Fun with NASA's Open Data Portal

Getting started with an API example using NASA's Near Earth Object Web Service

by Miskatonic October 24, 2018 1:00 pm UTC 7

Favorited Favorite 2

One of the most powerful tools available to tinkerers and prototypers is the Application Program Interface. An API is the structure by which two computer programs talk to each other – think of it as a traffic cop for information traveling between two applications.

In a workshop at SXSW EDU last year, I encountered NASA's open data platform. My mind was blown by the fact that a good percentage of NASA's projects reveal their data and code through an open API, and it's all fairly well supported and reasonably easy to get started with.

To begin, I navigated to the cover page for the Open Data Portal. From there I selected the data catalog; I'm pretty into the asteroid thing these days, so I used the Near Earth Object database.

From here I clicked on the API button:

This brought me to the opening page for the NeoWs.

If you go to the Getting Started documents, it will ask you to sign up for a development key. You don't necessarily need one; you can browse the API catalog and pull out example URLs and use them at will without a key. I chose object 2018GG – there's a nifty visual of this house-sized space rock here.

After pulling out the URL for the data on 2018GG, I put together a simple Python script that I ran in Trinket.

import urllib.request
contents = urllib.request.urlopen("https://api.nasa.gov/neo/rest/v1/neo/3542519?api_key=DEMO_KEY").read()
print(contents)

When I ran this, I actually got a return in the console matching all the data available for 2018GG:

I called in the urllib library and used the request method. Next, I created a variable and stored the return coming from the URL connected to 2018GG. Finally, I printed out what's stored in the variable. That’s a lot of data for this simple call. We can simplify things by loading everything into a Json object using the code below.

import urllib.request
import json
contents = urllib.request.urlopen("https://api.nasa.gov/neo/rest/v1/neo/3802394?api_key=DEMO_KEY").read()
varab = json.loads(contents)
print(varab)

This gave me a value and an object notation connected to the value, all comma-separated.

I wanted to find out what the key titles associated with each element of data are, so I wrote a short script that separates out and prints just the keys.

import urllib.request
import json
contents = urllib.request.urlopen("https://api.nasa.gov/neo/rest/v1/neo/3802394?api_key=DEMO_KEY").read()
varab = json.loads(contents)
print(varab.keys())

The printout now shows a whole list of the keys I can work with.

Continuing to narrow things down, I picked out a single element from my list of keys and added some text to it and built a small, very specific printout. I separated out the element line I really wanted by a set of asterisks for readability.

import urllib.request
import json
contents = urllib.request.urlopen("https://api.nasa.gov/neo/rest/v1/neo/3802394?api_key=DEMO_KEY").read()
varab = json.loads(contents)
print(varab.keys())
print('*' * 79)
print(('object 3802394 is: ')+varab['close_approach_data'][0]['miss_distance']['miles']+(" miles away!!!"))
print('*' * 79)

The fun part of all this is that not only can this be implemented in the Trinket environment like I've used here, but it's a great project for Python on the Raspberry Pi. Further steps might include a warning light or buzzer if an object got too close for comfort. This example would employ the GPIO pins on the Raspberry Pi and give a great overview of connected data through an API and physical computing. Many thanks to Cameron here at SFE for wisdom and help in getting this all going!

I hope you have fun with this and happy hacking.