D748: cleanupnodes: separate out bookmark destination calculation from actual update

[quark (Jun Wu)]

quark accepted this revision.
quark added inline comments.

INLINE COMMENTS

> scmutil.py:604
> +            # move bookmark backwards
> +            roots = list(unfi.set('max((::%n) - %ln)', oldnode,
> +                                  list(mapping)))

I thought this could be `O(changelog)`, but it's `O(1)` given generatorset implicitly requires the generator to be sorted.

REPOSITORY
  rHG Mercurial

REVISION DETAIL
  https://phab.mercurial-scm.org/D748

To: martinvonz, #hg-reviewers, quark
Cc: quark, mercurial-devel