Community Data Science Course (Spring 2016)/Day 4 Lecture

Lecture Outline

Introduction and context

• You can write some tools in Python now. Congratulations!
• Today we'll learn how to find/create data sets
• Next week we'll get into data science (asking and answering questions)

Outline

• What is an API?
• How do we use one to fetch interesting datasets?
• How do we write programs that use the internet?
• How can we use the placecage API to fetch kitten pictures?
• Introduction to structured data (JSON)
• How do we use APIs in general?

What is a (web) API?

• API: a structured way for programs to talk to each other (aka an interface for programs)
• Web APIs: like a website your programs can visit (you:a website::your program:a web API)

How do we use an API to fetch datasets?

Basic idea: your program sends a request, the API sends data back

• Where do you direct your request? The site's API endpoint.
  • For example: Wikipedia's web API endpoint is http://en.wikipedia.org/w/api.php
• How do I write my request? Put together a URL; it will be different for different web APIs.
  • Check the documentation, look for code samples
• How do you send a request?
  • Python has modules you can use, like requests (they make HTTP requests)
• What do you get back?
  • Structured data (usually in the JSON format)
• How do you understand (i.e. parse) the data?
  • There's a module for that!

How do we write Python programs that make web requests?

To use APIs to build a dataset we will need:

• all our tools from last session: variables, etc
• the ability to open urls on the web
• the ability to create custom URLS
• the ability to save to files
• the ability to understand (i.e., parse) JSON data that APIs usually give us

New programming concepts

• interpolate variables into a string using % and %()s
• requests
• open files and write to them
• A little bit about URLs

Introduction to structured data (JSON, JavaScriptObjectNotation)

• what is json: useful for more structured data
• import json; json.loads()
• like Python (except no single quotes)
• simple lists, dictionaries
• can reflect more complicated data structures
• Example file at http://mako.cc/cdsw.json
• You can parse data directly with .json() on a requests call

Using other APIs

• every API is different, so read the documentation!
• If the documentation isn't helpful, search online
• for popular APIs, there are python modules that help you make requests and parse json

Possible issues:

• rate limiting
• authentication
• text encoding issues