david-perez

I too was missing the ability to have interactive commands in subshells inside macros, so I came up with this ugly hack: ``` macro index L "unset wait_keyecho \"sales@\\njarvis\" |...

I agree with @Kestrer that the project's name is not ideal since there really isn't anything in it that intrinsically ties it to axum. I think the project's adoption would...

> RPC protocols like AWS JSON can work with this example. How? What does an HTTP body message look like in AWS JSON 1.x for that example?

But isn't #1287 for event streams? This is for a `@streaming` `blob` shape.

The example was changed in https://github.com/awslabs/smithy/pull/1458 to: ```smithy @http(method: "GET", uri: "/streaming-operation") operation StreamingOperation { input := {} output := { @required streamId: String @httpPayload output: StreamingBlob = "" }...

Issues with the same spirit: * https://github.com/awslabs/smithy/issues/1070 * https://github.com/awslabs/smithy/issues/1069 * https://github.com/awslabs/smithy/issues/1071

I agree that allowing empty path segments may inadvertently cause bad API design, but the Smithy spec does not disallow them. Perhaps Smithy CLI could emit a warning if it...

@gosar Yes, I think that should be the behavior, trailing slashes have meaning. It is also consistent with how websites serve HTML documents e.g. `/foo/bar/` will probably resolve to `/foo/bar/index.html`...

I've just stumbled upon a routing issue caused by allowing empty path segments. Consider the S3-like model with two operations: 1. `ListBuckets` with URI pattern `/`. 2. `ListObjects` with URI...

Actually, the issue still applies with a non-`streaming` member e.g. _optional_ `httpPayload`-bound members are allowed. What are their HTTP semantics? --- In restJson1 it's impossible to make the distinction, because...