Picking a Python Interpreter (3 vs 2)

../_images/34484834733_5b80f65ab1_k_d.jpg

The State of Python (3 & 2)

When choosing a Python interpreter, one looming question is always present:“Should I choose Python 2 or Python 3”? The answer is a bit more subtle thanone might think.

The basic gist of the state of things is as follows:

  1. Most production applications today use Python 3.
  2. Python 3 is ready for the production deployment of applications today.
  3. Python 2 reached the end of its life on January 1, 2020[6].
  4. The brand name “Python” encapsulates both Python 3 and Python 2.

Recommendations

Nota

The use ofPython 3 ishighly recommended over Python 2. Consider upgrading your applications and infrastructure if you find yourselfstill using Python 2 in production today. If you are using Python 3, congratulations — you are indeed a person of excellent taste.—Kenneth Reitz

I’ll be blunt:

  • Use Python 3 for new Python applications.
  • If you’re learning Python for the first time, familiarizing yourself with Python 2.7 will be veryuseful, but not more useful than learning Python 3.
  • Learn both. They are both “Python”.

So…. 3?

If you’re choosing a Python interpreter to use, Irecommend you use the newest Python 3.x, since every version brings new andimproved standard library modules, security and bug fixes.

Given such, only use Python 2 if you have a strong reason to, such as apre-existing code-base, a Python 2 exclusive library, simplicity/familiarity,or, of course, you absolutely love and are inspired by Python 2. No harm in that.

Further Reading

It is possible towrite code that works on Python 2.6, 2.7, and Python 3. Thisranges from trivial to hard depending upon the kind of softwareyou are writing; if you’re a beginner there are far more important things toworry about.

Implementations

When people speak ofPython they often mean not just the language but alsothe CPython implementation.Python is actually a specification for a languagethat can be implemented in many different ways.

CPython

CPython is the reference implementation of Python,written in C. It compiles Python code to intermediate bytecode which is theninterpreted by a virtual machine. CPython provides the highestlevel of compatibility with Python packages and C extension modules.

If you are writing open source Python code and want to reach the widest possibleaudience, targeting CPython is best. To use packages which rely on C extensionsto function, CPython is your only implementation option.

All versions of the Python language are implemented in C because CPython is thereference implementation.

PyPy

PyPy is a Python interpreter implemented in a restrictedstatically-typed subset of the Python language called RPython. The interpreterfeatures a just-in-time compiler and supports multiple back-ends (C, CLI, JVM).

PyPy aims for maximum compatibility with the reference CPython implementationwhile improving performance.

If you are looking to increase performance of your Python code, it’sworth giving PyPy a try. On a suite of benchmarks, it’s currentlyover 5 timesfaster than CPython.

PyPy supports Python 2.7. PyPy3[1], released in beta, targets Python 3.

Jython

Jython is a Python implementation that compilesPython code to Java bytecode which is then executed by the JVM (Java Virtual Machine).Additionally, it is able to import and use any Java class like a Pythonmodule.

If you need to interface with an existing Java codebase or have other reasons toneed to write Python code for the JVM, Jython is the best choice.

Jython currently supports up to Python 2.7.[2]

IronPython

IronPython is an implementation of Python for the .NETframework. It can use both Python and .NET framework libraries, and can alsoexpose Python code to other languages in the .NET framework.

Python Tools for Visual Studio integratesIronPython directly into the Visual Studio development environment, making itan ideal choice for Windows developers.

IronPython supports Python 2.7.[3] IronPython 3[4]is being developed, but is not ready for use as of September 2020.

PythonNet

Python for .NET is a package whichprovides near seamless integration of a natively installed Pythoninstallation with the .NET Common Language Runtime (CLR). This is theinverse approach to that taken by IronPython (see above), to which itis more complementary than competing with.

In conjunction with Mono, pythonnet enables native Pythoninstallations on non-Windows operating systems, such as OS X andLinux, to operate within the .NET framework. It can be run inaddition to IronPython without conflict.

Pythonnet is compatible with Python 2.7 and 3.5-3.8.[5]

[1]https://pypy.org/compat.html
[2]https://hg.python.org/jython/file/412a8f9445f7/NEWS
[3]https://ironpython.net/download/
[4]https://github.com/IronLanguages/ironpython3
[5]https://pythonnet.github.io/
[6]https://www.python.org/dev/peps/pep-0373/#id2