A Method for Discovering Key Test Functions of Large Object-Oriented Software Systems

Abstract

The invention discloses a method for discovering key test functions of a large-scale object-oriented software system. The invention can quickly and reasonably discover key test functions of large-scale object-oriented software systems with high accuracy. The present invention first draws the class collaboration network diagram of the software with classes as nodes and function calls between classes as arcs, and defines the importance of an arc as the number of arcs that can be traversed starting with the arc; then calculates The importance of each arc in the class collaboration network diagram, the arc whose arc head is the class node with the largest total number of reachable arcs is the key arc; finally, the arc head and arc tail information of the key arc are located in the software system For the corresponding two classes, the function call relationship that constitutes the coupling relationship between the two classes is the key test function. The invention overcomes two defects in the existing method, and the calculation complexity does not increase with the complexity of the software system, and can quickly and effectively find the key test function of the software system.

Description

technical field [0001] The invention relates to the technical field of software testing, in particular to a method for discovering key testing functions of large-scale object-oriented software systems. Background technique [0002] From the perspective of graph theory, a complex network is a graph with complex topological properties, and its typical characteristics are scale-free distribution with index characteristics and small-world effect. In a complex network system, there are a small number of key nodes that play a key role in controlling the entire system. The damage to these nodes often leads to the impact of large-scale nodes or even the paralysis of the entire system; while most of the nodes in the system are controlled by key nodes. Nodes that themselves have little influence on other parts of the system, and attacks on these nodes will not cause obvious failures in the system. Therefore, compared with random networks and regular networks, systems with complex net...

AI Technical Summary

Problems solved by technology

[0011] Therefore, there is a need for a fast and practical discovery method, especially when the scale of the software network is huge, if the time complexity of the method increases significantly with the number of software nodes, it will be of little practical value; the key test functions discovered by this method should be able to The accuracy truly reflects the functions that need to be maintained in the software system

Moreover, this method should weaken or rectify the deficiencies of the existing methods (using the flow betweenness of the arc as a standard to measure the key functions in the software): 1. Function abnormal information is likely to propagate along multiple paths rather than just the shortest path ;2. The propagation path of the exception is inaccurate (the exception passed into the class node by the arc entrance will not be passed out along some exits of the class)

Images