How to install Python the smart way

Learn how to get Python upwards and running on Windows, MacOS, or Linux—and avoid the biggest pitfalls along the way

Senior Author, InfoWorld |

Table of Contents
  • Choose the right Python version and distribution
  • Install Python on Windows the smart way
  • Install Python on Linux the smart manner
  • Install Python on MacOS the smart way
  • Install Python packages the smart way
  • Install multiple Python versions side-by-side
  • Upgrade Python the smart way

Prove More than

Python is easy to use, friendly to the beginner, and powerful enough to create robust software with for nigh any application. But information technology is still a piece of software like whatever other, meaning information technology can be complex to ready upward and manage.

In this article we'll walk through how to set up Python the correct way: how to selection the appropriate version, how to keep multiple versions from stepping on one another, and how to avoid all of the other sharp edges and potential pitfalls forth the way.

Choose the right Python version and distribution

For the sake of compatibility with third-party modules, it is always safest to cull a Python version that is one major point revision behind the current one.

At the time of this writing, Python 3.8.ane is the most current version. The prophylactic bet, then, is to use the latest update of Python 3.vii (in this case, Python three.7.6). You tin can always endeavour out the most contempo version of Python in a controlled way — eastward.g., in a VM or a test machine — but going one version back guarantees the best compatibility with common third-party Python packages.

Python likewise comes in a variety of distributions, in much the same way Linux does. Unlike Linux, though, Python offers ane, gold-standard, "official" edition you can ever fall dorsum on: CPython, the version provided by the Python Software Foundation at python.org. Again, this is the safest and most broadly compatible distribution, the one nobody gets fired for picking. (Y'all might want to investigate other Python distributions later, since they accost specific use cases you might take, but we won't consider them here.)

One key choice you'll demand to make, especially on Windows, is whether to apply the 32-bit or 64-bit version of Python. The about likely respond is 64-chip, for the following reasons:

  • Virtually mod operating systems use a 64-bit edition of Python by default. Windows users can run 32-flake editions of Python on 64-bit Windows, only at a slight cost of functioning.
  • 32-flake Python, and 32-flake apps generally, tin admission just 4GB of retentivity at a time. 64-fleck applications don't have this limit, hence many data analysis and auto learning tools for Python work best in 64-bit incarnations. Some are available simply in 64-flake versions.

The only time y'all should choose the 32-scrap version of Python is if you're stuck with a 32-bit version of Windows, or yous need to use a third-party module that is available simply in a 32-bit edition.

Install Python on Windows the smart way

Python installs on Windows in much the aforementioned fashion every bit any other application, by style of an installer that guides you through the setup process.

Past default the Python installer for Windows places its executables in the user's AppData directory, and then that information technology doesn't crave administrative permissions. If y'all're the just user on the system, you might want to place Python in a higher-level directory (e.thou. C:\Python3.7) to make information technology easier to find. The Windows installer lets yous specify the target directory.

Choose the right Python installer for Windows

Python.org offers a number of different incarnations of Python for Windows. In improver to the 32-bit ("x86") and 64-bit ("x86-64") versions already mentioned, y'all can choose from the embeddable zilch file, the executable installer, and the web-based installer. Here's what those are all about:

  • The executable installer is just an .EXE file that runs the setup process for Python. This is the easy default option, and the almost commonly used.
  • The web-based installer is the same as the executable installer, except that it separately downloads the bits needed to perform the install. This dramatically reduces the size of the actual installer, but of course requires a network connection.
  • The embeddable nada file is a self-contained, minimal re-create of the Python runtime that fits in a single folder with no dependencies. It'south useful to bundle in when y'all want to distribute a Python app manually, or when you need a quick, one-off Python install to test something on the fly. But the embeddable goose egg doesn't includepip or any of the other useful tools that come with a full install, so it's for proficient use only.

Install Python using a package manager for Windows

Still another option is to use one of the package management systems that exist for Windows. NuGet, the bundle manager for .NET, offers Python in its repository. However, Python is provided in that location mainly for the sake of using information technology as a component in a .NET application, not every bit a fashion to install a standalone example of Python for general use. Yous will likely detect your Python instance easier to manage if you install Python the regular way.

Chocolatey, a more general Windows packet management system, offers Python also. Chocolatey is a user-friendly way to run the Python installer and track the presence of the Python language runtime in your organisation — and thus a better pick than NuGet. Withal, it's best to avoid mixing and matching Chocolatey installs and regular installs of Python on the same organisation.

Install Python on Linux the smart style

Because Linux distributions differ significantly, the typical manner to install Python on Linux is to use the specific distro'due south package director. Ubuntu and Fedora, for instance, take entirely different procedures for installing Python. On Linux (and MacOS), the target directory for the install is usually predetermined and based on the Python version number, east.g., /usr/bin/python3.X on Linux, or /usr/local/opt/python/ on the Mac.

One manner to avoid dealing with the intricacies of Linux parcel managers is to use a containerized Python runtime. Containers run isolated from the remainder of the organization, so you demand non worry nearly different Python runtimes stepping on each others' toes. Even so, if your workflow doesn't already include containers, you'll need to devote time and energy to getting up to speed with Docker. (Annotation that you can use containerized Python on Windows besides.)

A tool named asdf-vm also comes in handy hither. Y'all can use asdf-vm to manage multiple Python runtimes on Unix-like systems (Linux and MacOS) — and multiple runtimes for Node.js, Cerise, Elixir, and many other languages also. So if you discover yourself juggling versions of other things likewise Python, y'all'll want to look into asdf-vm.

Install Python on MacOS the smart fashion

MacOS has traditionally shipped with a version of Python installed, but never more recent than Python 2.7. This created bug when Python 3 arrived, as the two versions oft conflicted. The official Python documentation has some notes to this effect, but doesn't provide any more detailed recommendations than to make sure you use the right path for the Python instance you lot want.

A mutual way to manage Python runtimes on MacOS is through the Homebrew package manager. Homebrew provides a consistent interface for downloading, installing, managing, and removing Python and other third-political party command-line apps.

Install Python packages the smart way

In one case you take a base of operations install of a Python version set, don't start installing packages directly into it with pip — no, not fifty-fifty if you plan on using Python for just i project. Fix your project directories, install Python virtual environments into them, then install packages into those virtual environments. This mode, the base installation stays clean.

For a high-level manner to manage multiple projects with virtual environments and dependencies, look into the Poesy project. Verse provides a command-line tool for managing virtual environments and dependencies at a high level.

Install multiple Python versions side-by-side

The single hardest upshot when dealing with Python installations is how to handle unlike versions of Python installed side-by-side. Two universal rules of thumb apply hither:

  • Always install each version in a different directory.
  • Ensure that any system paths are configured to point first to the version you want to run by default.

Running multiple Python versions argues strongly in favor of per-project virtual environments. When the virtual environment is activated, all Python activeness within the context of the projection is automatically directed towards the right version of Python,

Another selection Windows users have to control which Python version to use when multiples are installed is the py launcher app. During Python setup, you're offered the option to install the py launcher, a small executable that lets you lot select (via command-line flags) which version of Python to utilise for a given script. For case, to run pip for Python iii.7, you would enterpy -3.7 -thousand pip.

Upgrade Python the smart way

Minor revision upgrades for Python — e.1000., Python three.7.ii to Python 3.7.3 — are mostly easy plenty. On Windows, the installer detects the presence of the existing version and upgrades information technology. On Linux and MacOS, the installer or package director typically does the same thing.

However, whatever virtual environments you lot take created will also demand upgrading; they don't upgrade automatically. To upgrade Python in a virtual environs, just navigate to the virtual environment directory and entervenv--upgrade. Again, note that this works best just for minor point revision upgrades — like Python 3.7.2 to Python 3.7.iii.

If you're performing a major point revision upgrade, such as Python 3.seven to Python 3.8, your best bet is to use venv to create a new, separate virtual environment subdirectory in the project directory, reinstall whatever dependencies into it, and switch to using the new virtual environment. Most IDEs with Python support (e.one thousand., Microsoft Visual Studio Lawmaking) will discover multiple virtual environments in a project and allow yous to switch between them.

Serdar Yegulalp is a senior writer at InfoWorld, focused on machine learning, containerization, devops, the Python ecosystem, and periodic reviews.

Copyright © 2020 IDG Communications, Inc.