RFR: 8299015: Ensure that HttpResponse.BodySubscribers.ofFile writes all bytes [v2]

[Chris Hegarty]

On Tue, 20 Dec 2022 11:30:24 GMT, Dawid Weiss <duke at openjdk.org> wrote:

>> I'm on the fence a little on this point. I really dislike the potential for a busy loop, but also don't want the added complication of retry/backoff in this loop. But really the latter is the right solution. lemme see how it looks.
>
> Don't get me wrong - you folks are much deeper into low-level details than I am so perhaps it isn't possible that the OS or the hardware ever returns 0 here. From the public API contract perspective there's nothing preventing it from happening though. It's a bit strange for me to see the internal implementation accepts a stricter requirement.

I added a trivial retry mechanism see https://github.com/openjdk/jdk/pull/11722/commits/c1355274911537fd25eae873908b84617bbb2e58

-------------

PR: https://git.openjdk.org/jdk/pull/11722