Skip to content

kudashevs/fact-checker

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

47 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

A fact checking application

This is an example app for my presentation "Mocks aren’t Stubs. What are test doubles and how do we use them?"

Description

The goal of the application is to show how we can use different kinds of test doubles in different situations. The app provides a random interesting fact about cats by retrieving it from a public API, evaluating the truthfulness of this fact, and calculating a score.

How it works: We fetch a fact about cats from the public API. Then, we use an algorithm to assess the reliability of the fact, based on a fact’s score. Finally, the app generates an output that includes the fact with the opinion and the score.

Requirements

We are going to cover only few cases from the requirements:

  • should fetch a fact about cats
  • should assess the fact’s truthfulness and give a score
  • should log incorrect interactions with the public API
  • when a received JSON is unexpected (a business requirement):
    • notify ‘CTO’ about the problem via email
    • notify ‘programmers’ about the problem via email
    • notify ‘programmers’ about the problem via slack
  • ...

A simplified class diagram simplified-diagram

A simplified sequence diagram sequence-diagram

Disclaimer This application is supposed to be as demonstrative as possible. Therefore, it lacks separation of concerns (responsibilities) with clear boundaries, a flexible output, a mechanism for managing dependencies, etc. On the other hand, the app's logic is gathered in one place - the randomFact() method and its internal calls.

Tests

Let's examine possible test cases for our requirements:

Fetcher component

  • an expected JSON (happy path)
  • error requesting data with unspecified reason (fail path)
  • error requesting data due to a timeout (fail path)
  • an empty JSON (exceptional condition)
  • an invalid JSON (exceptional condition)
  • an unexpected JSON (exceptional condition/edge case)

Assessor component

  • an expected string (happy path)
  • an empty string (edge case)

List of tests

Fetcher component Test doubles
an expected JSON stub
error requesting data with unspecified reason stub / mock (for logging only)
error requesting data due to a timeout stub / mock (for logging only)
an empty JSON stub / mock (for logging only)
an invalid JSON stub / mock (for logging only)
an unexpected JSON stub / mock (for logging only) / spy (for notifier)
Assessor component Test doubles
an expected string stub / real implementation / never mock
an empty string stub / real implementation

License

The MIT License (MIT). Please see the License file for more information.

About

This is an example app for my presentation "Mocks aren’t Stubs. What are test doubles and how do we use them?"

Topics

Resources

License

Stars

Watchers

Forks