#Python versions install#

With pyenv you can install multiple Python versions on your machine and easily switch between them.

ThatвЂ™s where version management comes in, and my preferred tool is pyenv. As a data scientist I encounter this all the time - I frequently need to rerun old analyses or projects, so I need a way to support multiple Python versions on my machine.

#Python versions upgrade#

On the other hand, you donвЂ™t want to break all of your old code when you upgrade your installed Python version. One one hand, Python is an active language and youвЂ™ll want to take advantage of the newest features ( Python 3.10 was recently released!). The following example is taken from PEP 636: # Before structural pattern matching command = input('What are you doing next? ') command_parts = command.split() if len(command_parts) = 1: if command_parts = 'quit': print('Goodbye!') quit_game() elif command_parts = 'look': current_scribe() elif len(command_parts) = 2: if command_parts = 'get': obj = command_parts character.get(obj, current_room) elif command_parts = 'go': direction = command_parts current_room = current_room.neighbor(direction) # With structural pattern matching command = input('What are you doing next? ') match command.split(): case : print('Goodbye!') quit_game() case : current_scribe() case : character.get(obj, current_room) case : current_room = current_room.neighbor(direction) Python 3.You probably wonвЂ™t use the same version of Python for all of your projects for the rest of eternity. The elephant in the room is structural pattern matching ( PEP 634, PEP 635, PEP 636). We will get better error messages ( PEP 626) which is awesome! We also get improvements for type annotations рџЋ‰ PEP 618 is the release schedule of Python 3.10 and shows that itвЂ™s getting close! And Python 3.5 will reach it in September 2020 ( source).p Python 3.6 # Reasonable type annotation syntax > from typing import List > number : List = # f-string! > bar = 3 > f'foo Python 3.10 Please also be aware that Python 2.7, Python 3.1 / 3.2 / 3.3 and 3.4 reached their end of life a while ago. enum, pathlib and unittest.mock (IвЂ™ve covered patching and mocks seperately - itвЂ™s so awesome that Python now includes this рџҐі).

#Python versions generator#

PEP-380: yield from instead of iterating over a generator end yield -ing every value.

Range is the new xrange (the old range function was dropped).There are so many new features and improvements in Python 3 compared to 2.7: # python2 thinks this is fine, python3 throws an Exception > True = False > max() 'foo' > 'foo' > 2 True # Advanced iterable unpacking: a, b, *rest = range(10)

I really hope nobody is using Python versions below 3.3 anymore. For example, when you switch to 3.5.9, you also have pip for that version.

ItвЂ™s pretty neat that it switches also the tools you install with it. Here is how you use it: # See which Python versions are available: $ pyenv install -list # Install one $ pyenv install 3.5.9 # Use one: $ pyenv local 3.5.9 ItвЂ™s a super handy tool which letвЂ™s you easily switch the Python version. pyenvД«efore we start going into details, you must know about pyenv. But those are the ones I stumble over most of the time. Are you wondering why you should switch from Python 3.6 to Python 3.7 or any other version combination? Or are you thinking about supporting Python 3.5 and want to know which features you have to avoid? Then this is your article.