Replies: 9 comments 1 reply
-
Unit Tests
ConclusionUnit tests are fine as is, people can freely add them in the Rust source whenever they feel appropriate. |
Beta Was this translation helpful? Give feedback.
-
End-to-End
ConclusionThis is nice to have, not top priority right now. |
Beta Was this translation helpful? Give feedback.
-
Smoke Tests
|
Beta Was this translation helpful? Give feedback.
-
Testing Business Partner Applications
ConclusionWe need at least something, no final conclusion yet. |
Beta Was this translation helpful? Give feedback.
-
Prevent Performance Degradation
Conclusion
|
Beta Was this translation helpful? Give feedback.
-
Snapshots
ConclusionThis is currently a mess, but we should at least have a consensus on what to snapshot when we add feature. Further discussions needed. |
Beta Was this translation helpful? Give feedback.
-
Rust Integration Tests
ConclusionWe should clean this up. |
Beta Was this translation helpful? Give feedback.
-
JS Integration Tests
ConclusionThis is our preferred way of testing. We should improve our toolings for better developer experience. We also need a way to compare our test suite against webpack for making visibility on our compatibility with webpack. |
Beta Was this translation helpful? Give feedback.
-
It just came to me. I believe the key is to reduce complexity. IMO, we should remove all integration tests in Rust, which leads us to no more With a good design of the test system in js, it might be enough for testing correctness. I personally prefer testing in Rust. But It seems a mess in practice for such a project. |
Beta Was this translation helpful? Give feedback.
-
At the moment of writing, our test infrastructure need to be improved. We are spending way too much time doing ad-hoc debugging.
I'll use structured comments for gathering feed backs.
--
Overall Conclusion:
Need a guide.
Beta Was this translation helpful? Give feedback.
All reactions