Querying APIs with Python

A workshop for the Art of Data Visualization conference at Columbia University, 2016 and 2017. A project by Juan Francisco Saldarriaga ([email protected]).

1. Brief introduction to Python and APIs

• Intro to APIs
• Examples of things people do with APIs
• Why Python? Python 2.7 vs Python 3
• Open a Foursquare developer account
• Copy API key
• Python IDE:
  • Mac:
    • Terminal
    • Atom
    • Sublime Text
    • Canopy
  • Windows:
    • Canopy
    • Anaconda
• Overview of process and possible outcome
• Talk about authorization process (token vs. 2 step OAuth)

2. Basic concepts of programming in Python

• Functions
• Variables:
  • Types of variables
  • Operations between variables:
    • Problems with combining certain kinds of variables together (ie. string + int or float)
  • Global vs. local
• Comments
• Loops
  • Indentation matters
• Conditionals
• Libraries

3. Querying the API

A. Building the request

1. Import library
2. Build request (string)
3. Submit request
4. Get response
5. JSON format
6. Handling errors (try/catch)

B. Parsing the response

1. Import library
2. Examine response
3. Extract data from response

C. Importing base data

1. Open file
2. Reading data from file
3. Building multiple requests with data
4. Careful with API rate limits

D. Exporting data

1. Creating new file
2. Writing data to new file