The document discusses test automation architecture and strategies for building an effective test automation pyramid. It recommends starting with unit tests at the base to test individual classes in isolation using mocks and stubs. Integration and API tests should be the next focus to test application logic through service calls. GUI tests should be limited to a fixed number of end-to-end tests while avoiding fragile tests. Cross-functional involvement from different teams is also important for maintaining tests. The goal is to have a stable, well-designed test automation strategy that finds bugs efficiently without wasting resources on fragile tests.
Sencha Roadshow 2017: Sencha Upgrades - The Good. The Bad. The Ugly - Eva Luc...
A case study into the common problems faced by companies when trying to upgrade their legacy Sencha applications. Learn about the benefits of upgrading, the common issues faced and how to avoid them in the future.
Learn how you can improve the quality of web applications through Sencha Test 2.2. We’ll demonstrate how you can build robust tests using Page Objects, visualize tests using a tree view and utilize unique locators by using the DOM Tree directly from Sencha Studio.
Developing a test automation strategy by Brian Bayer
The document discusses developing a test automation strategy and considerations for regression testing, automation, and frameworks. It addresses why regression testing is important, factors for the automation strategy like business needs and ROI, and frameworks to consider like Cucumber, Robot Framework, and Spock. The strategy determines test maintenance and fits within the organization, development process, and technology stacks.
Udvid din test portefølje med coded ui test og cloud load test
The document discusses expanding a test portfolio with Coded UI tests and cloud-based load testing in Visual Studio. It covers automating tests with Microsoft Test Manager, creating automated UI tests using Coded UI, and performing load testing on applications hosted in the cloud using Visual Studio Online. Conducting load testing early and customizing tests to measure key performance metrics can help ensure applications meet expectations before public release.
Agile testing emphasizes:
- Collaborative, integrated teams that test early and often throughout development;
- Frequent deliveries of working software that is tested as part of being "done";
- The ability to adapt testing to changing business needs.
Some key principles of agile testing include shifting the focus of testing from the last phase to being integrated throughout development, having everybody on the team test rather than just designated testers, reducing the latency between implementing code and testing it, and treating bugs as issues to be fixed quickly rather than letting them linger.
Effective agile testing incorporates practices like automated unit and integration tests, test-driven development, automated system-level regression tests, acceptance test
The document discusses evolving test automation frameworks by starting simply and solving problems over time through abstraction, separation of concerns, and domain-driven design. It advocates designing frameworks that are readable, maintainable, and unaffected by changes to the application or business rules through techniques like page objects, domain objects, and defining tests in terms of actors performing work.
A Thin Automation Framework for Manageable Automated Acceptance Testing
The document discusses creating a thin automation framework for automated acceptance testing that is test-driven, layered, and uses static page objects to make tests faster to write, easier to maintain, and more reusable. It emphasizes layering tests, framework code, and the webdriver to decouple them and keep code cleaner. The framework utilizes tests, static page/component objects, workflows, utilities, and a webdriver wrapper.
The document discusses various free tools for testing web services: SoapUI for functional testing of SOAP and REST services, JMeter for load and performance testing, and REST-assured for Java-based acceptance testing of REST services. The presenter demonstrated features of each tool, including exploratory testing and mocking with SoapUI, smoke and load testing with JMeter, and the given-when-then DSL of REST-assured. The presentation concluded that these free tools can cover most web service testing needs.
This document provides an introduction and overview of using SoapUI for testing web services. It begins with background on XML, SOAP, REST, and WSDL. It then demonstrates how to create a SoapUI project from a WSDL, view and submit sample requests and responses, validate requests, and create multiple test cases. The document provides guidance on project structure, endpoints, properties, and navigating the SoapUI interface. Finally, it includes an exercise to create test cases for a weather web service.
In our ongoing effort to help developers and QA professionals quickly get up to speed with Selenium and Appium, we're thrilled to announce the availability of our three new technical training programs - Selenium 101, Appium 101, and Sauce Labs Onboarding. Led by professional instructors with in-depth knowledge of Selenium and Appium, class sizes are small and include lectures, demos, and interactive hands-on exercises.
This document provides an overview and agenda for a presentation on automation testing using IBM Rational Functional Tester. It discusses what automation testing is, why it is useful, and when it should be implemented. It also addresses common myths about automation testing and provides tips for successful automation. Finally, it covers features of IBM Rational Functional Tester, including how to set up a test environment and record scripts to automate testing.
This document provides an overview of test automation using Cucumber and Calabash. It discusses using Cucumber to write automated test specifications in plain language and Calabash to execute those tests on Android apps. It outlines the environments, tools, and basic steps needed to get started, including installing Ruby and DevKit, creating Cucumber feature files, and using Calabash APIs to automate user interactions like tapping, entering text, and scrolling. The document also explains how to run tests on an Android app and generate an HTML report of the results.
Test Automation With Cucumber JVM, Selenium, and Mocha
Join us for an overview of behavior-driven development and test automation, which aided in the production of a Visualforce/JavaScript application for an enterprise client. Using Cucumber JVM, Selenium, Jenkins, and Git - the team was able to catch regression errors during development. We'll provide an overview of the solution used and how it worked in a real-world environment.
The document discusses testing tools in Visual Studio 2010 that help developers get code right the first time. It describes features like code analysis, code metrics, test impact analysis, IntelliTrace for debugging, and performance profiling. The tools help improve code quality, find bugs more quickly, and reduce "no repro" issues. The presentation includes demos of these tools and how they can help foster better collaboration between developers and testers.
EggPlant Functional is an automation and testing tool that automates real user workflows and executes tests automatically. It takes an image-based approach, using visual recognition to interact with applications by looking for images, text, or areas on the screen. It works across different technologies and platforms by sending mouse and keyboard inputs through VNC or RDP to remotely control other systems. This allows it to perfectly recreate and test user experiences.
This document discusses using microservices for testing and provides examples of potential test-related microservices. It describes decomposing test activities like planning, implementation, automation, execution, triage, and reporting into discrete microservices. Examples of microservices are provided for various test activities like the Core Analytics Service, Test Generation Service, BenchEngine, Results Analytics Service, and Results Comparison Service. The document argues that a microservices approach can help share functionality across products and simplify testing processes.
AUTOMATION FRAMEWORK USING SELENIUM & TESTNG ALONG WITH MAVEN
We are creating an automation framework with the help of Selenium and TestNG software in which we are testing Facebook basic functionalities like:-
1 automatic login-logout
2 Friend list check
3 Invalid login
Effective performance evaluation as part of a CI approach - Mission Impossible?
A presentation on Continuous Performance Testing.
Performance in CI:
Background and challenges - Andy Still, Co-Founder, Intechnica
An example implementation:
Mark Smith, Online QA Manager, Channel 4
This document discusses test automation, including the purpose of test automation, the test automation process, and the test automation pyramid. The key points are:
1. Test automation aims to improve test efficiency, provide wider test coverage, reduce costs, and speed up testing.
2. The test automation process involves defining the test scope, designing tests, coding tests, setting up the test environment, running tests, and maintaining automation over time.
3. The test automation pyramid illustrates that unit tests should form the base, as they are quick to write and run, while user interface tests are at the top as they are more complex and time-consuming.
This document provides an overview of testing in the agile environment and automated testing tools. It discusses agile testing methodologies like test-driven development, acceptance test-driven development, and behavior-driven development. It also covers testing web and mobile applications. The document then focuses on automated testing, describing how it works and the types of tests that are normally automated. It discusses benefits and best practices of automated testing. Finally, it covers popular automated testing tools like Selenium, Katalon Studio, and Cucumber and provides pros and cons of each.
Why we need software testing?
- Find / prevent bug
- No more nightmare when you deploy to production
- Preserve product quality
End to End Test
- Testing the complete functionality of some application
- Real application
Automation testing involves developing and executing tests that can run unattended, comparing actual and expected results. The major objectives of automation testing are to speed up testing, reduce costs and time, and increase quality. Automation testing is faster than manual testing and can test more scenarios. However, high upfront investment is required for tools and training. Both manual and automation testing are needed as automation does not cover all test cases. Popular automation testing tools include Selenium, QTP, and TestComplete.
Reaching for Your Quality Stretch Goals: Testing at Realtor.comKlaus Salchner
A/B Testing
If you are not familiar yet, an introduction to A/B testing and how you can leverage this approach to truly measure customer impact before and after a change. It's a practice highly leveraged in the e-commerce and cloud space to truly measure the impact of a change and be able to iterate through it till you see the desired outcome.
Where in your stack to invest in test automation
This short talk will explain in which layer to invest in test automation and the pros and cons. Too many teams still invest heavily in automated UI testing which then results in large test automation suites once the platform grows while still not being able to catch critical quality issues before they reach customers.
Testing for reliability, resilience and recovery
Your customer experience is also impacted by how reliable your application is. How do you test for reliability. But also how do you build and test for resilience, as guaranteed reliability is unachievable and the closer you get the costlier it becomes. Lastly how do you test for recovery, so once an outage or partial outage happened how to you recover, and how do you prepare for that recovery.
Sencha Roadshow 2017: Best Practices for Implementing Continuous Web App TestingSencha
Learn how to create end-to-end functional tests quickly across multiple browsers simultaneously and scale the automated test suite to over thousands of test cases and cross-browser combinations for a complete regression cycle. We will demonstrate how we are able to locate a component, generate test code, and execute tests from TeamCity.
Sustainable Automation Frameworks by Kelsey ShannahanQA or the Highway
The document discusses sustainable automation frameworks for testing software. It addresses common problems such as hardcoding tests, having too many step definitions, and poor data management. The proposed solutions include using a page object model to organize test code, standardizing step definitions to represent business logic at a higher level, and having a consistent way to load and reuse test data. Maintaining an organized framework is emphasized as important for allowing tests to run quickly and changes to be made easily.
Sencha Roadshow 2017: Sencha Upgrades - The Good. The Bad. The Ugly - Eva Luc...Sencha
A case study into the common problems faced by companies when trying to upgrade their legacy Sencha applications. Learn about the benefits of upgrading, the common issues faced and how to avoid them in the future.
Sencha Roadshow 2017: What's New in Sencha TestSencha
Learn how you can improve the quality of web applications through Sencha Test 2.2. We’ll demonstrate how you can build robust tests using Page Objects, visualize tests using a tree view and utilize unique locators by using the DOM Tree directly from Sencha Studio.
Developing a test automation strategy by Brian BayerQA or the Highway
The document discusses developing a test automation strategy and considerations for regression testing, automation, and frameworks. It addresses why regression testing is important, factors for the automation strategy like business needs and ROI, and frameworks to consider like Cucumber, Robot Framework, and Spock. The strategy determines test maintenance and fits within the organization, development process, and technology stacks.
Udvid din test portefølje med coded ui test og cloud load testPeter Lindberg
The document discusses expanding a test portfolio with Coded UI tests and cloud-based load testing in Visual Studio. It covers automating tests with Microsoft Test Manager, creating automated UI tests using Coded UI, and performing load testing on applications hosted in the cloud using Visual Studio Online. Conducting load testing early and customizing tests to measure key performance metrics can help ensure applications meet expectations before public release.
Agile testing emphasizes:
- Collaborative, integrated teams that test early and often throughout development;
- Frequent deliveries of working software that is tested as part of being "done";
- The ability to adapt testing to changing business needs.
Some key principles of agile testing include shifting the focus of testing from the last phase to being integrated throughout development, having everybody on the team test rather than just designated testers, reducing the latency between implementing code and testing it, and treating bugs as issues to be fixed quickly rather than letting them linger.
Effective agile testing incorporates practices like automated unit and integration tests, test-driven development, automated system-level regression tests, acceptance test
The document discusses evolving test automation frameworks by starting simply and solving problems over time through abstraction, separation of concerns, and domain-driven design. It advocates designing frameworks that are readable, maintainable, and unaffected by changes to the application or business rules through techniques like page objects, domain objects, and defining tests in terms of actors performing work.
A Thin Automation Framework for Manageable Automated Acceptance TestingExcella
The document discusses creating a thin automation framework for automated acceptance testing that is test-driven, layered, and uses static page objects to make tests faster to write, easier to maintain, and more reusable. It emphasizes layering tests, framework code, and the webdriver to decouple them and keep code cleaner. The framework utilizes tests, static page/component objects, workflows, utilities, and a webdriver wrapper.
The document discusses various free tools for testing web services: SoapUI for functional testing of SOAP and REST services, JMeter for load and performance testing, and REST-assured for Java-based acceptance testing of REST services. The presenter demonstrated features of each tool, including exploratory testing and mocking with SoapUI, smoke and load testing with JMeter, and the given-when-then DSL of REST-assured. The presentation concluded that these free tools can cover most web service testing needs.
This document provides an introduction and overview of using SoapUI for testing web services. It begins with background on XML, SOAP, REST, and WSDL. It then demonstrates how to create a SoapUI project from a WSDL, view and submit sample requests and responses, validate requests, and create multiple test cases. The document provides guidance on project structure, endpoints, properties, and navigating the SoapUI interface. Finally, it includes an exercise to create test cases for a weather web service.
Selenium and Appium Training from Sauce LabsSauce Labs
In our ongoing effort to help developers and QA professionals quickly get up to speed with Selenium and Appium, we're thrilled to announce the availability of our three new technical training programs - Selenium 101, Appium 101, and Sauce Labs Onboarding. Led by professional instructors with in-depth knowledge of Selenium and Appium, class sizes are small and include lectures, demos, and interactive hands-on exercises.
This document provides an overview and agenda for a presentation on automation testing using IBM Rational Functional Tester. It discusses what automation testing is, why it is useful, and when it should be implemented. It also addresses common myths about automation testing and provides tips for successful automation. Finally, it covers features of IBM Rational Functional Tester, including how to set up a test environment and record scripts to automate testing.
This document provides an overview of test automation using Cucumber and Calabash. It discusses using Cucumber to write automated test specifications in plain language and Calabash to execute those tests on Android apps. It outlines the environments, tools, and basic steps needed to get started, including installing Ruby and DevKit, creating Cucumber feature files, and using Calabash APIs to automate user interactions like tapping, entering text, and scrolling. The document also explains how to run tests on an Android app and generate an HTML report of the results.
Join us for an overview of behavior-driven development and test automation, which aided in the production of a Visualforce/JavaScript application for an enterprise client. Using Cucumber JVM, Selenium, Jenkins, and Git - the team was able to catch regression errors during development. We'll provide an overview of the solution used and how it worked in a real-world environment.
Visual Studio 2010 Testing for DevelopersSteve Lange
The document discusses testing tools in Visual Studio 2010 that help developers get code right the first time. It describes features like code analysis, code metrics, test impact analysis, IntelliTrace for debugging, and performance profiling. The tools help improve code quality, find bugs more quickly, and reduce "no repro" issues. The presentation includes demos of these tools and how they can help foster better collaboration between developers and testers.
EggPlant Functional is an automation and testing tool that automates real user workflows and executes tests automatically. It takes an image-based approach, using visual recognition to interact with applications by looking for images, text, or areas on the screen. It works across different technologies and platforms by sending mouse and keyboard inputs through VNC or RDP to remotely control other systems. This allows it to perfectly recreate and test user experiences.
This document discusses using microservices for testing and provides examples of potential test-related microservices. It describes decomposing test activities like planning, implementation, automation, execution, triage, and reporting into discrete microservices. Examples of microservices are provided for various test activities like the Core Analytics Service, Test Generation Service, BenchEngine, Results Analytics Service, and Results Comparison Service. The document argues that a microservices approach can help share functionality across products and simplify testing processes.
AUTOMATION FRAMEWORK USING SELENIUM & TESTNG ALONG WITH MAVENSHASHI JAISWAL
We are creating an automation framework with the help of Selenium and TestNG software in which we are testing Facebook basic functionalities like:-
1 automatic login-logout
2 Friend list check
3 Invalid login
Effective performance evaluation as part of a CI approach - Mission Impossible?Intechnica
A presentation on Continuous Performance Testing.
Performance in CI:
Background and challenges - Andy Still, Co-Founder, Intechnica
An example implementation:
Mark Smith, Online QA Manager, Channel 4
This document discusses test automation, including the purpose of test automation, the test automation process, and the test automation pyramid. The key points are:
1. Test automation aims to improve test efficiency, provide wider test coverage, reduce costs, and speed up testing.
2. The test automation process involves defining the test scope, designing tests, coding tests, setting up the test environment, running tests, and maintaining automation over time.
3. The test automation pyramid illustrates that unit tests should form the base, as they are quick to write and run, while user interface tests are at the top as they are more complex and time-consuming.
4&5.pptx SOFTWARE TESTING UNIT-4 AND UNIT-5hemasubbu08
This document provides an overview of testing in the agile environment and automated testing tools. It discusses agile testing methodologies like test-driven development, acceptance test-driven development, and behavior-driven development. It also covers testing web and mobile applications. The document then focuses on automated testing, describing how it works and the types of tests that are normally automated. It discusses benefits and best practices of automated testing. Finally, it covers popular automated testing tools like Selenium, Katalon Studio, and Cucumber and provides pros and cons of each.
Why we need software testing?
- Find / prevent bug
- No more nightmare when you deploy to production
- Preserve product quality
End to End Test
- Testing the complete functionality of some application
- Real application
Automation testing involves developing and executing tests that can run unattended, comparing actual and expected results. The major objectives of automation testing are to speed up testing, reduce costs and time, and increase quality. Automation testing is faster than manual testing and can test more scenarios. However, high upfront investment is required for tools and training. Both manual and automation testing are needed as automation does not cover all test cases. Popular automation testing tools include Selenium, QTP, and TestComplete.
The document discusses 3 key ways that developing software for the cloud differs from traditional approaches:
1. Incremental delivery, with frequent small releases of new features rather than large periodic releases.
2. Increased automation, including automated testing and continuous integration/deployment pipelines to support more agile development and deployment.
3. Analytics of usage data to inform product decisions and ensure features are valuable to users. Developing with the cloud in mind requires rethinking processes to focus on agility, automation and data-driven insights.
Testing for Logic App Solutions | Integration MondayBizTalk360
In this Integration Monday session, Mike discussed the challenges and approaches for some of the common testing scenarios when delivering integration solutions with Microsoft Azure.
Karishma Kolli – Myth Busters on Test AutomationPractiTest
Slides from Karishma Kolli's Fall OnlineTestConf session – Myth Busters on Test Automation
There is a misconception in software world about automation testing that it is a “Magic wand that takes care of all your testing needs” but in reality it is NOT.
Test automation is at times mistaken to be the be-all and end-all testing solution. Why can’t it take care of all our needs? How and when should I choose automation? What’s my developer’s role in test automation? How much of ROI will I have with test automation? How can I get the best of both worlds(Manual and automation)?
These questions and more were addressed in this OnlineTestConf session.
For recording visit: www.onlinetestconf.com
Automated testing overview discusses the importance of software testing and automated testing. It defines software testing as verifying that software meets requirements and works as expected. The document covers different types of testing and why automated testing is needed to reduce costs, protect reputation, and address difficulties in testing. It provides examples of unit testing simple objects, objects with dependencies, and user interfaces to illustrate how to implement automated tests.
This document proposes an automated test architecture for the UI Builder application using REST API testing, UI testing, and integration testing. It recommends the Rest Assured library for API testing due to its ability to integrate with Jenkins and remove the need for manual HTTP calls. Protractor is proposed for UI testing as it allows testing Angular applications in JavaScript and integrates well with Selenium. Both test frameworks would be set up with their own Git repositories linked to Jenkins jobs. Integration testing would link the API and UI test repos. The test automation aims to provide faster feedback and reduce regressions as the application grows.
This document discusses test automation and continuous integration in agile development. It notes that early automation, continuously integrated and testable software, and fast feedback help ensure faster delivery and optimize business value. The main challenges are short testing time frames per story and multiple deployments per sprint. Building a test suite with unit, service, and UI tests provides faster feedback and easier refactoring. Collaboration between testers and developers and striking a balance between test levels is important. Continuous integration runs automated tests on every code check-in to test feature integration within sprints. Best practices include early QA involvement, defining test scenarios during backlog refinement, estimating automation effort, and creating automation tests in parallel with development. Benefits include faster test cycles within
Continuous integration (CI) implements continuous processes of applying quality control through small, frequent pieces of effort. CI aims to improve software quality and reduce delivery time by replacing traditional quality control after development. CI involves: writing automated developer tests; running private builds where all tests and inspections must pass; committing code frequently; and avoiding getting broken code. It provides feedback through mechanisms like email, IM, and SMS to keep the right people informed at the right time. CI helps reduce code complexity, maintain standards, and assess test coverage through continuous inspection along with compilation, testing and deployment.
Neev uses a scrum based Agile Development methodology, a proven Extended Delivery Center model of engagement - all designed to ensure high quality, timely deliverables.
The document discusses different types of automation frameworks including data driven, keyword driven, modular, functional decomposition, and hybrid frameworks. It provides details on the components, workflow, advantages, and disadvantages of a functional decomposition framework. This type of framework divides an application into logical functions that can be reused across multiple tests. Tests are created by calling a series of functions in the required sequence. The framework supports centralized test data management and flexible execution of tests in different environments.
Nitisak Mooltreesri from DST Worldwide Services spoke about automated load testing for continuous delivery. He discussed how load testing is important to find bugs under high user loads. His company performs daily automated performance tests using simulation approaches to test incomplete systems cheaply and reliably. This helps reduce performance issues by providing early feedback and catching problems before deployment.
Test automation lessons from WebSphere Application ServerRobbie Minshall
The document discusses WebSphere testing at IBM. It provides an overview of IBM's:
- Extensive testing resources including over 200 engineers and thousands of systems.
- Daily regression testing of over 1.7 million tests.
- Transition from waterfall to agile development which reduced cycle times and resources needed for testing.
- Use of cloud resources to speed up test deployment and automation.
- Focus on creating meaningful regressions through techniques like integration acceptance tests run continuously on each build.
3. Agenda
• Background - Why test automation, common
view of test automation?
• Define Problems - Issues most organizations
face with test automation.
• Fixing the above issues with architecture that
has the most ROI.
• Review.
• Q&A.
4. Why Test Automation?
Dev Team’s Perspective
• Increase effectiveness,
efficiency, and coverage of
software testing.
• For regression testing.
• Improve quality.
Boss’s Perspective
• Reduce operational cost.
• Increase revenue.
5. What comes to your mind when
you hear the term “We have
Automated Tests”?
8. Focused on GUI testing
• Hire consultants onshore/offshore with
mantra “Automate everything”.
• Qtp, Watir, Selenium Webdriver etc. use tools
specific to GUI tests.
• Every single small scenario gets added to GUI
automation.
11. When you need the test to run - It
never works.
Please work!
12. So many problems!
• Test execution is slow.
• Tests are fragile.
• Tests are not reusable.
• Tests needs more infrastructure.
• Test works on my box but fails
on daily run.
More time spent on figuring
why the tests are failing.
Resource tied up just reviewing
failures.
Maintenance takes way too
much time.
18. GUI Layer test
• Test the whole application end to end.
• Sensitive to UI changes.
• Does not require good code.
• Debugging application is harder based on failing test.
• Works for Legacy system and with newer software
architecture like SOA as it is independent of backend
technology. e.g. Open a browser, fill in the form and submit
the data.
19. API/Service Layer/Integration Test
• Test the application logic by making service calls.
• Faster than UI testing.
• Requires service to exist.
• E.g. Instantiate the calculator service and get it to
add two numbers.
20. Unit level test
• Test individual classes.
• Much faster than service level testing.
• Requires good design.
21. Overview of layers of test pyramid
Application UI
Tightly Coupled
Automated UI tests
Service Layer
Automated One Level
Below UI
DAO Layer
Fragile
Lag with Current Development
Robust & Stable
With Current Dev cycle
Fast Execution
UI Independent
23. Start with Unit Test
(Quality is not just QA’s responsibility)
• Isolation.
• Test small piece of code - preferably separate test for
each method.
• Biggest test should be one test for a class.
• Use mock services to remove dependency.
• Test only public endpoints.
• This forces to use good design.
24. Invest more in Service Layer Test
• Independent functional test.
• Use BDD tools like Cucumber, SpecFlow,
Jbehave.
• Use mocks but also have specific tests to
check third party calls.
• Add data validation tests.
• Tests for boundary condition.
• Tests for error handling - wrong data
type, missing data.
• Tests around calculations, business rules
etc.
Gherkin/Plain test
DSL(Domain Specific
Language)
Set up
projects/helpers,
mock
Framework
Nunit,Junit,Rspec,
xunit
GUI
Service Layer
25. Limit the Number of GUI test
• Ask if it can be tested in Service layer before starting.
• Start out with use and throw style test.
• Gradually build the framework.
• Use GUI test as an assistance for manual testing.
• Create fewer tests but with broader end to end
coverage.
26. Quick Guide to Building GUI Test
Framework
• Do basic meta - programming and
start automation. Don’t design
the entire framework right away.
• Re - factor often.
• Review test automation
architecture regularly.
• Run daily.
Gherkin/Plain test
Framework
Watir/Selenium
Webdriver/QTP
WorkFlows
Pages
Navigation
UI
Utilities
27. GUI
Service/Api
Layer/Integration
Unit Test
Unit Test
Business Logic
Data access
layer
Mocked Data
access layer
Mocked
Database
Unit tests that isolate
dependencies with fakes,
mocks, stubs etc
Database
Can use just unit test
framework like Nunit,
Xunit prefer BDD style
like Cucumber, specflow ,
Jbehave etc
End to end UI
test with tools
like Watir,
Selenium, QTP
etc.
Automated UI
Test
UI
Unit Test
Integration
Test
Integration
Test
Integration
Test
Business Logic Business Logic
Data access
layer
Data access
layer
Data access
layer
Mocked
Database
Database
Database
Data access
layer
Business Logic
30. Cross-Functional Team Involvement
• Request BA’s and product owners to
review your automated gherkin
often.
• Time box and rotate maintenance
tasks with developers and manual
testers.
• Pair up with manual testers
when writing different scenarios..
31. Review
• Prioritize Unit and API/Service test
automation.
• Limit the number of GUI tests, have
a fixed number of stable end-to-end
tests.
• It’s ok to use multiple technologies.
• It’s ok to use Record/Play tools like
Selenium IDE for throwaway tests.
• It is better to have 50 stable test
cases than 500 fragile ones that
breaks regularly.
• Cross functional team involvement
GUI test very important to catch system level test
Everything seems fine so far, you wanted to automate, you hired people and started automating, so what's the
problem ?
We forget or ignore fundamental
Investing just on GUI test, instant gratification
Defined by Mike Cohn in succeeding with agile. During the process of manufacturing a ballpoint pen, the cap, the body, the tail and clip, the ink cartridge and the ballpoint are produced separately and unit tested separately. When two or more units are ready, they are assembled and Integration Testing is performed. For example, whether the cap fits into the body or not.
Lets go ahead and define each layer. GUI test very important to catch system level test. UI tests are very brittle, so keep these tests to a minimum. These automation tests will need maintenance any time the UI changes, and because there are so many factors that come into play when you run a test that emulates clicks on a screen (such as network speed), such tests can result in false test failures. You can't ignore those test failures, but you don't want to end up spending more time troubleshooting and maintaining UI tests than you spend finding actual code defect
This provides many advantage of end to end testing without issue of UI tests. The service layer allows for testing business logic at the API or service level, where you're not encumbered by the user interface (UI).
Summary of above definition
This provides many advantage of end to end testing without issue of UI tests. The service layer allows for testing business logic at the API or service level, where you're not encumbered by the user interface (UI). DSL for setting values as per business criteria
Your focus now at the UI level is simply to ensure that the UI itself is working correctly. UI tests are very brittle, so keep these tests to a minimum. These automation tests will need maintenance any time the UI changes, and because there are so many factors that come into play when you run a test that emulates clicks on a screen (such as network speed), such tests can result in false test failures. You can't ignore those test failures, but you don't want to end up spending more time troubleshooting and maintaining UI tests than you spend finding actual code defect
