Zhejiang Smart Puhua Financial Leasing Co., Ltd. (hereinafter referred to as "Smart Puhua") was established in August 2013 with a registered capital of US$107.86 million. The company's shareholders are Zhejiang Zhongjian Investment Co., Ltd., BNP Paribas Personal Financial Group, Zhejiang Geely Automobile Co., Ltd. and Geely International (Hong Kong) Co., Ltd. Smart Puhua relies on the industrial resources of Geely Holding Group and the risk control system of BNP Paribas Personal Financial Group. It has branches in 29 cities and its business covers the country. Smart Puhua currently mainly provides individuals and enterprises with exclusive financial leasing products and services for brands such as Geely, Lynk & Co, Geometry, Yuan Yuan, Ouling and Qianjiang Motorcycle.
After years of development, Smart Puhua has developed strong product research and development, financial financing and risk management capabilities, and is committed to providing customers with professional, flexible and convenient financial leasing services. Its main businesses include direct leasing business, leaseback business, commercial vehicle business, motorcycle business and corporate business. Its platforms include "Huiyouqi", "Zhiweiban", "Huixiangjia" and "Huiyouche" , "Smart Peer" mini program, and the official WeChat service account "Smart Financial Leasing".
▲ Attached picture: Smart Puhua business model
1. Smart Puhua test platform selection ideas
Smart Puhua first came into contact with the MeterSphere open source continuous testing platform through an Internet search. At that time, Wisdom Puhua was trying to find a testing platform, hoping to provide support for the difficulties or difficulties encountered by the team during the software testing process. At that time, the challenges faced by the Smart Puhua test team included:
1. It is difficult to get started with automated interface testing, performance testing and UI automated testing;
2. Different test teams in regular iterations and core business system project teams use different test strategies, different IT management tools, and different methods and processes, resulting in low test delivery efficiency and inconsistent data export;
3. The pace of business requirements from entering the pool to online delivery and acceptance is getting faster and faster, the testing workload increases, team work continues to stay in the manual testing stage, automated test coverage is low, and high-quality software delivery is difficult to sustain;
4. The existing ZenTao, Luban, and Eolinker tools have single functions and are immature. Data links urgently need to be effectively integrated to provide strong support for the integration of R&D and testing;
5. There is a lack of tools and platforms that can effectively and accurately feedback personnel energy efficiency. The coordination and work arrangement of team personnel require the support of relevant data.
Taking into account the needs of business development, Smart Puhua's software delivery and testing team plans to build a unified testing platform to effectively standardize test management work. For the team’s internal testing platform, Smart Puhua’s specific needs include:
■ Project management: All testers use the test portal uniformly for test planning, tracking, execution and management. Establish isolated project test areas for each project and project members, create test accounts for each tester, and assign them to one or more projects. Members of each project can perform testing and management in their authorized project test areas. Test cases, test plans, perform interface testing and performance testing;
■ Visual display: For the leaders of the testing department, the overall testing situation of each project is displayed online in real time, and the version release and testing progress of each project are visually displayed, simplifying the communication link and reducing the workload of manually organizing test reports. It can intuitively display test quality and dynamically display test data and status information through visual charts. Visual charts can support user-customized arrangement and design without development;
■ Test tracking management capabilities: For test team members, during the development and testing phases, online management of test cases and test plans, execution of test plans, viewing of test reports, execution of interface tests and performance tests, tracking management capabilities cover test plans to test execution, Test report analysis and other stages. The platform needs to have test case management and defect management modules to identify, manage, and communicate any defects in the software life cycle to ensure that defects are tracked and managed without being lost;
■ Team collaboration ability: Improve the synergy between the test team and other teams and reduce daily communication costs;
■ Low learning cost and low migration cost: It should be simple and easy to use for most testers, and can quickly carry out functional testing, interface automation testing, performance testing and UI automation testing on the platform. Save team members’ learning time, put it into use quickly, and produce results quickly. Regarding how to store or use existing test scripts, we hope to achieve minimal migration or transformation workload and improve the level of automated testing;
■ Integrate into DevOps: Compatible with existing software release pipelines and improve the operating efficiency of software release pipelines;
■ Automated testing: including interface automation, UI automation, and performance test automation modules, which can implement regression of modules that have completed functional testing, discover potential defects as early as possible, and reduce the defect omission rate.
2. Comparison of using experience of testing tools and platforms
Based on its own work planning and needs, Smart Puhua's testing team finally chose the MeterSphere one-stop open source continuous testing platform as the enterprise's unified test management platform.
When using the MeterSphere platform for the first time, Smart Puhua's test team believed that the platform's interface was simple and its functions were easy to use. However, because it was positioned as a "one-stop" platform and covered many functions, it took a certain amount of time to become familiar with and use it. The following are some advantages and disadvantages of the MeterSphere platform and other similar tools it has used, summarized by the Smart Puhua test team.
1. MeterSphere open source continuous testing platform
Shortcomings: There is still room for optimization in use case writing, map writing use cases, use case import function, and XMind import function. The test cases are not intuitive and clear enough, and the corresponding solvers cannot be reminded after the bug is raised. Compared with Jira, the degree of freedom and scalability are smaller.
Advantages : low-coding design for interface testing, low learning cost; high integration, improving testing efficiency and helping to shift testing to the left. It turns out that the ZenTao, automation, and stress testing tools used by the team were all scattered and not integrated on one platform, making it difficult to manage them uniformly. As a one-stop testing platform, MeterSphere integrates use cases, interface automation, UI automation, and performance testing, effectively reducing management costs.
2. JMeter
Advantages: Open source software, easy to expand, lightweight design, can meet the interface and performance testing of multiple protocols.
Shortcomings: The learning cost is high, it is impossible to manage test scripts, plan test scenarios online, and have multiple people view the test process and test results in real time. It is inconvenient to carry out interface testing and performance testing in a team-based manner, let alone project-based testing. Iterations are carried out in real time and continuous testing is carried out. The interface that passes the test cannot be directly referenced and needs to be redefined and edited. Similarly, scenario planning for performance testing cannot be moved forward to the interface testing stage, and test scripts need to be prepared during the performance testing execution stage. Such requirements are very common for web platform system testing, but JMeter does not have such management convenience.
3. Postman
Advantages: The most mainstream interface testing tool, simple and easy to use.
Shortcomings: Postman is only a testing tool, suitable for use by a single person on a single machine. It is not suitable for teams of multiple people and multiple projects to work at the same time, and is not conducive to test management.
3. Implementation results of Smart Puhua using MeterSphere
Based on the MeterSphere continuous testing platform, Smart Puhua has realized docking with Eolinker, implemented interface import and automatic synchronization of interfaces, and improved the collaboration efficiency of development and testing. Some of the highlighted functions of the MeterSphere open source continuous testing platform have left a deep impression on Smart Puhua’s testing team and have also brought many practical benefits, including:
1. Test coverage statistics: interface testing, interface automation testing, and performance testing have been expanded, and UI testing can be improved in the future to increase test coverage;
2. Standardize the test process: Use the MeterSphere platform to define a test management model based on project dimensions. Project management → Prepare project test plan → Design test cases (including functional test cases, interface test cases, performance test cases) → Organize test case review → Execute the test → output the test report to further standardize the test;
3. Standardized test report: After each test plan is completed, the corresponding test report can be generated. The test quality feedback is more efficient and truly well-founded, and the test result data in the system is highly consistent;
4. Test data precipitation: Accumulate quantitative test data through the test platform to manage testers, thereby establishing a quantifiable evaluation system to measure the work of testers, laying the foundation for improving test energy efficiency;
5. Reduction in regression testing costs: return responsive test results through automated interface testing and automated performance testing. Automatically complete the test for multi-business scenario regression testing and feedback the corresponding test results to the R&D personnel. Test regression efficiency is significantly improved, and it also improves developers’ R&D efficiency;
6. Team collaboration: Through organizational management and project management, different organizations, teams, and personnel can perform their respective duties on the continuous testing platform. From product requirements to use case review, from use case execution to execution pass rate display, different organizations, teams, and personnel can perform their respective duties on the continuous testing platform. The work of personnel is connected in an orderly manner, and on this basis, the management of the entire test life cycle is realized, which strengthens the collaboration ability between personnel and reduces communication costs;
7. Test progress control: Visualize the test process and performance data, analyze it from multiple dimensions such as teams, personnel, projects, test cases, test processes, test results, etc., and intuitively display it through large-screen visualization, realizing the Overall control of project progress.
Up to now, the MeterSphere open source continuous testing platform has been running in Smart Puhua for more than seven months. The current number of users has reached 34, and it carries the use case management of more than 4 projects. More than 1,800 functional use cases and more than 1,500 have been created on the platform. Interface use cases, as well as more than 100 interface and scenario use cases, are also in the process of exploring the UI testing module of the MeterSphere platform, and have initially achieved results.
4. Suggestions and expectations
After a period of in-depth use, Smart Puhua has some expectations for the iterative optimization of the MeterSphere open source continuous testing platform.
First, there is room for improvement in software quality, and we hope that the platform’s functions can be used more smoothly; secondly, we hope that use case writing, map writing use cases, use case import functions, and XMind import functions can be optimized to a greater extent. Test cases can be more intuitive and clear, and the corresponding solvers can be promptly reminded when bugs are raised; thirdly, it is hoped that the MeterSphere platform can reach or even surpass the freedom and scalability of the project management tool Jira.
Lei Jun: The official version of Xiaomi's new operating system ThePaper OS has been packaged. The pop-up window on the lottery page of Gome App insults its founder. Ubuntu 23.10 is officially released. You might as well take advantage of Friday to upgrade! Ubuntu 23.10 release episode: The ISO image was urgently "recalled" due to containing hate speech. A 23-year-old PhD student fixed the 22-year-old "ghost bug" in Firefox. RustDesk remote desktop 1.2.3 was released, enhanced Wayland to support TiDB 7.4 Release: Official Compatible with MySQL 8.0. After unplugging the Logitech USB receiver, the Linux kernel crashed. The master used Scratch to rub the RISC-V simulator and successfully ran the Linux kernel. JetBrains launched Writerside, a tool for creating technical documents.