Once you learn to test your code, another thing you have to worry about is testing it as efficiently as possible. You can just write tests and hope that it ensures your code works effortlessly, but it may not always catch everything that can go wrong. Adopting some strategies and methods will be important in helping you determine how to go about testing your and managing that. 

There are a number of strategies that you can use and adapt. Early on in class, we went over some strategies very quickly that can be used or serve as a good base. Some of these consisted of black box and white box testing, dynamic and static testing, and some others. Each has their own way of testing, some can be better than others. Ultimately, it’s up to you to determine and figure out which one best suits you, the way you want to do things, and your code.

In this blog post, Janki Mehta writes about testing strategies, its key components, approaches, and some specific ones. They describe testing strategies as plans of “how software testing will be approached, executed, and managed throughout the software development life cycle.” They then describe the key components of the strategies, which are scope and overview, testing methodology, test environment specification, release control, risk analysis, testing tools, and review and approval. They dive into detail with the two approaches, proactive, where it focuses on detecting and catching bugs and defects before they even occur, and reactive, where it focuses on catching them afterwards. Some strategies they list were black box and white box testing, but some others too, like integration testing and regression testing, also describing what each of them are. At the end of the article, there are some final tips that can help easing the testing process and creating a comfortable environment when it comes to testing.

Knowing good strategies is important for testing code, and knowing multiple ones is even more important. It allows you to look around and figure out which fits best for your code and software, and adapt to it. Testing code can and probably will be difficult to get right all the time, learning all you can and having additional resources will be beneficial. I think that Mehta wrote a good article, it covers a lot of material, some of it can be seen as extra but it doesn’t hurt to know extra information, even if you can’t use it now. It’s clear the article is meant for a team to use as reference, but anyone can use it, not just a team. 

From the blog CS@Worcester – Cao's Thoughts by antcao and used with permission of the author. All other rights reserved by the author.

After exploring more of testing, and different strategies, I wondered about how testing could be automated. I came across a blog for Postman, on the TestProject website explaining Postman’s use in automated API testing. This blog is focused on providing a tutorial to understand how to use Postman to automate testing, rather than understanding various testing practices as seen in my previous blogs. The article’s title is Getting Started with using Postman for API Testing by Dave Westerveld

The blog first defines what an API is, as Postman is specifically for API testing. The blog defines API as Application Programming Interfaces. These APIs are extremely important in facilitating communication between software systems. Testing them is paramount for this reason, to assure they are working properly. This is where Postman comes in. It is an automated API tester, with a fully fledged application download. The article outlines the setup of Postman, and explains how to get it to cooperate with your system: First open the app of Postman that you’ve downloaded and then sign in. After this you are allowed to make API requests, the new button is then used and an API request is created. You name the request, and create a collection, where the requests are stored. Following setup, the article explains how to execute a Postman request. During setup, the article created the collection name JsonPlaceHolder, and then used that collection as the URL for the Request URL. After this, Postman will send the request for you. This can be used to compare results from different requests of an API and assure they are the same, although some manual input IS required in automated testing. However more intricate API Testing is shown in further tutorials with Postman.

This is a topic that interested me greatly, as I have been given many suggestions to look into API testing. Through the CS-443 course, I had been looking for more options to test and explore coding. Automated testing is something immensely interesting, and I appreciate the comprehensive guide provided by TestProject for Postman. This is of great value to me as I will continue to look through this guide and potentially post more blogs on this topic, as I explore automated testing. This first blog is mainly involving setup, and the basic concept of requests, but I look forward to further automated API testing using Postman, as these articles are very well structured.

Source:
https://blog.testproject.io/2020/07/15/getting-started-with-using-postman-for-api-testing/

From the blog CS@Worcester – WSU CS Blog: Ben Gelineau by Ben Gelineau and used with permission of the author. All other rights reserved by the author.

Black box vs. White box vs. Gray box Testing

Black box testing

Black box testing focuses on the behavior of the software on a surface level. This means the tester cannot see any technical details such as code or the structure. Testers are not concerned about the inner workings of the software, rather only on the inputs and outputs. Therefore black box testing comes from the perspective of the user which allows testers to uncover any bugs and/or design flaws that would directly affect the user’s experience.

Some black box testing procedures include boundary value analysis, equivalence partitioning, and decision table testing. Boundary value analysis tests whether the software results in the correct output when the input contains the lower or upper limit of the input range. Equivalence partitioning groups all possible inputs into categories, which can reduce the number of total test cases. Finally, decision table testing involves grouping inputs into a table and testing how the software behaves when those inputs are combined.

White box testing

White box testing is the opposite of black box, meaning testers are given the internals of the system such as the code, the structure, and how its implemented. White box testing allows testers to fully understand how the software is executed. Having access to the internals of the software can help testers find issues related to security vulnerabilities, broken data paths, etc. while black box testing could not focus on those type of issues.

White box testing can be completed with a few techniques. These techniques include statement, branch, and path coverage. Statement coverage ensures that every statement in the code is run and tested at least once. Branch coverage executes all possible branches (i.e. conditional branches) in the code to be tested. Path coverage uses test cases to cover all possible execution paths throughout the software.

Gray box testing

Gray box testing combines black and white box testing into one. Therefore testing can be done from the perspective of the user, while still having access to the software’s internal code. Gray box testing is helpful when identifying problems that may otherwise be hard to find with other types of testing.

A couple techniques that are used in gray box testing are Matrix and Regression testing. Matrix testing looks at all variables in the code and assesses its risk level. Doing this can identify unused/under-optimized variables. Regression testing checks that software changes or bug fixes do not create new errors.

Reflection

This article was chosen because it clearly described the different types of box testing, and what each does. While reading the material, I realized that what we have done in class (equivalence class, boundary value testing) are actually black box testing techniques. I enjoyed learning about the different techniques that are used, and the benefits of them. Although developers and testers are two different roles, developers sometimes play the role as the end user when black box testing, so having a basic understanding of the different types of box testing will be helpful in the future.

References:

https://www.shakebugs.com/blog/black-vs-white-vs-grey-box-testing

From the blog CS@Worcester – Zack's CS Blog by ztram1 and used with permission of the author. All other rights reserved by the author.

As my journey to becoming a full stack developer continues, I have begun to encounter the need for front-end testing. When finding an article, I realized that I did not have very much knowledge on what front-end testing entailed or how to implement it into my coding. I found the blog “Front End Testing: A Comprehensive Overview” by Kiruthika Devaraj (https://testsigma.com/blog/front-end-testing/) on my search for information on this topic.

The blog starts off by explaining that front-end testing is focused on the user’s experience. While past experiences have taught me that back-end testing specifically tests the functionality of the code, front-end, also known as the user-facing end of a website, testing has a higher focus on the user’s interactions. The author also delves into eight different types of front-end testing and the elements that make up these different types:

• Unit Testing is the analysis and testing of the individual components to ensure they each work as intended. 
• Acceptance Testing involves testing to ensure permissions, such as account accesses, are working properly. 
• Visual Regression Testing tests for visual changes in the front-end by comparing a reference picture to a baseline image. This can typically create brittle tests that can fail due to the slightest changes in an element’s location on a page.
• Performance Testing measures an applications stability and responsiveness under different levels of simulated stress and traffic.
• Accessibility Testing ensures that individuals with visual impairments or other additional needs can still use and access the application to the fullest of their capabilities.
• End-to-End (E2E) Testing involves testing the application from start to finish while testing all components and systems for how they work together.
• Integration Testing is when each individual module is tested to ensure it has been implemented correctly.
• Cross-Browser Testing tests that the application works correctly across multiple browsers and confirms compatibility.

Devarj also provides tips for better front-end testing, the first tip is to always use a testing framework. These frameworks, also called front-end tools, are what actually run the tests. Although countless front-end tools exist, the blog provides information to 3 popular options; Testsigma, Selenium, and Ranorex. Testsigma and Selenium are both open-source front-end testing frameworks that may be worth investigating further, while Ranorexis a commercial framework that may be more than what I am currently looking for. 

Front-End Testing Moving Forward

As I move to implement automated front-end testing into my projects, it is important to remember that the front-end testing is focused on ensuring the project is user-friendly, responsive, and visually appealing by testing the layout, design, and functionality across multiple browsers. Doing so will provide users with the best possible experience.

From the blog CS@Worcester – CS Learning by kbourassa18 and used with permission of the author. All other rights reserved by the author.

Exploring the Edges: A Deep Dive into Boundary Value Testing

In the realm of software testing, Boundary Value Testing (BVT) emerges as a cornerstone, spotlighting the crucial junctures at the extremities of software modules. This method, celebrated for its strategic focus, eschews the intricacies of internal code to scrutinize the behavior at pivotal points: the minimum, maximum, and nominal value inputs. Its essence captures the high propensity for errors lurking at the fringes, thereby underscoring its indispensability, particularly for modules sprawling with vast input ranges.

BVT transcends mere error detection; it embodies efficiency. By tailoring test cases to boundary conditions — including the crucial spots just above and below the extremes — it ensures a comprehensive examination without the exhaustive effort of covering every conceivable input. This approach not only conserves resources but also sharpens the focus on those areas most likely to be fraught with defects.

The methodology behind BVT is meticulous yet straightforward. It commences with the creation of equivalence partitions, a critical step that segments input data into logically similar groups. Within this framework, the boundaries are meticulously identified, crafting a testing landscape where both valid and invalid inputs undergo rigorous scrutiny. The beauty of BVT lies in its precision — it’s a targeted strike on the software’s most vulnerable fronts.

However, the path of BVT is not devoid of obstacles. Its lens, sharply focused on boundaries, may overlook the vast terrain of non-boundary inputs, a limitation particularly pronounced in Boolean contexts where only two states exist. Moreover, its efficacy is inextricably linked to the adeptness in defining equivalence partitions — a misstep here can skew the entire testing trajectory, transforming a potential asset into a liability.

Yet, the allure of BVT remains undiminished. In the high-stakes arena of software development, it stands as a sentinel at the gates of functionality, guarding against the incursion of boundary-related errors. Its role is pivotal, especially when the clock ticks against the backdrop of tight deadlines and expansive test fields.

BVT is more than a testing technique; it’s a strategic imperative. As software landscapes continue to evolve, the demand for precision, efficiency, and effectiveness in testing methodologies will only escalate. BVT, with its focused approach and proven efficacy, is poised to meet this challenge, proving itself as an invaluable ally in the quest for flawless software. In the hands of skilled testers, armed with a deep understanding and meticulous execution, it transforms from a mere method into a beacon of quality assurance. Boundary Value Testing (BVT) is a key black box testing method targeting software’s extreme limits to identify errors without delving into internal code, focusing on boundary conditions for efficient and effective error detection. It requires meticulous planning and execution, especially in defining equivalence partitions, to ensure testing success.

I read the blog on this link:
https://testgrid.io/blog/boundary-value-testing-explained/

From the blog CS@Worcester – Coding by asejdi and used with permission of the author. All other rights reserved by the author.

https://testsigma.com/blog/levels-of-testing/#:~:text=The%20most%20common%20types%20of,system%20testing%2C%20and%20acceptance%20testing.

Different levels of testing software is very important when it comes to quality assurance and each of the different levels is based on different combinations of things that help ensure the best conditions and functionality of your software. The four different levels of testing are as follows: “Unit Test” which tests an individual component, “Integration Test” which tests an integrated component, “system Test” which tests the entire system and “acceptance Test” which tests the final system otherwise known as the final product.

Unit Testing is used to outline the expectations of a program or outline its functionality which in essence makes sure that certain parts of the program function as intended. Integration testing allows you to test groups of functions to make sure that they interact properly with each other which is a step above unit testing. System testing is similar to integration testing as it tests a ‘group’ of functions but this group is now comprised of every part of the program/application. Acceptance testing then tests all functional and non functional aspects of the program in order to ensure proper functionality, security, etc.

I selected the levels of testing as a topic as it directly relates to the types of testing we have been doing in class. We have been testing functionality of different bodies of code so far during this semester and we have used different techniques including but not limited to equivalence class testing, edge case testing, worst case testing, boundary value testing, etc. I would say that most if not all of these testing types would fall under the unit testing level of software testing as we have for the most part testing individual system components being able to learn about the different levels of testing makes me excited to continue on through the levels of testing.

This source was very helpful when it came to learning about the other levels of testing we have not used during class yet and I would recommend this article to any of my fellow students although this article does not go extremely in depth when it comes to the different levels of testing it does give you a good general understanding of each level so that you can go on to learn more without getting confused between the different levels of testing. The diagrams and explanations of the software sequence are valuable resources for a software testing course.

From the blog CS@Worcester – Dylan Brown Computer Science by dylanbrowncs and used with permission of the author. All other rights reserved by the author.

Boundary testing is a technique that “involves evaluating data based on boundary values… such as minimums and maximums.” It involves testing the lower and upper limits of a domain of inputs, as well as values just inside and outside those limits. This technique helps improve software quality by detecting defects early in the development process, which ultimately saves time and resources. This is due to the fact that “the number of defects at the boundaries of an input domain tends to be higher than in other areas of the domain.” 

Going back to how it saves time and resources, it helps developers choose more efficient test cases, which cuts down on time and how much they have to do, especially with large sets of data and number of inputs to consider. Furthermore, this technique “allows for a lot of uniformity and the test cases are much smaller” meaning that it can be automated. This would save more time, possibly result in less errors than if a person were to write the test cases, and allow for developers to put their focus on other tests or somewhere else, resulting in increased productivity. 

Boundary testing involves determining the input value range, which is whatever values the system accepts. Values inside that range are valid while those outside are invalid. Test cases are then created based on the values at the extreme ends, end values, and values just before and after the end values. For example, with a range of 0 to 100, we test values less than 0 (-50), greater than 100 (150), values around and at 0 and 100 (-1, 1, 99, 101). 

Another example the article shares is one that allows users to create a password. The password has to be 5 digits meaning that the range is 10000 at minimum and 99999 at maximum. Any combination of numbers is valid as long as it is 5 digits long, meaning that 4 digit long and 6 digit long passwords are outside the range and are invalid. 

Boundary testing seems like a very useful testing technique with very little downsides. It’s a black-box testing technique which usually means that testers don’t need extensive knowledge of coding, computer science, or the system at hand. It also seems very easy to do as long as all the necessary knowledge for testing is provided. Seeing as how it can also be easily automated, it may be more efficient to focus on more complex or time-consuming testing techniques while this test is running in the background or on the backburner.

https://www.indeed.com/career-advice/career-development/what-is-boundary-testing

From the blog CS@Worcester – Kyler's Blog by kylerlai and used with permission of the author. All other rights reserved by the author.

Software testing is a crucial aspect of the software development lifecycle. Among various testing techniques, Boundary Value Testing and Equivalence Class Testing stand out for their effectiveness in identifying defects early in the development process. In this blog post, we delve into these two techniques and discuss their implementation using JUnit, a popular Java testing framework.

Boundary Value Testing: Boundary Value Testing focuses on testing the boundaries of input ranges. The rationale behind this technique is that bugs often lurk around the edges of acceptable input values. By testing values at the boundaries, we increase the likelihood of uncovering potential defects. For example, if a function accepts input within the range of 1 to 100, we would test values such as 0, 1, 100, and 101.

In JUnit, implementing Boundary Value Testing involves writing test cases that specifically target boundary values. By using assertions to verify the behavior of the function at these critical points, developers can gain confidence in the robustness of their code.

Equivalence Class Testing: Equivalence Class Testing aims to reduce redundancy in test cases by partitioning the input domain into equivalence classes. Each equivalence class represents a set of input values that should produce the same output when processed by the function under test. By selecting representative values from each equivalence class, testers can ensure adequate coverage without testing every possible input value.

In JUnit, implementing Equivalence Class Testing involves creating test cases that cover each equivalence class. Test inputs are chosen strategically to represent the entire range of possible inputs within each class. This approach not only improves test coverage but also makes test suites more manageable and maintainable.

Combining Boundary Value Testing and Equivalence Class Testing: While both techniques offer unique benefits, they are most effective when used together. By combining Boundary Value Testing to test the edges of input ranges and Equivalence Class Testing to cover representative values within each range, testers can achieve thorough test coverage with minimal redundancy.

Boundary Value Testing and Equivalence Class Testing are powerful techniques for improving the quality of software. By leveraging JUnit, developers can easily implement these techniques within their Java projects. By understanding the principles behind these testing strategies and applying them effectively, teams can build more robust and reliable software products.

Incorporating these testing techniques into your development workflow can help catch bugs early, reduce the risk of defects slipping into production, and ultimately enhance the overall quality of your software.

Link: JUnit 5 – User -guide/

From the blog Discoveries in CS world by mgl1990 and used with permission of the author. All other rights reserved by the author.

While looking for more articles online involving testing, I came across SauceLabs. This website is another compendium of blogs, similar to stickyminds mentioned in my last blog. The layout and articles on this website were an excellent selection of resources dedicated to testing and quality assurance. While running through many bugs in my most recent projects and assignments, I was searching for an article dedicated to programming bugs. I found an article that fit my needs, dedicated to the differences between severity and priority of bugs, which is something I had not previously considered. The article’s title is Understanding Bug Severity vs. Priority: Key Differences and Best Practices by Chris Tozzi

The article provides (as the title indicates) a breakdown of bug severity and bug priority in software testing. It explains that bug severity is primarily concerned with the impact of a bug on the system’s functionality. Issues are categorized into the following levels of severity: critical, major, minor, and trivial. Each of these have a different level of impact on the system. For example, critical bugs may cause system crashes, while trivial bugs have a small functionality and performance impact. Bug priority focuses on the order in which bugs should be addressed, considering various factors that will impact the job as well as the program, such as deadlines, or bugs that are caused by higher priority bugs. Bugs with high priority require immediate attention, and those with low priority can be addressed later as the program develops and is tested further. It is an extremely important skill to be able to identify the severity and priority of bugs throughout the system and its development, so as to facilitate a smoother and more efficient development cycle. The article also provides some examples, such as a critical severity bug that causes data loss. This bug would ALSO have high priority because it is causing a lot of damage to the system’s structure.The article offers very useful recommendations for bug reports such as: establishing clear classification guidelines for bugs (priority/severity levels), communicating efficiently within teams, or reassessing bug statuses frequently.

I chose this blog because of my most recent experience in my assignments and projects. I had found that many bugs relied on each other, and I had never previously taken the initiative to learn more about bugs. When working with a team in industry, I will now have an understanding of how bugs are classified, as well as how to manage bug reports. This is something I had very little experience with prior to this article, and I will make sure to put extra effort into bug reports in my own projects with peers. I will continue to look into articles about bugs, testing, and QA to further improve my knowledge.

Source:
https://saucelabs.com/resources/blog/bug-severity-vs-priority

From the blog CS@Worcester – WSU CS Blog: Ben Gelineau by Ben Gelineau and used with permission of the author. All other rights reserved by the author.