Andrea Peruffo

@dcais I think you are talking about the `java-generator` here and not the CRD generator, can you please confirm? I tested the java-generator` and for the `class` field you get...

cc. @EricWittmann

> It's most likely caused because the bundled application actually has two copies of the abstractions Haven't realized! Why? > register the serialization providers through a derived request adapter Do...

Sorry for the delay @baywet , the linked PR is still relying on `defaultInstance`, I have not tested, but I'm not sure how it addresses this issue. > NEVER have...

I like it, it's possibly an improvement for all implementations 🙂

> would be a breaking change for other languages at this point I think that it can be rolled out as a non-breaking change: 1. keep the default instances, only...

I quickly looked into this, and, to have a fully functional Java JDBC driver based on the SQLite3 C compatibility layer there are a few missing APIs: - [x] sqlite3_changes64...

> Have you tried to use [HeadersInpectionOption](https://github.com/microsoft/kiota-http-dotnet/blob/main/src/Middleware/Options/HeadersInspectionHandlerOption.cs) to do that? No, I haven't, but an interceptor needs some kind of synchronization in addition to the request and it's inherently not...

I confirm: https://github.com/microsoft/kiota-http-dotnet/blob/0212e818931b31a2d9e0ab4d768b31f23fe5f386/Microsoft.Kiota.Http.HttpClientLibrary.Tests/Middleware/HeadersInspectionHandlerTests.cs#L43-L55 This API has some shortcomings: - it's bound to the http client implementation - it's extremely low level - requires merging information coming from 2 different sources...

Thanks for the feedback! > I'd expect GitHub's pagination logic to be implemented in GitHub's core library for the time being. The problem of having the implementation tied to the...