Method and system for processing loop codes in software testing

Abstract

The invention discloses a method for processing loop codes in software test. The method includes: searching loop bodies of a code under test according to a program control flow graph; determining a loop body prepath of a current loop body; determining all single paths of the current loop body, combining the single paths to a combined path, and when an assembled path of the loop body prepath and the combined path is reachable, determining the combined path as a determined path of the current loop body; when an assembled path of the loop body prepath and the determined path is determined to be reachable, generating a test case; executing the code under test according to the test case, and capturing an actual execution path of the current loop body; and repeating the steps until all the loop bodies of the code under test are processed. The invention further discloses a method for processing the loop codes in software test. The method and the system are high in flexibility and capable of saving labor cost and avoiding path explosion.

Description

technical field [0001] The invention relates to software testing technology, in particular to a processing method and system for loop codes in software testing. Background technique [0002] Software testing is the process of analyzing or executing the source program (the code under test) in a controlled environment. quality. The accuracy of software testing is closely related to the comprehensiveness of test path generation. [0003] In the prior art, a test path is generated by traversing the nodes in the program control flow graph, and the source program is executed along the test path using test cases. When the nodes in the program control flow graph are all sequential nodes, that is, the source program is all sequential statements, a clear test path can be generated; but when there are branch nodes in the program control flow graph, that is, there are loop statements in the source program, now There are conventional processing methods such as 0-1 strategy or 0-K stra...

AI Technical Summary

Problems solved by technology

When the nodes in the program control flow graph are all sequential nodes, that is, the source program is all sequential statements, a clear test path can be generated; but when there are branch nodes in the program control flow graph, that is, there are loop statements in the source program, now There are conventional processing methods such as 0-1 strategy or 0-K strategy. The loop statement is treated as a judgment statement, which invisibly misses the generation of some test paths. The incomplete generation of test paths reduces the accuracy of software testing.

In addition, loop statements can also be processed through manual methods such as manually specifying test paths, which increases labor costs; moreover, manual methods have poor flexibility and are difficult to adapt to the needs of different engineering projects

Images