Testing determines whether software runs correctly based on specific inputs and identifies defects that need to be fixed.
As software scales in codebase size, it's impossible for a person or even a large team to keep up with all of the changes and the interactions between the changes. Automated testing is the only proven method for building reliable software once they grow past the point of a simple prototype. Many major software program development failures can be traced back to inadequate or a complete lack of testing.
It's impossible to know whether software works properly unless it is tested. While testing can be done manually, by a user clicking buttons or typing in input, it should be performed automatically by writing software programs that test the application under test.
There are many forms of testing and they should all be used together. When a single function of a program is isolated for testing, that is called unit testing. Testing more than a single function in an application at the same time is known as integration testing. User interface testing ensures the correctness of how a user would interact with the software. There are even more forms of testing that large programs need, such as load testing, database testing, and browser testing (for web applications).
Python software development culture is heavy on software testing. Because Python is a dynamically-typed language as opposed to a statically-typed language, testing takes on even greater importance for ensuring program correctness.
The Minimum Viable Test Suite shows how to set unit tests and integration tests for a Flask example application.
Good test, bad test explains the difference between a "good" test case and one that is not as useful. Along the way the post breaks down some myths about common testing subjects such as code coverage, assertions and mocking.
Python Testing is a site devoted to testing in - you guessed it - the Python programming language.
The case for test-driven development by Michael DeHaan explains how automation is the only way to build software at a large scale.
Google has a testing blog where they write about various aspects of testing software at scale.
Still confused about the difference between unit, functional and integration tests? Check out this top answer on Stack Overflow to that very question.