---
title: "Twitter API Data"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Twitter API Data}
  %\VignetteEngine{knitr::rmarkdown}
  \usepackage[utf8]{inputenc}
---

This vignette demonstrates the functionality of the `roomba` package on Twitter API data.

## Downloading Data

For more information on downloading Twitter data, please check out the [`httr`](https://github.com/r-lib/httr/blob/master/demo/oauth1-twitter.r) package.

The output `req` will be a nested list; you can save it using `write_rda()` from the `tidyverse` package.

```{r, eval = FALSE}
library(httr)
library(jsonlite)

oauth_endpoints("twitter")

# edit the keys with own information
myapp <- oauth_app("twitter",
                   key = "EOy06ORJM56b8mk1yoUo6bnjG",
                   secret = "8z4PMPIJrXKYE9JrALjI4TnzDJksB8xRphHj0L5JpWpSiEtbs6"
)

twitter_token <- oauth1.0_token(oauth_endpoints("twitter"), myapp)

req <- GET("https://api.twitter.com/1.1/statuses/home_timeline.json",
           config(token = twitter_token))

stop_for_status(req)

content(req)
```

## Example

We provide actual Twitter data as an example, which can be loaded using `data(twitter_data).`

```{r, message = FALSE}
library(ggplot2)
library(magrittr)
library(roomba)
```

Using the `roomba()` function will gather information based on your variables of interest (in this case, followers_count and friends_count). From there, you can use other `dplyr` functions on your data.

```{r}
twitter_data <- twitter_data

twitter_data %>% roomba(cols = c("followers_count", "friends_count")) %>% 
  ggplot(aes(x = followers_count, y = friends_count)) +
  geom_point() + 
  theme_minimal()
```