Python: best_practices[1]

Write better Python code in the future.

Part 2 of our snippets and learning moments in Python. Topics I recently remembered or picked up. Here is the recollection of all this posts:

Visual Studio Code [color scheme: Night Owl (No Italics)]

ASSIGN TABLES

Convert your assignments into tables by using spaces.

# fine
group = context['group']
asset = context['asset']
task = context['task']
version = context['version']

# clear table: var | value
group   = context['group']
asset   = context['asset']
task    = context['task']
version = context['version']

# too much
task           = context['task']
version_number = context['version']

ONE-LINER

Simple one liner help readability (not everyone is a fan).

for path in paths:
    # 1 line
    if not path: continue
 
    # 2 lines
    if not path: 
        continue

PRECHECK

return/continue/break in one line instead of if-tabs.

# usual way
def change_path(path):
    if path:
        new_path = path.split('_')
    else:
        return

# precheck
# checks without additional tab levels
def change_path(path):
    if not path: return
    new_path = path.split('_')

IF SEQUENCE

Find a part in a sequence.

file_name = 'char_mike_v001_test'
group, name, version, comment = file_name.split('_')
text = 'mike'

# Too long and complicated
if group == text or name == text or comment == text:
    print('in file_name or')

# Search in list, tuple, set, ...
if text in [group, name, comment]:
    print('in file_name list')

# Search in string
if text in file_name:
    print('In file_name str')

LIST COMPREHENSION

Create a simple list in one line (faster and easier to read ... mostly).

# loop list (3 lines)
squares = []
for nr in range(10):
    squares.append(nr * nr)


# list comprehension
# one line - still easy to read
squares = [nr * nr for nr in range(10)]
# [0, 1, 4, 9, 16, 25, 36, 49, 64, 81]

DOCSTRING

Explain important processes in more detail.

This is a must in complex scripts and on a professional pipeline TD level.

def rename_read_nodes(name, read_nodes):
    """ Renames all given read nodes while adding count.

    Args:
        name [str]: future name added with a count e.g. bg_001
        read_nodes [list]: Nuke read nodes
        
    Return:
        [bool]: success
    """
    # process rename read_nodes
    return True

UNICODE

Unicode aims to list every character used by human languages.

# ASCII is the old character encoding standard used by default
# Limited range of characters
print('This is a normal message.')

# Unicode with e.g. German Umlaute
print('mögen - gären - süß')

# Defining the encoding standard (PEP 263)
# by using this comment at the beginning of the file
# -*- coding: utf-8 -*-

IMPLICIT IF

Implicitly exists or not.

path = ""

# explicit empty path/string
if path == "":
    print("No path")

# implicit empty path
if not path:
    print("No path")

# implicit existing path
if path:
    print("Path is {}".format(path))

PSEUDO CODE

Note your script step by step before filling out the blanks.

# GET selected nodes

# FIND all assets that start with "char_"

    # DELETE char asset

CONSTANTS

Constants are UPPERCASE variables which never change.

# UPPERCASE indicates constants which are defined in the beginning
# but are never overwritten.
PROJECT_PATH = '/project/'
RESOLUTION = [1920, 1080]

- Alex

PS. For more content subscribe to my Newsletter.