Table of contents
The first step, test activities decompose quality phase
The second step, each stage of work must serve the quality
In summary, quality depends on the ability of the team
Foreword:
Testing methodology is the cornerstone of software testing, which provides a set of systematic methods and guidelines for planning, designing and executing testing activities to ensure the quality and stability of software. The goal of testing methodology is to find potential bugs and problems and provide reliable test results to help the development team improve and optimize the software.
The most difficult proposition in the testing industry is not testing technology, but testing quality.
Everyone is familiar with this scenario: production problems occur, problems are solved, production accidents are resumed, and responsibilities are assigned to people.
If your team often encounters this situation, it is not bad luck, and there is no need to burn incense and worship Buddha, but there is a problem with the quality system.
There are many factors that affect the quality, and the most important one is-testing methodology.
The first step, test activities decompose quality phase
Different companies may have differences, take the author's company as an example:
- Requirements and technical solution review;
- test design;
- offline test;
- online test;
- Online monitoring.
The second step, each stage of work must serve the quality
Requirements and Technical Solution Review
- External review: testing is not a passive acceptance, it should be brainstormed together with product and development to discover design defects, technical risks and hidden dangers, influence of related parties, etc. (Those who do not have this ability can conduct practical training through internal review);
- Internal review: The team conducts internal review and training of technical solutions to identify concerns and risk points (teams with generally high capabilities can be eliminated).
Purpose: This stage is the cornerstone of quality. By shifting testing to the left, we can discover requirements design flaws, technical solution risks, interface design flaws, performance design flaws, and related party dependencies as early as possible, understand testing concerns, and focus on testability and other issues.
test design
- Use case design: In addition to the understanding of the business, solid basic skills are also required (boundary value, equivalence class division, orthogonality, etc.);
- Scenario design: normal scenarios, abnormal scenarios, compensation scenarios, each key node of the scenario flow must list inspection items, how to simulate special scenarios is a challenge;
- Data preparation: For joint debugging between front-end and back-end, between components, and between own and third parties, relevant solutions, such as Mock interface and Mock data, should be prepared in advance for testing.
Purpose: This stage is the skeleton of quality. Through test design, more test points are covered, more scenarios are simulated, and more adequate test preparations are made to ensure quality and win more precious time for testing.
offline test
- Interface testing: It needs to follow strict interface testing specifications, such as: required items, value range, default value, paging, single interface time-consuming, redundancy, linkage, data landing correctness, security, etc.;
- Single-point coverage: Execute strictly according to test cases, such as: whether functions and requirements are consistent, db data correctness, robustness, security, friendliness, memory leaks, etc.;
- Horizontal coverage: For a scenario, the key nodes involved from the beginning to the end must be covered by checkpoints, including the correctness of function implementation, data reading, data calculation, and data writing;
- Vertical coverage: normal scenes, abnormal scenes, and compensation scenes must be covered;
- Exploratory testing: In addition to the above, exploratory testing can be conducted based on personal experience;
- Regression test: pull regression test set, and ensure horizontal coverage, vertical coverage, automatic regression, etc. of the main process;
- Performance testing: front-end performance testing (when using asynchronous requests, what situations can only use synchronous requests, rendering, compression, what is the correct posture), back-end performance testing (how to analyze the results and locate problems).
Purpose: This stage is the shaping of the quality. Through the full preparation of the test design, the strict and three-dimensional execution of the offline test, most of the problems are found and solved.
online test
- New function test: pull the online quick verification test set, and ensure the horizontal and vertical coverage of the main process;
- Regression test: Pull the online regression test set and ensure the horizontal and vertical coverage of the main process;
- Performance test: full link stress test (data isolation).
Purpose: This stage is the final state of version quality. Online testing is mainly to ensure the impact of code deployment, production configuration, and production environment on quality.
Online monitoring
Purpose: This stage is for quality compensation, quick response and resolution, and reduction of losses caused by production accidents.
In summary, quality depends on the ability of the team
First of all, it is necessary to find a suitable methodology, and secondly, with the same methodology, the implementation effect still depends on the ability of the person. Therefore, never neglect the cultivation of people.
As someone who has been here, I also hope that everyone will avoid some detours
Here I will share with you some necessities of the way forward in automated testing, hoping to help you.
(software testing related materials, automated testing related materials, technical questions and answers, etc.)
I believe it can make you better progress!
Click on the small card below
