Skip to content

Conversation

@jamesls
Copy link
Member

@jamesls jamesls commented Sep 11, 2024

We should validate that if an error matcher references an error code, then the operation actually has a modeled error with a matching code.

This won't pass as is, so I'm adding this as a reference for the time being. We likely won't be able to get all the service models updated immediately, so we may need to allow list existing cases that are valid but not fully modeled if you do want to pull this in.

Side note, looks like when this file was migrated to pytest a few years ago (#2495) the parameterization logic was changed such that instead of 1 test per individual waiter, it's now 1 test for all waiters within a service. This will still catch any issues, it just means that it'll fail at the first waiter validation error is encounters in the service so there may actually be more waiters that are invalid for the service.

Results

Looks like there are a few additional failures, with a mix of error and unmodeled behavior:

Looks like there's two errors, one's being tracked, and one from rds that's not.

We should validate that if an error matcher references an error code,
then the operation actually has a modeled error with a matching code.

This won't pass as is, so I'm adding this as a reference for now.
We likely won't be able to get all the service models updated
immediately, so we may need to allow list existing cases that are valid
but not fully modeled.
@jamesls jamesls force-pushed the waiter-error-validate branch from 9d51bce to 74745b9 Compare September 11, 2024 21:31
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant