Warning

This is an early prototype. There aren't any known bugs, but there are very few supported API calls at this point. Stay tuned or submit a pull request!

Overview

EVEnado is an asynchronous EVE API library. It automatically caches all API responses (including error responses) for the minimum allowed time. The cache system is pluggable, and EVEnado comes with a both a Redis and SQLite3 implementation.

One thing to note is that for some API calls, caching is mandatory. As in, the request will fail if you re-request before the expiration time, which may be up to 24 hours. As a result, memcache-style data stores are not suitable for use as a cache plugin.

Usage

Usage is straightforward and best explained by example:

from evenado.client import APIClient
from evenado.sqlite import SqliteCacheProvider
from tornado import gen, ioloop

@gen.coroutine
def async_main():
    client = APIClient(SqliteProvider(), keyid=..., vcode=...)
    for char in (yield client.characters()):
        print char.name

if __name__ == '__main__':
    ioloop.IOLoop.instance().run_sync(async_main)

Of course you can also use the library within the context of a Tornado HTTPServer or anything else which can use Tornado coroutines.

Cache Providers

The Redis cache provider is suitable for a 10kqps+ multi-machine webserver environment. Even with only a single Redis process, this will not be your limiting factor given the number of subscribed EVE players.

The SQLite3 provider is a good choice for scripts, or low-traffic single-machine webservers. By default, all programs using this provider on the same machine will share a cache. Currently, SQLite I/O takes place on the main thread which will cause performance issues with concurrent access on a slow disk.