By Lisa Tagliaferri and Melissa Anderson
A popular open-source programming language, R specializes in statistical computing and graphics. It is widely used for developing statistical software and performing data analysis. The R community is known for continuously adding user-generated packages for specific areas of study, which makes it applicable to many fields.
In this tutorial, we’ll go over how to install devtools and use it to install an R package directly from GitHub.
To follow along with this tutorial, you will need an Ubuntu 18.04 server with:
Once these prerequisites are in place, you’re ready to begin.
We’ll be installing devtools from the interactive shell, but before we do, we’ll need to install these system dependencies:
- sudo apt install build-essential libcurl4-gnutls-dev libxml2-dev libssl-dev
With these dependencies in place, we’re ready to continue.
Through devtools we’ll be able to install and build packages directly from GitHub, BitBucket, local files, and install specific versions from CRAN. To make devtools available system-wide, we’ll enter R’s shell as root:
- sudo -i R
From within the monitor, we’ll use the
install.packages() function to install
devtools from the official Comprehensive R Archive Network (CRAN).
Installation may take a while. When it’s finished, near the end of the installation output, we should see:
Output... ** testing if installed package can be loaded * DONE (devtools)
Next, we’ll put it to the test as we fetch and build a project directly from GitHub.
In this step, we’re going to install the latest development build of Shiny, a web application framework for R, directly from GitHub. We’ll do this using the
install_github function provided by devtools. A GitHub package is defined by its author (
rstudio) and its name (
shiny) which you can find from the GitHub URL of the main project page:
Use the following command to install:
Installation has successfully completed when we see the following lines near the end of the output and are returned to the R prompt:
Output. . . ** testing if installed package can be loaded * DONE (shiny) >
We can see the specific version of Shiny we’ve installed with the following command:
In the next step, we’ll take a quick look at Shiny. We’ll need to do a couple of system-level tasks first, so we’ll exit the monitor with the following command or by using CTRL-D.:
Either of these will prompt you to save your workspace image, R’s working environment that includes user-defined objects. This isn’t necessary for our purposes, so you can safely enter
Since we’ll be using a web browser to look at an example of what Shiny can do, we’re going to make sure that web traffic is allowed.
Shiny is a web application framework, so in order to view its examples in a browser, we’ll need to ensure that web traffic is allowed on our firewall. The built-in web server randomly chooses a port each time it is started unless we initiate it with a specific value. In order to make firewall management easier, we’ll specify port 4040 when we run our example.
Let’s check the status of the firewall, if we have it enabled:
- sudo ufw status
If you followed our prerequisite tutorials, only SSH is allowed, as indicated in the following output:
OutputStatus: active To Action From -- ------ ---- OpenSSH ALLOW Anywhere OpenSSH (v6) ALLOW Anywhere (v6)
You may have other rules in place or no firewall rules at all. Since only SSH traffic is permitted in this case, we’ll add port 4040 and check the status when we’re done.
- sudo ufw allow 4040/tcp
- sudo ufw status
OutputStatus: active To Action From -- ------ ---- OpenSSH ALLOW Anywhere 4040/tcp ALLOW Anywhere OpenSSH (v6) ALLOW Anywhere (v6) 4040/tcp (v6) ALLOW Anywhere (v6)
With the firewall configured, we’re ready to take a look at Shiny.
We’ll connect to R, this time as a regular user. Be sure to type
R as title case.
Next, we’ll load the Shiny package:
Shiny includes eleven built-in examples that demonstrate how it works. We’ll load the library, then run the first example. Because we are working on a remote server, we’ll specify the host address in order to browse from our local machine. We’ll also set
FALSE so it doesn’t try to launch a browser on the remote server.
- runExample("01_hello", host = "203.0.113.0", port = 4040, launch.browser= FALSE)
OutputListening on http://203.0.113.0:4040
Let’s visit this URL in a web browser:
We installed Shiny to illustrate using devtools to install directly from a GitHub repository. Then we ran the example Shiny application without elevated privileges to verify that all users have access to the package.
For a production situation, we would install from CRAN unless we had a compelling reason to install from the repository.
Let’s take a moment to return to the stable package. First, we’ll interrupt the server with
C, then exit the R shell with
D and re-enter it as root:
- sudo -i R
We can install the stable version with the following command, which will overwrite the installation from GitHub that we completed in the earlier step.
Let’s verify the change in version:
The output indicates that instead of
18.104.22.16800, the version we installed from GitHub, we’re now running the stable release.
Note: We can also find out more information about a package’s version from the system command line from its DESCRIPTION file.
- cat /usr/local/lib/R/site-library/shiny/DESCRIPTION
In this tutorial, we’ve installed the latest Shiny package directly from GitHub and learned how to reinstall its stable release from CRAN.
If you’re interested in learning more about Shiny itself, you can take a look at RStudio’s Shiny tutorial. You may also be interested in installing the open source RStudio Server, an interface to a version of R running on a remote Linux server, which brings an IDE to a server-based deployment.
Thanks for learning with the DigitalOcean Community. Check out our offerings for compute, storage, networking, and managed databases.
This textbox defaults to using Markdown to format your answer.
You can type !ref in this text area to quickly search our full set of tutorials, documentation & marketplace offerings and insert the link!
Click below to sign up and get $200 of credit to try our products over 60 days!