Greg Wilkins

Hmmm the text is rather complex and on my third read through I'm still not sure I entirely understand. I'm wondering if we are trying to be too smart by...

@stuartwdouglas I'm OK with not calling WL.onError or RL.onError if they have been closed, which means: - a RL is closed if: - InputStream.close() has been called - a read...

@stuartwdouglas Can you wake up this PR and let's get it merged.

We just did some testing on the [2] issue with `getHttpServletMapping`. It appears the currently undertow is implementing it according to the spec, but that tomcat and jetty are implementing...

@markt-asf @stuartwdouglas I'd like to make progress on this one and have opened https://github.com/eclipse-ee4j/jakartaee-tck/issues/585 as I think the TCK is wrong. If you are in agreement, the I'll create PR...

I've opened https://github.com/eclipse-ee4j/servlet-api/issues/385 to get the TCK updated

I've updated the table in the original post with some questions. It would be good to close this one off and get our consensus for it captured in documentation and/or...

@markt-asf any chance you could update the table with your feedback. I tried, but there were just a few too many cells that didn't seam right (e.g. "Named Forward -...

@markt-asf my understanding of that TCK test is that the servlet throws the wrapped exception - ie it throws `new ServletException(new IllegalStateException())`. To my mind it is wrong to unwrap...

Jetty's handling of 100-Continues is to wait until either an InputStream or a Reader is obtained from the request to indicate that the filter/servlet/async is willing to read a body....