Api test automation of search functionality with artificial intelligence
Mysiuk R,Yuzevych V,Mysiuk I,,,
DOI: https://doi.org/10.15407/jai2022.01.269
IF: 14.4
2022-06-20
Artificial Intelligence
Abstract:One of the steps in software development is to test the software product. With the development of technology, the testing process has improved to automated testing, which reduces the impact of the human factor on error and speeds up testing. The main software products for testing are considered to be web applications, web services, mobile applications and performance testing. According to the testing pyramid, when testing web services, you need to develop more test cases than when testing a web application. Because automation involves writing software code for testing, the use of ready-made tools will speed up the software development process. One of the most important test indicators is the coverage of search functionality. The search functionality of a web application or web service requires a large number of cases, as you need to provide many conditions for its operation through the free entry of any information on the web page. There is an approach to data-based testing, which involves working with a test data set through files such as CSV, XLS, JSON, XML and others. However, finding input for testing takes a lot of time when creating test cases and automated test scenarios. It is proposed to use artificial data set generators based on real values and popular queries on the website to form a test data set. In addition, it is possible to take into account the probable techniques of developing test cases. It is proposed to conditionally divide the software for testing into several layers: client, test, work with data, checks and reports. The Java programming language has a number of libraries for working at each of these levels. It is proposed to use Rest Assured as a Restful client, TestNG as a library for writing tests with checks, and Allure report for generating reports. It is noted that the proposed approach uses artificial intelligence for automated selection of test cases when creating a test to diversify test approaches and simulate human input and behavior to maximize the use of cases.
computer science, artificial intelligence