SFTP in R on a Mac

[This article was first published on R on Jason Bryer, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

I am working on a project where I need to upload PDFs generated from Rmarkdown to a SFTP server. The sftp R package is a nice wrapper to the RCurl package for handling SFTP access. But to my surprise, SFTP support is not included on Macs by default through the curl command. After some research I found the curl-openssl formula that includes SFTP support. However, since curl is a build-in program for Mac OS brew install will not install it into the PATH environment, therefore not being directly available. This function will help configure RCurl on a Mac to use the curl-openssl version so we can have SFTP access.

# First, need to install a version of CURL that supports SFTP
# brew install curl-openssl
# Verify that SFTP is a supported protocol
# /opt/homebrew/Cellar/curl/7.82.0/bin/curl -V

#' Configures CURL with openSSL support for Macs.
#' This function will try the following:
#' 1. Verify that this is being called on a Mac.
#' 2. Check to see if sftp is already available (returns gracefully so to be
#'  integrated in setup scripts).
#' 3. Checks to see if Homebrew is installed even if not currently on the PATH.
#' 4. Install curl-openssl if not already installed.
#' 5. Modify the PATH to include the openssl version of curl.
#' 6. Install RCurl from source.
#' 7. Verify sftp is available.
#' @param path the path where Homebrew packages are installed.
#' @return TRUE if sftp is available.
configure_curl_openssl <- function(
    path = '/opt/homebrew/Cellar'
) {
    if(Sys.info()['sysname'] != 'Darwin') {
        warning('This funtion only works on Mac OS.')
    if(system('which brew') == 1) {
        if(file.exists('/opt/homebrew/bin/brew')) {
            # Homebrew is installed but not on the PATH
            PATH <- Sys.getenv("PATH")
            Sys.setenv(PATH = paste("/opt/homebrew/bin", PATH, sep = ":"))
        } else {
            stop('Could not find brew. Try installing from https://brew.sh')
    if('sftp' %in% RCurl::curlVersion()$protocols) {

    curl.versions <- list.dirs(path = paste0(path, '/curl/'),
                               recursive = FALSE,
                               full.names = FALSE)
    if(length(curl.versions) == 0) { # Try install curl
        message('curl-openssl not found, trying to install using Homebrew...')
        system('brew install curl-openssl')
        curl.versions <- list.dirs(path = paste0(path, '/curl/'),
                                   recursive = FALSE,
                                   full.names = FALSE)
    if(length(curl.versions) == 0) {
        stop('Could not find or install curl-openssl.')
    version <- curl.versions[length(curl.versions)] # Use the latest version
    if('package:RCurl' %in% search()) { # Detach the RCurl package first
        detach('package:RCurl', character.only = TRUE)
    PATH <- Sys.getenv("PATH")
    Sys.setenv(PATH = paste0("/opt/homebrew/Cellar/curl/", version, "/bin:", PATH))
    message('Resinstalling RCurl from source...')
    install.packages('RCurl', type = 'source')
    if(!'sftp' %in% RCurl::curlVersion()$protocols) {
        stop('Could not configure RCurl with openssl, sorry.')

Simply running the command should do the setup. However, if things go wrong I tried to indicate where in the process something went wrong. You should only need to run this once per R installation since once RCurl has been installed from source built against the curl-openssl version of CURL, it should remember to use that version.


Once done, we can verify that SFTP access is available.

'sftp' %in% RCurl::curlVersion()$protocols # Verify sftp is available

To leave a comment for the author, please follow the link and comment on their blog: R on Jason Bryer.

R-bloggers.com offers daily e-mail updates about R news and tutorials about learning R and many other topics. Click here if you're looking to post or find an R/data-science job.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

Never miss an update!
Subscribe to R-bloggers to receive
e-mails with the latest R posts.
(You will not see this message again.)

Click here to close (This popup will not appear again)