Python Tutorial: How to Set the Path and Switch Between Different Versions/Executables (Mac & Linux)

Python version mix-ups on Mac and Linux usually come down to one thing: the command line is pointing at the wrong Python executable. When “python” lands on Python 2, “python 3” isn’t found, f-strings fail, or pip-installed packages won’t import, the fix is to identify which interpreter the system is actually running and then adjust your PATH (or your editor settings) so the intended version wins.

A common starting point is the difference between “python” and “python 3” on these systems. Many Macs ship with Python 2, so typing `python` may drop into Python 2 even after installing Python 3. Installing Python 3 often creates a separate `python3` command, so checking `python3` can confirm whether the newer interpreter is available. From there, the workflow becomes about locating executables and understanding why the shell chooses one over another.

To see where a command lives, the transcript uses `which python3` and `type python3`. Those commands reveal the executable path—often something like `/usr/local/bin/python3`—and the key reason it works is that directory order matters in the PATH environment variable. PATH is a colon-separated list of directories the shell searches in sequence. If multiple versions exist (for example, one in `/usr/bin` and another in `/usr/local/bin`), the first match in PATH gets executed. The tutorial also demonstrates running Python by full path (e.g., `/usr/local/bin/python3`) to bypass PATH ambiguity.

When PATH doesn’t include a desired install location—such as an Anaconda install—full paths still work. In the example, Anaconda’s Python executables live under an `anaconda/bin` directory, and running `/.../anaconda/bin/python` selects a specific Python version (including Python 3.6). If the goal is to make that version reachable via `python` or `python3`, the directory must be added to PATH ahead of other competing entries.

The transcript then walks through editing `~/.bash_profile` (or `~/.bashrc` on Linux) using Nano to prepend the correct Python bin directory to PATH. It emphasizes not overwriting PATH accidentally: the correct pattern appends the existing `$PATH` using `${PATH}` and then exports the updated variable. After restarting the terminal, `type python3` and `echo $PATH` confirm that the shell now resolves `python3` to the intended location.

To make the workflow smoother, aliases can map `python` to `python3` and `pip` to `pip3`, and those aliases can be made permanent by adding them to the same bash profile. For deeper debugging, the tutorial uses Python’s standard library module `sys` to print `sys.executable`, which reveals the exact interpreter currently running. This becomes crucial for import errors: pip and Python are separate commands, so `pip install Django` may install into a different Python’s `site-packages` than the one being used for imports. Using `pip show <package>` and comparing its “Location” with `sys.executable`’s `site-packages` directory pinpoints mismatches.

Finally, the transcript addresses editor confusion: even if the terminal uses the right Python, editors like Sublime Text may run a different interpreter via build systems. Creating or switching build systems to point at the correct Python executable (including virtual environments) resolves issues like f-strings not working. Virtual environments themselves automatically adjust PATH when activated, so `which python` and `sys.executable` help verify the active environment’s interpreter.