Brian Tarricone

use `fun` for declaring static class methods, not `class`.

6

Hope you don't mind my sending a PR for this suggestion, if you're interested in it. First off, a huge thanks for writing this book! I've gone through the chapters...

Use DOB rather than Age for pets

2

The current card uses DOB for humans, but has an "Age" field for pets. Certainly this is approximate for many pets, since the exact date of birth is often unknown,...

Allow custom class names for generated clients and server resources/handlers

4

Currently guardrail uses the last component of the custom package name (as specified using `x-{jvm,scala,java}-package` or `tags`) as a prefix for the generated class name. It would be useful to...

Consider: ``` Request: type: object required: - type properties: type: type: string discriminators: propertyName: type FooRequest: - $ref: "#/components/schemas/Request" - type: object required: - foo properties: foo: type: string BarRequest:...

I was working on generating protocol modules for a random project (sadly the API itself isn't well suited to guardrail at present), and found some things that made it a...

This was a bug in the Java Jackson generator as well (see https://github.com/guardrail-dev/guardrail/pull/1168), but the same fix won't work there, as Scala doesn't have the primitive vs. reference type distinction....

Right now we have most of our vendor extensions prefixed with the language or platform in use. So `x-scala-type` describes a custom type for Scala clients, and `x-jvm-type` does the...

Tag & x-(jvm|scala|java)-package conversion to valid identifiers is brittle

1

If someone uses a tag/package name like "this is a fun tag", we'll fail to generate a valid identifier out of that (in Java at least), and guardrail will give...

From some real-world usage, the `fold` method on Java client response classes isn't quite as useful as it is with Scala. Lack of named arguments makes it more error-prone, and...

Handing of required/nullable is in schemas not correct

11

After looking at [a SO post](https://stackoverflow.com/a/45577763/1626395), I've realized we're not quite handling `required` correctly. The `nullable` property (which defaults to false) was added in OpenAPI v3 (there is no analog...