GitHub user shlxue edited a comment on the discussion: About unit testing and UI standardization to simplify UI scaffolding code
Reason: Unit test coverage of the code. The test report can provide a more direct indication of the resolution of the problem Our customers often need to provide verification results for each upgrade change to confirm that the problem can be truly solved. Generally, the corresponding verification results are provided through unit testing or manual testing and verification by testers. There are too many problems with manual testing or simply executing some hpl pipelines: * The testing process is often a black box * Differences in understanding of the problem can also lead to inaccurate results * Takes up too much labor cost In contrast, we want to use unit test coverage as a validation result. It is more accurate. For different pipeline engines, we want all hpl pipelines to work properly and have the same output. However, this may not be the case: some hop plugins often introduce some heavy and relatively heavy framework dependencies due to the needs of the runtime environment, such as the incorrect version of some commonly used libraries, which may lead to errors in the running results; Some cases can be performed normally, but may be inaccurate due to various reasons. In our application, we need to collect the execution results and performance data of all plug-ins to locate the cause of the problem when there is a problem.  So for code, unit test coverage is a very important report. This report is required for code reviews prior to going live to production. GitHub link: https://github.com/apache/hop/discussions/4745#discussioncomment-12103087 ---- This is an automatically sent email for [email protected]. To unsubscribe, please send an email to: [email protected]
