Method for automatic classification of regression testing case

Abstract

The invention provides a method for automatic classification of a regression testing case. The method comprises the steps that (1) a testing case subset which is influenced by software changes is selected from an original testing case set based on control flow analysis technologies; (2) the testing case subset which is influenced by software changes and the regression testing case to be classified are run on a historical software version and a current software version respectively, branch covering vectors are constructed, and an Euclidean distance function is used to compute the distance between the testing case to be classified and the testing case subset influenced by the software changes; and (3) according to the distance information, a K nearest neighbor algorithm based on weights is used to predict the category of the regression testing case to be classified. According to the invention, the problem that automation efficiency is not high during classification of the regression testing case based on testing prediction generated from manual operations under current regression testing backgrounds can be solved. The method does not need to construct the testing prediction and can greatly increase the automation degree and efficiency of regression testing.

Description

A Method for Automatic Classification of Regression Test Cases The invention belongs to the technical field of software testing, and in particular relates to a method for automatically classifying regression test cases, which is used to reduce the energy consumption of regression testing and improve the automation efficiency of regression testing. Background technique Software testing is an important activity to ensure software quality and reliability. Due to reasons such as adaptive maintenance, complete maintenance and corrective maintenance, software is in dynamic evolution rather than static existence. Once the software changes, regression testing (Regression testing) is required to test the software to ensure that the new changes have no side effects on the software. Regression testing has been proven to be one of the most expensive activities in the process of software development and maintenance, accounting for 50% of the total cost of software maintenance and 80% of...

AI Technical Summary

Problems solved by technology

With the continuous popularization of agile development methods, software regression testing has become more frequent, and regression testing needs to consume more testing resources

This makes the traditional regression test case classification method less and less able to meet the needs of actual software regression testing.

Images