Android automated testing method based on particle swarm optimization algorithm

Abstract

The invention discloses a particle swarm optimization algorithm (PSO) based Android automatic testing method. Triggering of an application control can be carried out in a modularized manner; relatively efficient input test data can be generated; and software exception can be relatively efficiently triggered when convenience is achieved. The method comprises the following steps: (1) mounting an app on a simulator, starting view server service by a viewer module, and obtaining control hierarchy tree information of the app through a hierarchy viewer; (2) with the obtained control hierarchy tree information as input, organizing a tree structure convenient to traverse by a build tree method of a PSO module; (3) traversing a control hierarchy tree constructed by the build tree, and searching whether a historical test case exists or not in a test case module after traversing a node every time; and (4) calling a test case set stored in the test case module, thereby implementing automatic testing.

Description

technical field [0001] The invention belongs to the field of software automation testing, and relates to an Android automation testing method based on a particle swarm optimization algorithm. Background technique [0002] The current automated testing of Android applications still requires the manual participation of a large number of testers, the test cycle is long, the efficiency and coverage are low, and the use of existing testing tools will lead to code redundancy. The keyword-driven testing framework is passed Methods for manipulating test case generation using test directives. Currently, keyword-based testing frameworks and testing methods are widely used in automated testing of Android applications. Robotium is an open source testing framework based on the encapsulation of the Instrumentation framework, and is mainly used to simulate user scenario testing. Use Robotium to easily write robust and effective automated black-box and white-box tests for Android applicat...

AI Technical Summary

Problems solved by technology

First of all, such automated testing of Android applications still requires the manual participation of a large number of testers, the testing cycle is long, and the efficiency and coverage are low.

Secondly, the existing Android automated testing framework has great limitations in generating test cases for input classes, mainly relying on empirical judgment, which greatly reduces the efficiency and coverage of automated testing