space-track-cache

This is the code that runs the Exclosure pull-through TLE cache.

Often we want to do analyses of TLE data that spans many months/days/years. The API rate limits of space-track.org makes pulling this sort of data down quite cumbersome.

To speed this up, (and minimize our impact on their servers) we implemented a lambda+s3 cache for simple TLE queries.

These queries only allow you to pull all of the TLE data aligned to a particular UTC date, and are cached in s3 and reused to avoid pulling on space-track.

To break this down concretely the following diagram:

========== --1--> ==============        ===================
| Client |        | AWS Lambda | <--3-- | Space-Track.org |
========== <--5-- ==============        ===================
                     ^    |
                     |    4
                     2    |
                     |    V
                ==================
                | S3 Cached TLEs |
                ==================

1. The client requests TLE data for a particular day
2. The lambda attempts to pull data from the cache
3. If it isn't present in the cache, it pulls it from space-track
4. The TLE data from space track is inserted in the cache
5. The TLE data is returned to the request maker

At the time of writing the s3 cache has >90% of all days available. Additionally, the client interface will self-throttle in the unlikely case where a large number of non cached entities are desired.

Using the client

Install this module by: pip install stcache

Use it like:

import stcache
from getpass import getpass

username = input("SpaceTrack Username:")
password = getpass("SpaceTrack Password:")

print(stcache.TLEClient(username, password).get_tle_for_day(2001, 1, 1))

Development

For development, you'll want to install both the dev and server extra requirements.

pip install -e ".[dev,server]"

Note that the server extra does not install the server package, just the requirements to run the server.

Set up your pre-commit hooks:

pre-commit

Unit tests

You can run the tests locally with

pytest