Geoff Baskwill

Results 26 comments of Geoff Baskwill

This would be a brilliant addition. I would love to be able to use this to import existing resources into stacks. What would you expect to happen in the `UpdateIfExists`...

Something like this, perhaps, but really hoping it's already supported somehow and doesn't need to be added: ```diff /// Configures the HTTP bindings of an operation. @trait(selector: "operation") @tags(["diff.error.remove"]) structure...

@mtdowling thanks! Does this approach make sense (adding a `headers` map to the `@http` trait)? If so, I could take a look at building it after I've finished the PR...

Great questions! Here are some proposed answers: 1. What's the behaviour if someone tries to bind multiple members to the same header / query parameter / payload member with the...

> We don't allow conflicts today in header bindings, so I think at least starting with that stance here would make sense. Sounds good. Pointers on how / where this...

I think I've got the parsing figured out. Next step will be to look at consuming the parsed model in the OpenAPI projection. re: validation: looks like [HttpHeaderTraitValidator](https://github.com/awslabs/smithy/blob/main/smithy-model/src/main/java/software/amazon/smithy/model/validation/validators/HttpHeaderTraitValidator.java) has a...

Going back to the original problem statement, I think I missed a key requirement: documentation. Using a simple map makes it impossible to have documentation for the headers; you get...

An alternative that also solves the documentation problem would be a map where the value is a structure: ``` @http(method:"POST", uri: "/", headers: { "api-version": { value: "v1", description: "The...

Got what I was looking for: ``` { "name": "api-version", "in": "header", "description": "The API version.", "schema": { "const": "v1", "description": "The API version." } } ``` not sure why...

I think I've got most things working for this, just need to implement the validator per @mtdowling 's comments earlier: > We don't allow conflicts today in header bindings, so...