Thanks, both look good.

Done. The refactor part was extracted into https://phab.mercurial-scm.org/D11818.

Yes, I think in this case a separate changeset for the type refactor (of the old code) would make sense.

@Alphare, Ok. I understood you before, but those commits were not suitable for independent submission.

In D11722#181171, @aalekseyev wrote:

@Alphare, folded commit seems better than the separate ones in this case, though, since the second commit is simply incorporating review comments, not really an independent change.
I ran hg fold --from .^ followed by hg phabsend . to make this less confusing.
If you were thinking of a certain logical split of this commit into two that I'm not aware of, please let me know.

@Alphare, folded commit seems better than the separate ones in this case, though, since the second commit is simply incorporating review comments, not really an independent change.
I ran hg fold --from .^ followed by hg phabsend . to make this less confusing.

Thanks!

I took the liberty of assigning this as the parent of D11817 as if you did the phabsend of the range, this way it's more obvious for other reviewers/archival reasons. You can see the stack tab appeared

@martinvonz @durin42 pinging you because you were both involved in the last diff

@Alphare, I re-created the commit and was able to make a phabricator revision https://phab.mercurial-scm.org/D11817.

✅ refresh by Heptapod after a successful CI run (🐙 💚)

✅ refresh by Heptapod after a successful CI run (🐙 💚)

✅ refresh by Heptapod after a successful CI run (🐙 💚)

Trying to send this, I get:

$ hg update 356d27775359
$ hg phabsend .^ .
D11766 - updated - 48486:e53ef79cd9d5 "sparse: demonstrate a bug when used with safe-share"
abort: Conduit Error (ERR-CONDUIT-CORE): Graph cycle detected (type=5, cycle=PHID-DREV-z7jbc5yf2jun6v2pbnqj, PHID-DREV-z7jbc5yf2jun6v2pbnqj).
EXIT STATUS 255