Just to note, this is the kind of change that if anyone else on Earth had made this PR, it would have been rejected out of hand. The entire module is rewritten, with no review or even discussion. It was merged within 1 hour of posting, late at night for most contributors.

Realistically, the change is probably fine and there probably aren't any backdoors hidden in 1800 lines of changes, because Raymond probably retains full control of his account and is a trusted contributor. However, I do not believe this should pass without comment and until someone can actually review this change, I would be a lot more comfortable if it were reverted.

This has been nagging at me enough that I have looked through it myself and find no evidence of foul play, but it's also impossible to tell what if anything actually changed and in a change this size (even looking at individual commits) I could very easily have missed something. From a security viewpoint, this PR is terrifying and is strong evidence that we should revisit our policy of not requiring reviews regardless of the major inconvenience that poses.

(cc @hugovk as 3.14 RM)