On 31/05/2017 09:38, Andrzej Krzemienski wrote:
3. Under the assumption that error_code_extended's use of a static global ring buffer would be controversial in this review, I hacked a quick and dirty solution expecting to have to remove it. Now we know that few disapprove, a more serious implementation will be needed. https://github.com/ned14/boost.outcome/issues/52
I am not sure of that. It was my impression that many potential reviewers were put off by documentation and legal/structural issues, and stopped with their review at that point. Once you have move thes obstacles out of the way, you will likely draw more reviewers, and who knows what they have to say about the ring buffer.
FWIW, I don't have a problem with the ring buffer per se, but I think it's currently too small to be effective.