INTRODUCTION

SOME QUESTIONS FOR YOU:

• What is testing?

• Why do you think we should test software?

Bugs

Faults

Failures

Devs

            Defect Prevention
        

            Fault Tolerance

        

            Defect Removal
        

REMEMBER OUR TARGET AS SOFTWARE ENGINEERS: REDUCE AND MINIMIZE IMPACT OF DEFECTS

Bugs

Faults

Failures

Devs

TESTING IS MOSTLY ABOUT DETECTING FAILURES SO WE CAN REMOVE DEFECTS

• Testing is using our software to observe deviations from the expected behaviour (a failure)
• If a failure is observed, we need to look for the reason for that failure (bug) and remove it so the failure doesn't happen again
• Testing helps to remove defects after they are already part of our code

BY REMOVING THE DEFECTS WE REMOVE THE FAILURES LINKED TO THEM

INPUT

OUTPUT

Look inside your code for the origin of the failures and when detected, remove it

Is the output different to the expected one?  FAILURE

INPUT

OUTPUT

Limitations of Testing

1 MINUTE PAGE

GOING BACK TO THE QUESTIONS

• What is testing?

• Why do we test?

• Testing is the execution of software and the observation of its behavior or outcome

• It's used to detect failures and fix the linked defects (sometimes it's also used to demonstrate quality or proper behaviour)

TYPES OF TESTING

1 MINUTE TO THINK ABOUT ALL THE POSSIBLE TYPES OF TESTING YOU KNOW

DEPENDING ON THE ITEM TO BE TESTED

UNIT TESTING

COMPONENT TESTING

INTEGRATION TESTING

SYSTEM TESTING

DEPENDING ON THE WAY TO DECIDE WHEN TO STOP TESTING

COVERAGE BASED

USAGE BASED

• Controlled Software Release to "Beta" Users
• The software is released iteratively checking after every iteration if the targets (e.g. number of bugs reported, etc.) have been met and we can ship the product to non-beta users
• PROS: It resembles the REAL environment
• CONS: It must be used with care, assessing which users and at which moment should we involve

• A set of test cases are pre-defined
• They are executed
• When the results of the execution, meets the targets, then we are ready to ship
• PROS: It's a very controlled way to execute tests
• CONS: It's far away to the users

DEPENDING ON THE PURPOSE

SMOKE TESTING

CERTIFICATION TESTING

REGRESSION TESTING

DEPENDING ON HOW TESTS ARE DEFINED

FUNCTIONAL TESTING

STRUCTURAL TESTING

DEPENDING ON WHO EXECUTES THE TEST

AUTOMATED

MANUAL

KEY CONCEPTS

Test Case

WORK IN PAIRS:

What do you think is a test case?

Can you provide an example?

Test Case

A test-case is the description of the procedure to test a specific functionality of the system, explaining clearly what the  expected outcome of the execution

In software, a test-case is usually:

• Preconditions that should be established before the test is conducted.
• Clear sequence of actions and input data that constitutes the test sequence.
• Expected Result.

TEST CASES EXAMPLE

Test Cases for a function that calculates the addition of 3 numbers

Test Execution

WORK IN PAIRS:

What do you think is a test execution?

Which information should we register for a test-execution?

• Identification

• Timing

• Tester

• Transactions

• Results (and steps to reproduce)

• Screenshots, logs, etc.

Test Execution is the act of performing the tasks related to execute a test case.

A Test Case can be executed as many times as needed and hence have multiple test executions. 

TEST EXECUTION INFO

TESTING ACTIVITIES

TEST PLANNING

• Types of testing

• How to define test cases

• When to execute tests

• Resources

TEST EXECUTION

• Ensure resources assignment

• Execute Test

• Record Test Execution Results

TEST ANALYSIS & FOLLOW-UP

• Defect Fixing: For test-executions that led to a failure, look for the linked defects and fix them

• Management Decisions: Study how testing worked for this product, how are defect distributed, etc. and take remedial actions

EXAMPLE

• Define Test Cases for every User Story in every sprint
• QA Engineers are going to define the Test Cases
• Rate of Test Cases per User Story should be at least 5
• Developers are responsible of executing tests but after a user story is completed, QA engineers will also verify
• After every 3 sprints, the QA team will run a regression test. The regression test is a selection of the User Stories tests.
• Before releasing the software a regression test will be executed, no major failure should be detected.

PLANNING

EXECUTION

• The Scrum Master ensures that test are executed as planned
• Results of every test execution are registered 

FOLLOW-UP

• Team fixes defects detected
• Based on Test Execution Results, Product Owner decides when User Stories are completed
• User Stories with high % of defects detected are assessed for further understanding the reasons

STEPS TO FIX A DEFECT

WHICH STEPS YOU THINK SHOULD BE FOLLOWED TO FIX A DEFECT?

TEST CASES VS. TEST COVERAGE

Exhaustive Testing is impossible... 

1) Which test cases to select

2) How many tests cases are enough

An Example

In the book "The Art of Software Testing", Glenford Myers poses the following functional testing problem: Develop a good set of test cases for a program that accepts three numbers, a, b, c, interprets those numbers as the lengths of the sides of a triangle, and outputs the type of the triangle. Myers reports that in his experience most software developers will not respond with a good test set.