Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

A system to efficiently identify test cases per code submission

IP.com Disclosure Number: IPCOM000232229D
Publication Date: 2013-Oct-27
Document File: 4 page(s) / 73K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a system to improve test efficiency by identifying a list of automated testcases that hit the code that is being changed. The approach uses code coverage in parallel with the development process, and converts the coverage output to a consumable form and automatically passes it to an automation runtime environment for execution.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 49% of the total text.

Page 01 of 4

A system to efficiently identify test cases per code submission

A variety of coverage metrics and criteria can be measured in a software program.

Some of the metrics include the path coverage, condition coverage, function coverage, statement coverage, etc. within the program. A method is needed to enable the use of

this information on a daily basis in order to reduce the cost of testing.

In the current software process, after the developer submits the code for testing, the test cases to be executed are selected based on evaluation of the changes or functional area. Test engineers are dependent on developers to have full understanding of the associated functional area as well as all surrounding areas to help identify the impact of the proposed changes. After an interaction between the

developer and test engineer, a list of test cases is developed based on domain knowledge and expert judgment. This might lead to having test cases that are not directly related to exact modified code or that might miss the related ones. During this process, the side effects of the code submission on some other functional areas cannot be ascertained. Thus, test selection is an inexact science having many loopholes. This has a direct effect on the quality of the final product. This process also increases the submission cost.

No current approaches answer the question, "How can an organization plan tests to target code change?" Current trial and error processes are lengthy and costly, and cannot be carried out in frequent intervals. In the end, the set of changes suspected for failing test cases does not point to a particular submission, but only provides multiple suspected changes introduced over a period of time. Such methods cannot be incorporated into any automated system during the software development process.

The proposed solution is a system to improve test efficiency by identifying a list of automated testcases that hit the code that is being changed. This suggests the use of code coverage in parallel with the development process. Some available tools take in the list of functions as input and find out the test coverage for the same. However, current approaches provide no consumable output that can be passed directly into the testing cycles. The novel system converts the coverage output to a consumable form and automatically passes it to an automation runtime environment for execution. This allows automatic testcase execution without consuming an excess of resources.

The system performs as an end-to-end tool that takes a code change, queries code coverage data to provide a list of testcases that cover the same code as a given code submission, reserves a VMware* image for automated testcase execution, and attaches the results of this test execution to a fix record that contains the original code submission.

This system reduces the resource costs generated by the creation and execution of regression tests for every code submission. Automation of t...