README.md

Way to SW

The purpose of this simple project is to learn most of the parts used at SW.

We'll build a simple API to download videos from Youtbe.

API endpoints:

curl http://myapi.com/download -H 'Content-Type: application/json' -d '{"url": "http://youtube.com/....."}'

    HTTP/200 OK

    {
        "task": "<UUID>"
    }


curl http://myapi.com/tasks/<UUID>

    HTTP/200 OK

    {
        "status": "pending"
    }

curl http://myapi.com/tasks/<UUID>

    HTTP/200 OK

    {
        "status": "ok"
        "url": "http://myapi.com/videos/<url de la video>"
    }


curl http://myapi.com/videos/

    HTTP/200 OK

    {
        "videos": [
            {"url": "http://myapi.com", "orig_url": "http://youtube.com/....."}
        ]
    }

Step 1

Build the base system.

Database schema

At least : Table videos uuid UUID primary key src_url VARCHAR url VARCHAR

Features

Things to do in no particular order:

- A PostgreSQL database contains all the videos
- Videos are stored on the file system
- REST API build upon
  [flask-restful](http://flask-restful.readthedocs.io/)
- RabbitMQ + [Celery](http://www.celeryproject.org/) for all asynchronous
  operations
- API parameters validation with
  [Voluptuous](http://alecthomas.github.io/voluptuous/)
- Data serialization with
  [marshal](http://flask-restful-cn.readthedocs.io/en/0.3.4/fields.html)
- Packaging with `setuptools` (look at `console_scripts`)
- Everything must work inside a virtualenv

Step 2

Add database versionning with http://alembic.zzzcomputing.com/en/latest/.

Given an empty database, doing an alembic upgrade head must create all the database tables.

Step 3

Add a title field to the database to store the video title. Update the database schema with an alembic migration.