A quick fix for your rstudio::conf FOMO. Try it on shinyapps.io or run it yourself!
UPDATE 2018-02-09: RStudio::conf 2018 is now over!
This repo now includes the tweets I gathered during the conference, so you do not need to worry about authenticating with Twitter.
You can run the app locally with a simple shiny::runGitHub("rsconf_tweets", "gadenbuie")
To run this on your own, you need to create an OAuth twitter token for rtweet
.
I followed Bob Rudis's (hrbrmstr) excellent guide from 21 Recipes for Mining Twitter with rtweet.
To have this app recognize your twitter PAT, you have a couple options:
-
You can follow all of the steps in Using OAuth to Access Twitter APIs,
-
You can save your
twitter_token
tortweet.rds
in the app directory -
You can code up another alternative by setting
.TWITTER_PAT
in a file calledtwitter_secrets.R
.
(The first one is the best answer.)
I used the following packages to make this, all of which can all be installed from CRAN:
packages = c("shiny", "rtweet", "dplyr", "stringr",
"purrr", "httr", "DT", "shinythemes",
"glue", "simpleCache")
install.packages(packages)
I used a simple twitter search for anything tagged or related to rstudioconf
.
Ininitally I was just looking at anything tagged #rstudioconf
, but I borrowed the search terms from Michael Kearney.
rstudioconf <- c("rstudioconf", "rstudio::conf",
"rstudioconference", "rstudioconference18",
"rstudioconference2018", "rstudio18",
"rstudioconf18", "rstudioconf2018",
"rstudio::conf18", "rstudio::conf2018")
The app is set to pull in new tweets if the app was last loaded more than 15 minutes ago. To get new tweets, just reload!
Check out init.r
for more details.
@kearneymw posted a script that collects tweets and makes awesome plots at https://github.com/mkearney/rstudioconf_tweets