Ludovic Orban

@sbordet this seems related to #11016, maybe we did not fill all the gaps?

This seems to be another manifestation of https://github.com/eclipse-ee4j/jersey/issues/5675

@gregw If my understanding is correct, you're proposing to manage static content with a mechanism vaguely similar to the the 11.0.x handler one: pass the request/response/callback trio to some `ContentFactory`...

Unfortunately, you cannot have `RetainableByteBuffer.writeTo(Content.Sink sink, boolean last, Callback callback)` as `RetainableByteBuffer` is in `jetty-util` and `Content.Sink` is in `jetty-io`. This is the core reason that lead to the way...

@gregw Making `HttpContent` extend `RetainableByteBuffer` would mean that `HttpContent.getByteBuffer()` cannot return null, but this behavior is (so far) expected as `HttpContent`'s byte buffer is meant to represent cached content that...

@gregw There is no need for `HttpContent` to extend from `RetainableByteBuffer` for it to expose a `writeTo(Content.Sink sink, Callback callback)` method and to remove the `getByteBuffer()` method. The `IOResources` helper...

The `HPACK - max dynamic table size change required` is reported by a class that seems to be part of [Netty](https://github.com/netty/netty), and I see no trace of Jetty's involvement. Aren't...

Oh okay, got it, and sorry for the confusion. Indeed it looks like your Netty client doesn't like something about H2 so it aborts its connection by sending Jetty a...

I don't think I agree with your statement that `ContentSinkOutputStream#close()` and `ContentSinkOutputStream#flush()` should block conditionally. But it seems that you have found a real issue with our implementation of `ContentSinkOutputStream`:...

Vlad, Throwing more specific exceptions could be done, obviously as long as that doesn't break anything. I'd recommend you to open a feature request and start working on it in...