prep counter with failure

[max-hoffman]

No description provided.

[zachmu]

I like the sentiment of making sure that all tests pass with prepared statements, but I think this is going to cause as much confusion as it catches bugs. The thing with the magic number especially -- people are just going to change it instead of adding new prepared tests.

A better way of assuring prepared coverage would be a separate harness combined with code gen, with a manually updated list of test methods that can't be used with the prepared harness.

[max-hoffman]

@zachmu If i codegen it, then I have to do it separately for memory_engine_test.go and dolt_engine_test.go, which is probably 2-3 days of automation. You don't think this is a useful stop-gap until we have a bit of time to rework testing?

[zachmu]

I would hold off on these checks until we can do it a better way

…

On Fri, Jul 29, 2022, 07:44 Maximilian Hoffman ***@***.***> wrote: @zachmu <https://github.com/zachmu> If i codegen it, then I have to do it separately for memory_engine_test.go and dolt_engine_test.go, which is probably 2-3 days of automation. You don't think this is a useful stop-gap until we have a bit of time to rework testing? — Reply to this email directly, view it on GitHub <#1137 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AADT5FORCF4YGASM4AZVO7LVWPU4FANCNFSM546QSOVA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[timsehn]

Should we just close this?

[max-hoffman]

Fine w me, the checks in the Dolt repo have worked well enough on their own.