Michael Oberegger

Perhaps the policy could be built on top of (or align to) [`keyv`](https://www.npmjs.com/package/keyv). Could help get it working "out of the box" with other tools. By default, it would cache...

I've been beating my head against this too. Biggest problem for me is the for some reason the `body` is wrapped in a promise and I don't understand why.

Hey @sebakerckhof. Do you have a branch available that you could push? Not sure if I will be much use, but I could certainly try to help here.

@sebakerckhof > Also note that dart-sass, while faster than the old ruby implementation is still substantially slower than libsass. I saw this note in the [README](https://www.npmjs.com/package/sass), just in case you...

For posterity, I have also created [an issue](https://github.com/davishmcclurg/json_schemer/issues/192) in `json_schemer`.

> Maybe you can make use of one of the [JSON schema conditionals](https://json-schema.org/understanding-json-schema/reference/conditionals) instead of using discriminator? Haven't looked into this yet, so unsure if this would work; I am...

> So something a long the lines you did, but the goal would be to keep validation intact somehow. I just realized I didn't explain what the patch did 😆....

Tried it out and still see the unions being dereferenced. There are a couple differences between my setup and your integration test, though. One difference is that my union is...

Just a friendly followup to let you know that I haven't forgotten about this! I was hoping to get to it sooner, but we were ramping up for a release...

I figured out how to reproduce what is causing the `KeyError: key not found: "$ref"` error from `json_schemer`, and was able to reproduce it in one of the existing [tests](https://github.com/ahx/openapi_first/blob/main/spec/middlewares/response_validation_spec.rb#L334-L338)....