API Introduction

API Packages

api_list <- c("httr","jsonlite")
for(i in api_list){
  if(!(i %in% installed.packages())){
    install.packages(i)
  }
}
library(httr)
library(jsonlite)

Types of API Requests

• GET, PUT, and POST (likely more).

API Requests from RapidAPI

Link: https://rapidapi.com/api-sports/api/api-nba

url <- "https://api-nba-v1.p.rapidapi.com/seasons"

api_response <- VERB("GET", url, 
            add_headers('X-RapidAPI-Key' = Sys.getenv('RAPID_API'), 
            'X-RapidAPI-Host' = 'api-nba-v1.p.rapidapi.com'), 
            content_type("application/octet-stream"))
api_response

## Response [https://api-nba-v1.p.rapidapi.com/seasons]
##   Date: 2023-09-05 22:36
##   Status: 200
##   Content-Type: application/json
##   Size: 112 B

content(api_response, "text")

## No encoding supplied: defaulting to UTF-8.

## [1] "{\"get\":\"seasons\\/\",\"parameters\":[],\"errors\":[],\"results\":8,\"response\":[2015,2016,2017,2018,2019,2020,2021,2022]}"

WHAT Happens

  Response [https://api-nba-v1.p.rapidapi.com/seasons]
  Date: 2023-09-04 20:05
  Status: 200
  Content-Type: application/json
  Size: 112 B

• The status of the response should be 200 indicating that the request was successful and a response was generated.

• There should be information on the content type as well.

• A quick preview of the content can be achieved by using the code snippet below and the output is shown for demonstration purposes.

  content(api_response, "text")
  [1] "{\"get\":\"seasons\\/\",\"parameters\":[],\"errors\":[],\"results\":       8,\"response\":[2015,2016,2017,2018,2019,2020,2021,2022]}"

WHAT to do with the Data

• Original response is raw UNICODE data found in the response list

typeof(api_response)

## [1] "list"

WHAT is a json format?

• The json format which I initially learned about when studying python through the University of Michigan Runestone platform is a format of storing data as key-value pairs often also referred to as the dictionary format.

  {
    "name": "swaggy T"
    "number of years": 32
  }

How to get to JSON format?

rawToChar(api_response$content)

## [1] "{\"get\":\"seasons\\/\",\"parameters\":[],\"errors\":[],\"results\":8,\"response\":[2015,2016,2017,2018,2019,2020,2021,2022]}"

• The output here is not a typical looking dictionary or json format, rather a character vector that resembles the JSON format

How to get usable data

nba_data <- fromJSON(rawToChar(api_response$content))
nba_data <- as.data.frame(nba_data$response)
print(nba_data)

##   nba_data$response
## 1              2015
## 2              2016
## 3              2017
## 4              2018
## 5              2019
## 6              2020
## 7              2021
## 8              2022

• Try using the janitor package (perhaps use tidyverse to pipe into clean_names()) to clean the data.
• Did not work for me and likely won’t work for you.

Conclusions

• Demonstrated the use of R functions httr and jsonlite as clients for APIs hosted by RapidAPI.
• Demonstrated fine skill in observing the output and determining the right subset approach in obtaining desired data from chosen APIs
• Next will be using these APIs to build apps that automatically update data