Radosław Waśko

 This is what the annotations look like on a PR. I think this could be a helpful improvement when looking for test failures - instead of having to dig...

I did not manage to cleanup and finish this yet because I was focusing on more pressing tasks. I will re-visit this once I'm back from vacation.

After the changes now the folding of benchmarks is working again and now we get a nice list of test failures in the checks and files changed panels, next to...

> Is it possible to skip checking the data first, but catch the "non-nullable column" error and convert it to a better error? We could do that, but it is...

I think we should remove the default. But even if we do - I realized that does not solve the problem fully yet - the user can still explicitly choose...

> This should be a general solution for Postgres, SQL Server, and Snowflake. The batched insert is more or less a general solution. The problem highlighted in this ticket is...

We may want to start with tweaking the batch size for each backend, to possibly try getting better performance.

Encountered it yet again in https://github.com/enso-org/enso/actions/runs/10245645438/job/28392624237?pr=10733#step:7:2072 run GitHubInitial Tableau Reading Support · enso-org/enso@872384fHybrid visual and textual functional programming. Contribute to enso-org/enso development by creating an account on GitHub.

> [This](https://github.com/enso-org/enso/actions/runs/10262829302/job/28393327668#step:7:2115) is confusing. It seems to suggest that the problem is with the test that should be pending? What do you mean? Do you mean the ``` INFO ide_ci::program::command:...

> However at the end I found valid code that ends the line with `*`: > > ```ruby > > a f = f 5 > > a 10 >...