Tommy Smith

Since there is a workaround for now, @s3rius, I will leave this issue hanging open and return to it with an attempted fix when I get the opportunity to in...

It was my takeaway from that issue, and this issue too, that the intended FastAPI behaviour would be to have `request` refer to both `request` and `websocket`, which is to...

@s3rius, I've done some more digging into this issue to find, with the help of @kristjanvalur's changes, that the FastAPI dependency resolution successfully resolves either the `Request` of `Websocket` type...

Hi @EteimZ, sorry that your contribution was left to go stale! We've managed to get the client into a state of `mypy` compliance so would you be willing to regenerate...

@EteimZ, great news! Thanks for your help on this 😁 I will review once I get the opportunity as we're heavily into a sprint cycle currently!

@EteimZ, I believe that the stubs all look fine provided that you called the standard `mypy.stubgen` command 😁 The only thing that jumps out is that the `METADATA.toml` seems outdated...

Alas! I found issues of inconsistency between the stubs and the package due to our misconfiguration of `mypy` within the `pyproject.toml` file. I have a PR setup to rectify this,...

@EteimZ, looking at typeshed's `CONTRIBUTING.md` file [here](https://github.com/python/typeshed/blob/main/CONTRIBUTING.md#the-metadatatoml-file), I think we need to add the stubs used by this library into the `requires` field. These, as listed in `requirements-devel.txt`, are: -...

@EteimZ, there's one last thing that needs fixing. A better solution than adding `# type: ignore` to the auto-generated `weaviate_grpc` files is to create stubs for this folder. Could you...

Yes, I believe the stubs need to be propagated into the typeshed repo in order for them to be available to users when performing mypy static type checking When you...