iamdanfox

I don't consider resolving the java case of importing identically named types from entirely disparate APIs `com.palantir.product1.Row` and `com.palantir.product2.Row` to be a goal of this issue (indeed changes to conjure/generators...

I don't think @ferozco is suggesting removing packages from generated Java code - we're still in sync that in java-land we need packages. The long-term proposal is just that since...

If we left conjure-java unchanged but major rev'd the IR to remove java-style packages, then sure that would end up with with conflicting classes (e.g. Row) and unhappy devs. To...

No Rob, "Conjure definition" was still referring to the [IR](https://github.com/palantir/conjure/blob/af223471fa7273e2232ac6f93483735c7bd88672/conjure-api/src/main/conjure/conjure-api.yml#L5)! Don't worry I'm not going back to the YAML parsing hell.

@shzhng how would you want a typescript or python client to interact with an API that has a rich 'UserPrincipal' type? ```yaml default-auth: header: type: UserPrincipal markers: - Auth param-id:...

One further argument in support of this wire-spec change is that Chrome actually strips the `Content-Type` header if a request body is empty!

Similar to #23

@markelliot I think the intention of this integer64 proposal is that it is sent over the wire as a quoted string but in all practicable languages (Java, Golang, Python, Rust),...

This is a tricky question because for some conjure-generators this is not a problem at all, and for others it's quite tricky to handle. (We've seen something similar in python...

I think it a sensible outcome that when an _orderable_ version is required, .dirty is rejected because they are not orderable. We specifically made sls-packaging lenient to avoid this problem....