Introduction to Logging

Welcome to another Print Logs in Python by using the Logging Module.

What is Logging?

In Python, logging is typically a way to track events, that is what is happening, and whenever any program/script runs. Also, the calls of logging are added by software developers in their software to print useful steps during software code execution along with information that can be used later to track down the execution.

Proper logging can help you develop a better understanding of the flow of the program and it is very helpful in discovering those scenarios that may not have been needed or thought of during development.
Logging is very useful in debugging problems in a program.
Logs help developers get an extra set of eyes to look at the flow of the application, as they can store information, like which user ID or IP is accessing the application.
It is used for logging user data and metrics in the software application. Errors can be debugged easily, however, you can also use the data to analyze the performance of the application to plan for scaling.
An event is described as a descriptive message that can optionally contain variable data, these are data, different for each occurrence of the event.
Events or Logs printed, also have importance, and are known as Log level.

Most new developers, use the print() method in their Python code to print statements and variables, to debug their code. However, making use of Logs is very much the right method for that.

Python Logging Module

In Python, the Logging module is an inbuilt module and is powerful and ready to use. Also, the module allows writing logs either to a file or console or to any other output stream.

In addition, this module is also, used by many third-party Python libraries. This is your project uses several third-party libraries, then you can use the logging module, as this log handling can be the same in your project and the third-party libraries you employed.

Check out the example to see how to write the above.

import logging

From the above example, after we imported the logging module, we can use the different methods provided by the logging module to print the logs.

Different Levels of Log Message

There are five standard levels by default in the Logging Module, which mainly indicate the severity of events. Thus, logging functions are named after the level. Check the illustration below for levels in order of increasing severity.

The table below shows the meaning of each level.

Level	Time to use
DEBUG	It is mainly used to provide detailed information, typically of interest only when debugging problems.
INFO	It is used to confirm that things are working as expected. That is the usual information.
WARNING	It tells that something unexpected happened, but not too severe that it may affect the normal functioning of the program or software.
ERROR	It can be used to log more serious problems like errors or 1111exceptions leading to functionality getting broken.
CRITICAL	It indicates a super serious error like the application not getting started or the database being unavailable to set up the connection and many more.

One can use any of the types of log levels in its python code to log different information.

An important question here is 'what logs will be printed', actually they depend on the configuration of the logging.

Note that the default level is a warning, meaning that only events of this level and above will be tracked, and they are warning, Error, and Critical.

Python Logging Methods

In Python some convenient functions are present for simple logging usage such as; debug(), info(), warning(), error() and critical(). They are discussed in the table below:

How to Log?	Task to perform
print() method	It is used to display normal messages on a console for the user information.
warning.warn() and logging.warn()	If you want to log a warning regarding a particular runtime event. You can use the warning.warn() method in python code if the issue is avoidable and the changes must be made to eliminate the warning. You can also, use the logging.warning() method in the case there is some small issue/error that doesn't have any direct impact on the functioning of the code, however, must be tracked as it may cause a problem later. E.g. a module version mismatch, and more.
Raise an Error/Exception	This report is an error regarding a particular runtime event.
logging.error() or logging.critical()	In this case, you desire to report without raising an exception. Also, if you have done proper exception handling, but, must still log the exception which is handled for it to be found later and fixed. The method logging.critical() can be used for critical errors like program startup failed or database connection failed, and many more.
logging.info() or logging.debug()	The logging.info() method, is used to report events that occur during the normal operation of a program, for example, status monitoring or fault investigation. The method logging.debug(), is typically used for very detailed output for diagnostic purposes.

Logging Example in Python

Check out this example, in which we will use the print log messages related to the different log levels.

import logging

logging.debug('It is a debug message')	# it will not be printed
logging.info('It is an info message')	# not printed
logging.warning('OOPs!!! It is a warning')	# it will be print because it is default level
logging.error('Oops !! an error message')	# will be printed 
logging.critical('Oh!!!! it is a critical message')	# will be printed

Output:

WARNING:root: OOPs!!! It is a warning
ERROR:root: Oops !! an error message
CRITICAL:root: Oh!!!! it is a critical message

The output for the code above shows the severity level before each message along with the root. The root is the name that the logging module gives to its default logger.

In addition, the output format illustrates the level, name, and message, and all are separated by a colon(:) and thus is the default format for logs, but the format can be changed.

Note that the debug() and info() messages did not get logged. This is because, by default, the logging modules log the messages with a severity level of warning or just as mentioned above.

However, we can change that by configuring the logging module to log events of all levels if it is needed again.