Source code change influence analyzing and testing method for software units and components with combination of phase

Abstract

The invention discloses a source code change influence analyzing and testing method for software units and components with combination of phase. The analysis and test on unit grade source code change influence comprise the following steps of: constructing a program unit flow graph; establishing a linear independent test route set; checking a node of a modified code in the flow graph and determining the node as a change point; using a route selector to select a route containing the change point from the linear independent test route set so as to form a influenced route set; executing the route in the influenced route set; and comparing the execution result with expected results. The analysis and test on component grade source code change influence comprise the following steps of: establishing a program scheduling graph, wherein the scheduling graph uses a scheduling structure of nodes and boundary description programs; establishing a linear independent test sub-tree set; confirming a unit at which the modified code is located and taking the unit as the change point; using a sub-tree selector to select a sub-tree from the linear independent test sub-tree set which contains the change point so as to form an influenced sub-tree set; executing the sub-tree in the influenced sub-tree set; and comparing the execution result with expected results.

Description

technical field [0001] The present invention relates to the field of program development and testing. More specifically, the present invention relates to a method for analyzing the impact of source code changes. In addition, the present invention also relates to a method for implementing tests after adopting the source code change impact analysis. Background technique [0002] The development based on the existing system source code needs to modify a large number of software codes. The software debugging work needs to modify the code at any time, and the software maintenance and upgrading process also needs to modify the code. Code changes are always accompanied by software research and development activities. more and more important. Code changes usually produce "side effects", which may affect the internal and external interfaces, functional characteristics, and performance of the software. If the impact of code changes is not fully analyzed and tested, it may lead to Exc...

AI Technical Summary

Problems solved by technology

[0007] It is very difficult to analyze and test the impact of code changes, and there is currently a lack of systematic methods to guide

Traditional change impact analysis and testing methods rely on the coder's experience and familiarity with the code. Manually analyzing the code blocks or execution paths affected by the change makes it difficult to comprehensively evaluate the impact of the change, and it is also difficult to fully evaluate the code. test

The effect of the analysis method based on experience is related to the personal ability of the coding and debugging personnel, which has a large randomness and blindness, and there is no automatic tool support, which is inefficient and cannot provide a systematic solution

As for the research on the impact analysis of object-oriented code changes, the target of the analysis is the class, which is limited to object-oriented codes, and the change impact analysis cannot be performed on non-object-oriented codes, and this method does not carry out the test process and method after the change impact analysis. illustrate

Images