Python Modules

Python modules are installed using the DistUtils tools. This can either be done through the packaged setup.py script within the package distribution tarball:

python setup.py install

Or without downloading the package, you can use the installer tools pip:

pip install <package_name>

Or easy_isntall:

easy_install <package_name>

The above commands will install the module in the system-wide site package directory usually located at /usr/local/lib/pythonX.Y/site-packages. However, on the any machine where you do not have superuser privileges, including the Research Computing clusters, this will throw a permission error. Therefore, you need to modify the commands instructing python to use an user scheme installation.

Python User Flag

An easy way to install Python modules locally (within your user directory) is by using the User flag with the DistUtils installation methods:

python setup.py install --user

pip install --user <package_name>

The User Flag directs python to install the package in a user site package directory usually located at /lib/pythonX.Y/site-packages, with usually being a hidden directory in the user’s home directory. This is generally the preferred method of locally installing modules as this user site package is by default in the python library path: meaning modules installed this way can be used as if they were installed system-wide.

Python Home/Prefix Flag

Another way to install Python modules locally is by using the Home flag or the Prefix flag with the DistUtils installation:

$> python setup.py install --home=<dir>

Or

python setup.py install --prefix=<dir>

Using pip:

pip install --target <dir> <package_name>

Using easy_install:

easy_install --install-dir=<dir> <package_name>

With

representing the directory location you want the package installed into. These flags essentially do the same thing by directing Python to install the module in the specified directory. These directories will not be searched by default with Python. Therefore, in order to use these modules in your Python scripts you will have to modify the PYTHONPATH environment variable to include the specified directory. Or alternatively, modify sys.path from within your python script (for this method, consult python documentation.

export PYTHONPATH=<dir>

Virtual Environments

The problem with Python User, Home and Prefix flags are that they will all be searched secondary to the system-wide site package. This is a problem if you are trying to install locally a different version of a module already installed system-wide. A way to get around this is by using Python Virtual Environments.

Installing R Packages in default Directories

The two usual ways to install R packages from CRAN is either executing

install.packages('<Package Name>')

from the R prompt. Or using the command

R CMD INSTALL <package>

from the command line. The second command only works if you have already downloaded a copy of the package from CRAN or an external site. The first will automatically download the package from CRAN. R will automatically detect that you do not have permissions to write in the system-wide R library folder and will prompt if you would like to install in a local directory from within your home directory. This folder will also be checked automatically when you run R for packages, allowing you to use anything you install in this way.

Installing R Packages in non-default directories

Using install.packages()

To use install.packages from the R prompt, before you start R you need to modify the R_LIBS environment variable

export R_LIBS=<dir>

Then inside the R prompt you execute install.packages() as normal (see above).

Using R CMD INSTALL

For ‘R CMD INSTALL’ command, you can specify the path with the l flag (lowercase ‘L’)

R CMD INSTALL -l <dir> <package_name>

Using Installed Packages from non-default directories

To use locally installed packages, before you execute R you just need to modify the R_LIBS environment variable to tell R where to search for local packages.

export R_LIBS=<dir>