Embedded software intelligent testing method based on Agent

Abstract

The invention relates to an embedded software intelligent testing method based on Agent, which is an improved software testing method using a Linux system as an embedded operation system core, mainly comprising three Agent modules: a pre-processing Agent module, a program instrumentation Agent module and a dynamic test information analysis Agent module. The pre-processing Agent module comprises a lexical analysis Agent and a grammar analysis Agent that are used for generating a grammar tree structure of a source program; the program instrumentation Agent module comprises a grammar tree analysis Agent and an automatic instrumentation Agent that are mainly used for analyzing the grammar tree, finding out the key point of the source program, and automatically inserting probe functions in the source program so as to acquire the dynamic information during the operation of the program; and the dynamic test information analysis Agent module comprises a cover ratio analysis Agent and an inquiry Agent that are mainly used for testing the sentence and the branch cover ratio of the source program, and automatically inquiring the relevant conditions of the uncovered target in the source program when the cover ratio cannot satisfy the requirements. The invention can be applied to different programs simply and conveniently, thereby greatly improving the efficiency of the software automation test with strong commonality.

Description

technical field [0001] The invention relates to a software testing method, in particular to an Agent-based embedded software intelligent testing method. Background technique [0002] As an important branch of computer software, embedded software has been widely used in consumer electronics, transportation systems, industrial control, medical equipment and military fields. As a very popular operating system type at present, Linux is widely used in the embedded operating system kernel. With the development of embedded technology, the continuous growth of embedded applications and the continuous improvement of the complexity of embedded systems, the scale and complexity of embedded software are also required to continue to increase. The quality of embedded software and the development cycle have a great impact on the final quality of products. and time-to-market play a decisive role. [0003] Software testing is an important means to ensure the correctness of software systems...

AI Technical Summary

Problems solved by technology

[0006] (1) The software reliability evaluation is only an overall concept, and the number of software defects is predicted, but the location of the defects cannot be specifically pointed out;

[0007] (2) The effect of the hidden code error test is not good, and it will cost a lot of manpower and material resources to test again;

[0008] (3) There are too many steps of human participation, which is very inconvenient to use;

[0009] (4) The obtained coverage index is relatively simple and cannot fully reflect the quality of the software

[0010] (5) Most testing software is expensive and difficult to popularize on a large scale

Images