Test case generation method and deice, equipment and storage medium

Abstract

Embodiments of the invention disclose a test case generation method and device, equipment and a storage medium. The method comprises the following steps of: obtaining user behavior data on a terminaland denoising the user behavior data; carrying out training according to the denoised user behavior data so as to obtain a page prediction model, and generating a corresponding relationship between page skipping and user operation; and generating a test case on the basis of the page prediction model and the corresponding relationship between page skipping and user operation. According to the method, the user behavior data is denoised so as to improve the test case generation correctness, the page prediction model and the corresponding relationship between page skipping and user operation are obtained on the basis of the user history behavior data, so that possible operations of users can be predicted, automatic cases can be executed, case quantity can be decreased and feasibility of the cases can be ensured. Moreover, through importing an extension page and a loop occurrence number threshold value, a balance point between prediction and real user operation is realized, case generationfailures caused by incorrect prediction are decreased, and a case set is supplemented at the same time.

Description

technical field

[0001] Embodiments of the present invention relate to testing technology, and in particular to a test case generation method, device, equipment and storage medium. Background technique

[0002] With the increasingly complex functions of application programs (APP) and the increase of customized versions, the requirements for testing efficiency of APPs on terminals are also getting higher and higher. At present, the test efficiency is mainly improved by generating automated test cases.

[0003] At present, the Monkey test simulates the user's key input, touch screen input, and gesture input, etc., to test how long the terminal will be abnormal, but the Monkey test random click cannot truly simulate the user's usage scenario. Using the Appium test framework to manually write automation use cases can solidify some user scenarios, but due to the need to manually write use cases, the maintenance cost of automation use cases is increased. In addition, there is also...

Examples

Embodiment 1

[0031] figure 1 It is a flow chart of the test case generation method provided by Embodiment 1 of the present invention. This embodiment is applicable to the situation of automatically generating test cases. The method can be executed by a test case generation device, which can be installed on a terminal. like figure 1 As shown, the method includes:

[0032] S101. Acquire user behavior data on a terminal, and perform denoising processing on the user behavior data.

[0033] Wherein, the application server can collect the user behavior data in the terminal installed with the application program, and the user behavior data includes: operation time, operation page and operation control, that is, which control is clicked on which page by the user at what time. The behavior of the user in the process from starting the application to exiting the application is a set of behaviors of the user, which can be expressed as a session (Session), each session is set with a unique identifier...

Embodiment 2

[0040] This embodiment provides an implementation manner of denoising processing on the basis of the foregoing embodiments. The denoising processing includes at least one of the following: continuous error point data filtering, invalid data filtering and loopback error point data filtering. Each will be described below.

[0041] (1) Data filtering of continuous error points

[0042] Continuous delay data filtering is mainly to filter repeated click noise. For example, if the user clicks "Back" twice in a row, the second click is a delay, and the second click operation needs to be filtered.

[0043] Continuous error point data filtering includes: reading each operation in the user behavior data in turn as the current operation; comparing the current operation with the previous operation; if the current operation is compared with the previous operation, the rest of the operation information is the same except for the operation time , it is determined that the current operation...

Embodiment 3

[0062] On the basis of the foregoing embodiments, this embodiment provides an implementation manner of generating a corresponding relationship between page jumps and user operations in S102. Figure 5 is a flow chart of the test case generation method provided by Embodiment 3 of the present invention, such as Figure 5 As shown, the method includes:

[0063] S501. Acquire user behavior data on a terminal, and perform denoising processing on the user behavior data.

[0064] S502. According to the denoised user behavior data, train a page prediction model.

[0065] like Image 6 As shown, a group of behaviors of user 1 and a group of behaviors of user 2, where user 1 performs operations a, b, and c in sequence on page A, operation c triggers a jump to page B, and performs operation d on page B , trigger jump to page C, execute operation e on page C, trigger jump to page D, user 2 is similar to this, no more details. Depend on Image 6 It can be seen that the page transfer i...

technical field

[0001] Embodiments of the present invention relate to testing technology, and in particular to a test case generation method, device, equipment and storage medium. Background technique

[0002] With the increasingly complex functions of application programs (APP) and the increase of customized versions, the requirements for testing efficiency of APPs on terminals are also getting higher and higher. At present, the test efficiency is mainly improved by generating automated test cases.

[0003] At present, the Monkey test simulates the user's key input, touch screen input, and gesture input, etc., to test how long the terminal will be abnormal, but the Monkey test random click cannot truly simulate the user's usage scenario. Using the Appium test framework to manually write automation use cases can solidify some user scenarios, but due to the need to manually write use cases, the maintenance cost of automation use cases is increased. In addition, there is also...

Examples

Embodiment 1

[0031] figure 1 It is a flow chart of the test case generation method provided by Embodiment 1 of the present invention. This embodiment is applicable to the situation of automatically generating test cases. The method can be executed by a test case generation device, which can be installed on a terminal. like figure 1 As shown, the method includes:

[0032] S101. Acquire user behavior data on a terminal, and perform denoising processing on the user behavior data.

[0033] Wherein, the application server can collect the user behavior data in the terminal installed with the application program, and the user behavior data includes: operation time, operation page and operation control, that is, which control is clicked on which page by the user at what time. The behavior of the user in the process from starting the application to exiting the application is a set of behaviors of the user, which can be expressed as a session (Session), each session is set with a unique identifier...

Embodiment 2

[0040] This embodiment provides an implementation manner of denoising processing on the basis of the foregoing embodiments. The denoising processing includes at least one of the following: continuous error point data filtering, invalid data filtering and loopback error point data filtering. Each will be described below.

[0041] (1) Data filtering of continuous error points

[0042] Continuous delay data filtering is mainly to filter repeated click noise. For example, if the user clicks "Back" twice in a row, the second click is a delay, and the second click operation needs to be filtered.

[0043] Continuous error point data filtering includes: reading each operation in the user behavior data in turn as the current operation; comparing the current operation with the previous operation; if the current operation is compared with the previous operation, the rest of the operation information is the same except for the operation time , it is determined that the current operation...

Embodiment 3

[0062] On the basis of the foregoing embodiments, this embodiment provides an implementation manner of generating a corresponding relationship between page jumps and user operations in S102. Figure 5 is a flow chart of the test case generation method provided by Embodiment 3 of the present invention, such as Figure 5 As shown, the method includes:

[0063] S501. Acquire user behavior data on a terminal, and perform denoising processing on the user behavior data.

[0064] S502. According to the denoised user behavior data, train a page prediction model.

[0065] like Image 6 As shown, a group of behaviors of user 1 and a group of behaviors of user 2, where user 1 performs operations a, b, and c in sequence on page A, operation c triggers a jump to page B, and performs operation d on page B , trigger jump to page C, execute operation e on page C, trigger jump to page D, user 2 is similar to this, no more details. Depend on Image 6 It can be seen that the page transfer i...

Images