Original posted on 2017-01-08
Updated posted on 2017-07-12
With a single
brew upgrade you can save yourself a headache and valuable time when you need newer software versions.
Besides that, R binaries from CRAN and OS X stock Python can give problems with different libraries.
Here are some examples of unexpected situations of the mentioned versions:
I built my blog with Pelican. To use Pelican you need to run
pip install pelican but that library figured as "not installed" after running that command on a fresh OS X installation.
Some R packages will not work properly. For example, data.table shows this message when its loaded after running
This installation of data.table has not detected OpenMP support. It will still work but in single-threaded mode. If this is a Mac and you obtained the Mac binary of data.table from CRAN, CRAN's Mac does not support OpenMP.
Compiling R allows the user to use libraries such as OpenBLAS that can enhance performance for some operations.
You'll need Homebrew and XCode installed. How to do that? check these links
Its possible to install just Command Line Tools for XCode from Apple Developers instead of full XCode. The script below contains a line to install that.
This script will install R and Python. R will be installed with OpenBLAS, OpenMP and rJava enabled.