1.
2.
Proposed in 2001
3. Characteristics of Agile Testing
The second point: The division of strict stages in traditional testing is not emphasized in agile testing (unit testing, integration testing, system testing, acceptance testing, there is no such stage division)
The third point: For example, after the function of a module is written, it is necessary to start testing, which is basically the same as the H model we mentioned earlier.
The fourth point: For example, using the Kanban method in the agile method, timely feedback the current situation to the project team members.
4.
The last heavyweight document, detailed documents, test plans, functions, etc. will appear at each stage
Usable software is more important than complete documentation
5. Script Based Testing - SBT
Also called ST
The emphasis is to do the test design first, and then execute the design.
Scripted refers more to test cases in manual testing; in automated testing, it refers to the scripts of automated testing.
ET is an exploratory test.
ET generally explores our system under test, uses our system under test with problems, discovers test points during the exploration process, and finds out test problems.
During testing, test design and test execution are performed in parallel. The execution of ET will be more free for the tester, so the ET will be more dependent on the responsibility of the tester
ST and ET are generally used in the project
From the implementation of the actual project, pure scripted is a complete ST, which is completely executed with reference to test cases, and the use case writing will be very detailed.
Freestyle ET refers to ET that is completely free, without any documentation to back it up, and it does not record the main points of the testing process. These are two extremes.
Vague Scripted: The use case writing is relatively vague, for example, the description of the execution steps for the expected result will be relatively simple.
Fragmentary test cases: Test cases will not be classified as documents, and may only describe the test ideas in one or two sentences, or can be regarded as a list of test points.
charters: more inclined to ET, ET will list a detailed task list, point out the test object, test strategy and point out possible risks, reference documents, etc. in the list.
Roles: It is just a given role for the tester. The tester starts from this role, and the test progress and quality are controlled by the tester himself.
Compare ST and ET
Advantages of exploratory testing
SUT: soft and test test system
shortcoming:
The SUT system under test, because exploratory testing emphasizes the continuous interaction with the system under test, has a deep learning process, so its role is limited when the system is not fully available.
Because of the testing process and the degree of completion, it is difficult to have a quantitative indicator
More reliant on tester personal skills and creativity
local exploration
For input, the basic tasks during testing include: accepting input, generating output, storing data, and performing operations. When testing, consider the test points from the perspective of input order, input content, and output exceptions.
State: Temporary state (valid at runtime, valid at stage), permanent state (database save, file save). The status information can help us to judge the test input and test output more effectively.
Code path: more about test coverage (see the white-box approach for details)
User data: The test simulates real data as much as possible, or constructs reasonable test data
Execution environment: operating system, network topology, third-party systems, configuration data, running devices
global exploration
Let the testers test the system under test like a tourist tour, and divide the test system into different areas for testing according to different attributes.
Business area: some functions that users may use from startup to shutdown
Hotel area: The software rests or does not have the function of actual running time, generally it is a background process or a scheduled task
History area: some functions in the legacy code of the version history, or functions with many problems in previous tests
Tourist area: functions that new users will use or pay more attention to, such as new user registration or novice guide
Entertainment Zone: Auxiliary features or functions beyond the primary function of the system
Dilapidated area: some functions that have been abandoned or hidden in the system, generally not mentioned in the user manual
Exploratory testing process:
know you: You need to understand the focus of the test task, the main test direction, and the system environment
learning: cherish each other to learn and explore the system under test, understand the business logic and function of the system
coverage: In the main implementation stage, the test acceptance of the main function points and the coverage of the test points are completed.
deep: On the basis of the coverage of the previous stage, further dig deeper things
close: Summarize the previous test work, sort out the test information, and analyze whether there is any omission according to the record summary
Clean-up: retest for omissions in close
6. RBT Risk-Based Testing
What are the risks?
Quality risks of the system under test: software functionality, applicability, software missing, etc.
Risk management: Insufficient personnel skills, insufficient manpower, lack of testing environment, unclear requirements of the system under test, problems associated with third parties
identification of risks
possibility:
Project cycle time is tight and risk is higher.
Demand changes rapidly and risks are high.
Geographical dispersion, the R&D team is too scattered and affects efficiency
severity:
The higher the frequency of use, the greater the risk.
Failures are more visible, failure visibility
Business losses, the greater the loss, the more serious the problem (billing function)
Negative organizational impact and damage (Pepsi-Cola logo appears in Coca-Cola system; fast broadcast is banned)
Advantages of RBT
Prioritize high-risk testing, increase testing effort, and reduce risk
Based on human nature, I will be lazy at first, and I will do simple things first.
However, RBT has relatively high requirements for the ability to identify risks in the project, and has more detailed requirements for functions and various elements. If the accuracy rate does not meet the requirements, it is easy to produce evaluation deviations.
7. MBT - Model-Based Testing
A type of software testing in which test cases are derived in whole or in part from a model. This model describes some aspect, usually a functional part, of the system under test.
The model here refers more to the model of the demand function point rather than the process model (wv)
More biased towards automated testing