card, but the cost of running Catsnap will be tiny--a few cents a month, typically.

Once you have an AWS account, you can create a bucket in S3 where your images will be stored. Pick a name that makes sense to you. You may also want a second bucket for local/development use.

Optionally, you may wish to create a cloudfront distribution so people can download your images lightning-fast.

You need python 2.7. Clone the repo and run python setup.py develop to fetch the requirements.

Catsnap depends on Postgresql and Redis, so make sure they're installed and running. Next you need to do some configuration. Copy catsnap/config/example.config.yml to catsnap/config/config.yml and read through it. Most of the example settings will be fine for development, but you'll need to make a couple of changes. There are instructions in the example config.

You're nearly ready! Create a postgres database for catsnap to use, and set up its schema by running yoyo -b apply -d postgresql://localhost/catsnap migrations (you may need to edit that database url, depending on your postgres configuration).

You're good to go! Start the web service with gunicorn -k flask_sockets.worker catsnap.app:app -b 0.0.0.0:5000 and the background worker with celery -A catsnap.worker worker and visit localhost:5000 to get snapping!

Once you've installed the dependencies as described in Running Locally, you can launch the test suite with nosetests.

There are a number of tests that launch a browser, and they're somewhat finicky to get working. They require geckodriver, at minimum. [INSERT DEBUGGING ADVICE HERE].

If you want to skip the browser tests entirely, run nosetests --exclude splinter. You'll lose JavaScript coverage, but the Python code will still get tested.

Production use is pretty similar to dev. There are a couple of caveats:

If you'll be putting Catsnap behind a reverse proxy like Nginx, be sure to tell it to forward HTTP Upgrade headers to catsnap's /task_info route.

If you're using Heroku, you won't be able to deploy a config.yml. You can configure catsnap with environment variables (heroku config); see example.config.yml for instructions on translating option names to env vars.