ronkapoor86 commented on PR #14108: URL: https://github.com/apache/iceberg/pull/14108#issuecomment-3383375551
> > Thanks @ronkapoor86 , I don't have much to say as I am not a test writer expert, and I would be curious to know what the others think about your test, I tried to think about a way of testing what @nastra described [here](https://github.com/apache/iceberg/pull/14108#discussion_r2371288262) which is _please make sure that the test actually reproduces the original issue and fails without the fix vs passes with the fix_, in the sense that the fix we are talking about here is to fix something that stopped to work properly after using spark 4.x.x but was working before. To reproduce the original issue it would mean that we keep in the code something that does not make sense anymore, or maybe we reproduce it only in the test? > > The test just needs to fail if the production code change is removed. Yup, thats the idea with this new test. If we remove the unix_date func we'll get the same repeated error that arises from spark about casting. Tried to make it simpler than before -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
