Greg Wilkins

In some ways this is related to #11094 where I'm adding async APIs to RetainableByteBuffer. A HttpContent is kind of similar

Building on the work in #11598 that makes RetainableByteBuffer a more generic buffer API, we should look at changing `HttpContent` to either *have-a* `RetainableByteBuffer` or *extend* `RetainableByteByffer`. The key API...

The API of `HttpContent` currently has methods including: ```java long getContentLengthValue(); ByteBuffer getByteBuffer(); void release(); ``` and the new `RetainableByteBuffer` has methods including: ``` long size(); ByteBuffer getByteBuffer() throws BufferOverflowException;...

@lorban I've had a quick start at the approach above (`HttpContent extend RetainableByteBuffer`) and that first part looks pretty trivial. Do you want to do it, or shall I turn...

@lorban you can have a getByteBuffer() that returns null (although technically it would be better for it to now throw BufferOverException), but the HttpContent type needs to implement the writeTo...

@lorban but why isn't a HttpContent is-a RBB, just like a chunk is RBB? It has the same signature?

@scrat98 please note #11598 which deprecates ChunkAccumulator. It is still draft status, but we are working on improving this area of the code. However, we will still look at a...

Some progress here: #10897

Your XML is badly formed as the `TotalCapacity` and `UsedCapacity` elements are not closed. You also may need to turn off validation. I fixed the XML and wrote the following,...

The Jetty's team analysis of this issue is that we have detected an error while the request is asynchronously waiting (probably a HTTP/2 stream reset - which is a semantic...