All tests are passing for me under python3, including with this patch. Pushed, thanks!

This is stalled. I've got some bugs in the logic I can't track down and
this is second on my priority list at work. I do intend to come back to it,
but it'd be fair to mark it all as changes requested.

Seems to need rebased?

Something in this stack (that I didn't push) breaks the phabricator extension. See phabricator.py line 170:

@eh.wrapfunction(localrepo, "loadhgrc")
def _loadhgrc(orig, ui, wdirvfs, hgvfs, requirements):

So you'll need to do....something there. Maybe look for other wrappers in extensions too while you're there...

I queued a mailing list version of this series. Should I drop it?

Yeah, if you’re seeing these changes I’m pretty sure you have a too-old black...

In D8691#129807, @graemenail wrote:

Thanks for the prompt review.

In D8691#129796, @durin42 wrote:

(If anything, ~/Library/Preferences/org.mercurial-scm.hgrc might make sense, but that seems silly to me.)

I agree that ~/Library/Preferences would be the default XDG analogy in macOS. However, the patch was more motivated by relocating the configuration file.
Am I to understand that setting HGRCPATH is the preferred approach for relocating hgrc on macOS?

Yes: XDG isn't a thing on macOS, and we won't respect it there. Perhaps the help text should be improved instead.

LGTM, but needs rebased.

My feelings on this are "obviously yes" but I thought that so quickly I lack confidence that I'm not missing something. :)

I'm a fan of this, but for default, not stable. Is there a specific reason you need a new feature on the stable branch?

This is still valid.

Doesn't apply?

Okay, now localrepo correctly uses the same context across whole operations. This almost certainly avoids some bugs, given that committing a memctx was (surprise!) reading the mergestate from disk.