How do I install "R" on OS X? Ideally using Homebrew?
There seems to be very little information online.
Ask Different is a question and answer site for power users of Apple hardware and software. It only takes a minute to sign up.
Sign up to join this communityHow do I install "R" on OS X? Ideally using Homebrew?
There seems to be very little information online.
Install Homebrew (if needed)
ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"
Install R
brew tap homebrew/science
brew install r
Error: homebrew/science was deprecated. This tap is now empty as all its formulae were migrated.
while trying out brew tap homebrew/science
– Daniel
Mar 18 '18 at 12:10
Assuming you just want to install “R” on OS X and are not interested in homebrew:
Just download the binary from CRAN https://cloud.r-project.org/
See also the R for Mac OSX FAQ that includes information on installation.
Note also the comments below suggesting that homebrew is often not the best option.
brew
fan, but it was NOT the right way to install R in my case because of gcc
dependency catastrophe; save yourself.
– hoc_age
Feb 16 '16 at 15:48
After following Matt Burns' answer, you can also install the R.app GUI via brew cask
:
$ brew tap caskroom/cask
$ brew cask install r-app
$ open /Applications/R.app
brew linkapps r-gui
instead of merely brew linkapps
.
– Steve S
Aug 27 '15 at 18:45
brew install r-gui
: Error: No available formula with the name "r-gui". It was migrated from homebrew/science to caskroom/cask/r-app. You can access it again by running: "brew tap caskroom/cask".
.
– nbro
Feb 26 '18 at 11:28
Jasons-MacBook-Pro:20190402_Tuesday_PM_captures harrison$ brew cask install r-app Error: Cask 'r-app' is unavailable: No Cask with this name exists.
– Jason Harrison
Apr 15 '19 at 4:08
You can download R for Mac OS X simply from here : http://cran.cnr.berkeley.edu
Hit the button : Download R for (Mac) OS X
Install the PKG file that came in the download.
This website might help to go ahead and download home-brew / install home brew as well.
I'm a fan of RStudio. It's an IDE that wraps R, makes visualization, organization, debugging, and other tasks much easier. Or, you can just use it as if it were a simple install of R. There's a Mac binary available from that website.
1.- http://cran.r-project.org/bin/macosx/ and download in Mac-GUI-1.62.tar.gz. 2.- INSTALL
You need R built and installed as a framework: see the 'R Installation and Administration Manual'. A CRAN binary install of R suffices.
Only Mac OS X 10.6 and higher are supported, and only 64-bit R. The project is called "R.xcodeproj" and requires Xcode 3.2 or higher.
The project can be built by selecting "R" target and "Build" inside the XCode GUI.
Supported configurations are: SnowLeopard64 (release, current OS X, default) Lion64 (release, OS X 10.7+, Xcode 4.5+) MLion64 (release, OS X 10.8+, Xcode 4.5+) Debug (with debugging output, current OS X)
The configurations differ mainly in the SDK selected (recent versions of Xcode only support the current and immediately previous SDKs, so for example in Mar 2013 the default would build for 10.8, but configuration Lion64 allows building for >= 10.7).
To build the project from the command line in the Mac-GUI directory use something like: xcodebuild -target R -configuration SnowLeopard64
To build the R for Mac OS X FAQ use either xcodebuild -target Docs or manually in docs folder makeinfo -D UseExternalXrefs --html --force --no-split RMacOSX-FAQ.texi
The resulting html FAQ file will be found in Mac-GUI/docs directory.
Note about binary compatibility:
The general rules for R apply, that is binary compatibility is given only if the major and minor version numbers match - only the patch level may differ. When using the X.Y.Z version form it means that X.Y must match. For example R-GUIs linked to 3.0.x and 3.1.x are NOT binary compatible.
The compiled R.app is usually bound to a specific version, such as 3.0.1. If you upgrade R removing the older version, let's say using R.app built for 3.0.0 and updating R to 3.0.1, you may need to fix the absolute path to libR.dylib. The nightly builds use a generic path /Library/Frameworks/R.framework/Resources/lib/libR.dylib which points to the latest version of R, but this is done by an additional call to install_name_tool in the building script. Release versions of the GUI use a fixed-version path as they come with a specific R version (in fact the default behavior doesn't depends on the GUI, but on libR.dylib - changing its own reference entry changes the way R.app is linked).
I stumbled onto this trying to install R in general and I ended up using MacPorts. If your using MacPorts, try
sudo port install R
Taken from http://johnlaudun.org/20140721-install-r-with-macports/