Getting your Python setup right on your system is important; otherwise, you may find yourself in the realm of confusing error messages. The following instructions focus on getting Python 3 running on Debian/Ubuntu Linux-based systems as well as on macOS via MacPorts or HomeBrew.
Tip: If you are running on an Intel-based computers (e.g., most Macs and Windows machines), you can use a Python distribution called Miniconda. This is often the easiest approach to getting a modern Python on your machine. If that's not suitable for you for any reason, continue on reading.
It is possible to use your default OS-provided Python if it is version greater than 3.x. Many distributions still ship with version 2.7 out of the box and with pip command versions corresponding to version 2.7. On some systems, a version of 3.x is available using the commands python3 and pip3. Others let you install Python 3 and pip via their package manager and use the package manager select default versions and/or names. This is true on Debian/Ubuntu based systems as well as macOS's MacPorts.
Debian-based systems still ship with older versions of Python by default. We want to use a modern Python 3.x. Make sure it is installed and available along with the related version of pip.
sudo apt install python3 python3-pip
If you are using macOS and MacPorts you can take a similar approach:
sudo port install python3
sudo port install py38-pip
At this point you should be able to envoke the Python interpreter using the python3 command. You should see the version of Python and pip with these commands:
python3 --version
python3 -m pip --version
If you are using macOS and HomeBrew, the corresponding commands are as follows:
sudo brew install python3
HomeBrew's Python 3 package installs pip3 by default. Unless you configured your copy of HomeBrew to install it elsewhere, programs should end up in /usr/local/bin on your computer. Look for a version of pip there (e.g., by running the command ls /usr/local/bin/pip* to see what gets listed).
Often you'll need some additional modules with your
Python development environment. To make sure we're using the right
Python and pip you can use the -m interpreter option to
install the desired module. Example of installing py_dataset
and Elasticsearch python modules.
python3 -m pip install py_dataset
python3 -m pip install elasticsearch