Chris Russell

@lpinca To make the preferred usage easier, how do you feel about making `abortHandshake()` part of the public API? This way those who want to implement custom handshake preprocessing are...

>I prefer developers to write their own abortHandshake() helper if they have to. It's trivial to do. I think "if they have to" is they key phrase. The current recommendation...

It is still forced reimplementation of existing, proven, tested, and versioned behavior. It seems unfortunate that developers do not have access to invoke the built-in abort behavior. :( I don't...

Most of the async work I do is request-response. Is there anything community members like me can do to help advance this?

To me, the `request-response` information should not interact with the `publish` nor `subscribe` information and thus, the `request-response` details should not have any interaction with the `pub-sub` oriented sections. I...

@shockey I'm confused why this has been tagged as a `Feature` instead of `Bug`. According to the [Specification](https://swagger.io/specification/#schemaObject), `schema` is listed as supporting `anyOf`, `oneOf`, and `allOf`.

Applying the `oneOf` to type within `components/schema` **does not** work around this problem. ```yaml components: responses: ... ServerWriteError: description: Write Failed content: application/json: schema: $ref: '#/components/schemas/ServerWriteErrorResponse' schemas: ... ServerWriteErrorResponse: oneOf:...

@shockey That makes sense... thanks for the clarification :)

One of the challenges with this package is that a modified version if the nodejs is embedded within it. What we really need is very clean, easy to follow, instructions...

@squaremo Any 2018 thoughts on this? The AMQP libs that I've encountered in other languages all support the transaction semantic. In some of your comments on this thread, I see...