Software Testing Interview (3)

I. Mainstream automated testing technology

Selenium: Web UI Automation Testing Application

Appium: APP UI automation testing application

Requests: HTTP 1.1 interface automated testing application

Second, when writing test cases, what methods are generally used to design use cases?

Boundary:

A lot of errors are found on the boundaries of the input or output ranges, not inside the input and output ranges, so designing test cases for various edge cases can catch more errors.

Equivalent Class:

The equivalence class test method is to divide all possible input data, that is, the input field of the program into several parts, and then select a few representative from each partSexual data as test cases.Using the equivalence class division method to design test cases goes through two steps of dividing equivalence classes and selecting test cases.The ideal test is the best test quality using the least amount of test data.

Type division: valid equivalence class; invalid equivalence class

Design test case steps: 1. Determine requirements 2. Determine valid equivalence classes and invalid equivalence classes 3. Design test cases for each equivalence class

Case: QQ login

#QQ login: QQ number of 6-10 digits, including 6 digits and 10 digits, QQ number must be an integer and cannot start with 0
#Valid equivalence classes: 6 digits, 7 digits, 8 digits, 9 digits, 10 digits (cannot start with 0)
#Invalid equivalence classes: 6 digits, 7 digits, 8 digitsNumbers, 9 digits, 10 digits (starting with 0); decimals, letters, special characters, Chinese characters and combinations of the above

The difference from the boundary value method:

The equivalence class division method can select any data in the equivalent range as a representative, and the boundary value analysis method requires that each boundary value should be used as a test condition; boundary value analysisThe method not only considers the input conditions, but also considers the test situation that the output produces.
Scene Analysis:

A process test from the starting point, through a series of operation steps to achieve a certain result, to the end point.The scenario method is mainly used for smoke tests.After passing the scene test, more detailed tests are carried out by other methods.

Cause and effect diagram:

Cause-and-effect diagram method is a method of designing test cases by graphically analyzing various combinations of input, and it is suitable for checking various combinations of program input conditions.

The equivalence class division method and the boundary value analysis method both focus on considering input conditions, but do not consider various combinations of input conditions, and the mutual constraints of input conditionsrelationship, which may ignore the combination of multiple input conditions that go wrong.Therefore, consider using a form that is suitable for describing a combination of conditions to generate multiple actions to design test cases—causal diagrams.

Error inference:

Based on experience and intuition to speculate all possible errors in the program, so as to design the method of test cases in a targeted manner.

According to experience, list all possible errors in the program and special cases that are prone to errors, and select test cases according to them.
Orthogonal:

Orthogonal method, also known as orthogonal experiment method, is using the smallest set of test procedures to obtain the largest test coverage.

3. Why do interface testing be done during the testing process?

1. You can bypass the front end and find many bugs that cannot be found during page operation

2. Check the exception handling capability of the system

3. Check the security of the system

4. Guarantee the quality of the server

4. How to deal with unreproducible bugs during the testing process?

1. The bug found in which version will be reproduced in which version

2. Communicate with developers and work together to reproduce bugs

3. Record the operation steps and the specific information of the bug, and keep an eye on it in the follow-up test

5. How to submit a complete BUG information?

1. The title of the defect

2. The specific reproduction steps of the defect and the specific description of the defect

3. The found version of the defect

4. The developer corresponding to the defect

5. The severity level of the defect

6. Priority of defects

7. Attachment information of defects (including screenshots, logs, etc.)

6. How should I test if I give you a pen?

Reference Software Quality Model (six features and twenty-seven sub-features)

7. How to test the compatibility of the Web?

1. Understand the mainstream browsers

2. Verify the main process and the display effect of the page

If it is not normal, such as the layout of the page is misplaced, some styles fail to load, some functions on the page are invalid, etc.

8. What is the operating principle of Selenium?Why can web automation be realized?