1. The Twitter API and Authentication
Congratulations on interacting with your very first APIs and getting data from them! You're on the home stretch now.
2. Herein, you’ll learn
As a final deep dive, you're going to stream data from the Twitter API. You'll learn how to filter incoming tweets for keywords, you'll learn about the principles of API authentication and OAuth. You'll also learn the basics of the package
3. Herein, you’ll learn
tweepy, which many people in PythonLand use to interact with the Twitter API.
4. Access the Twitter API
One of the first major differences between the Twitter API and all the APIs you have seen so far is that you were able to access all the others anonymously and Twitter requires that you have an account.
In order gain access to the Twitter API, one needs to create a twitter account if you don't already have one,
5. Access the Twitter API
log into Twitter Apps and click "Create a New App" - you'll need to agree to a variety of terms and conditions here,
6. Access the Twitter API
then , go to your "Keys and Access Tokens" tab and Copy your API key, your API secret,
7. Access the Twitter API
your Access Token and your Access Token secret. These are the Authentication credentials that will allow you to access the Twitter API from Python.
In the following interactive exercises, we won't require that you create your own Twitter account and App: we'll do a mock run-through of how you would stream data and analyze it as if you had done so.
8. Twitter has a number of APIs
It is now important to mention that Twitter has a number of APIs. Firstly, they have a REST API; we won't go into the gory details of REST APIs here but I'll say two things - one: REST is short for Representational State Transfer; two: Twitter's REST API allows the user to "read and write Twitter data". In order to "monitor or process Tweets in real-time",
9. Twitter has a number of APIs
that is, to stream Twitter data, however, we'll want to use Twitter's Streaming API. In particular,
10. Twitter has a number of APIs
we'll use the public stream, which Twitter's API documentation defines as "Streams of the public data flowing through Twitter." The Public Stream itself contains a number of options. As we want to read and process tweets,
11. Twitter has a number of APIs
we'll want to use the GET statuses/sample API, which "Returns a small random sample of all public statuses."
12. Twitter has a number of APIs
If you wanted to access absolutely "All public statuses", you would need to use Twitter's Firehose API, which is not publicly available and would most likely cost you a pretty penny.
13. Tweets are returned as JSONs
One last point to note before we begin streaming tweets: tweets are returned to us as JSONs and they contain numerous possible fields. Check out the Twitter tweet field guide here. You can get tweet text, user, language, time of tweet,
14. Tweets are returned as JSONs
among many other fields.
Let's now see how to access and stream data from the Twitter API! For first-time Python tweet-streamers, I usually recommend the package tweepy,
15. Using Tweepy: Authentication
which has a nice balance of usability and capability.
Let's now use it to stream some tweets! First off, we need to create variables to store our access token and secret, plus our consumer key and secret.
16. Using Tweepy: stream tweets!!
To stream, we create an instance of tweepy's Stream class, passing our consumer key, consumer secret, access token, and access token secret. You can then stream tweets that contain keywords of choice by applying the filter method to the object stream!
In the following exercises, you'll practice writing Python code to stream tweets and then you'll do some basic analysis of these tweets to see how often particular keywords are mentioned.
17. Let's practice!
Enjoy!