write-logs-like-pros

What and Why?

Every automated module and scheduled code will at some point run into problems that will cause exceptions to occur. Often times, we want to log our code to monitor its progress. Using Try-Catch the classic way works fine, but is messy and often ruins the flow of the code.

Clean-Code Architecture

As you've probably read elsewhere, a function should do one thing, and only one thing. “Functions should do something, or answer something, but not both.” As he states, a function should change the state of an object, or return some information about that object, but not both. - Clean Code

How it Works

We learned from the previous quote from Martin C.Roberts that every function has only one task to do and can only succeed/fail at that spesific task; So, only one exception can happen and one success message if nothing goes wrong.

Usage

We have a simple module in logger.py that writes logs into a file:

import logging

# Create the logging module
logging.basicConfig(
    format = '%(asctime)s - %(levelname)s - %(message)s',
    datefmt = '%m/%d/%Y %I:%M:%S %p',
    filename = 'logs.log',
    level = logging.DEBUG
)
_logger = logging.getLogger(__name__)

Using the classic Try-Catch architecture:

from logger import _logger # Our logger.py file that goes along our project files

# Our function
def square(value: int): return (value * value)

try:
    result = square('Whoops!') # We can't multiple string to string
except:
    print('Cannot multiply non-int values')                       # Print on console
    _logger.error('Cannot multiply non-int values')               # Write logs in file

Using decorators:

from logger import logger # Our logger.py file that goes along our project files

# Our function but with a decorator with custome success/error messages
@logger(on_success = 'Success', on_failure = 'Cannot multiply non-int values')
def square(value: int) -> int:
  return (value * value)

square('Whoops!') # We can't multiple string to string

Further Customizations