[Updated] D10866: revlog: add a way to keep track of older uids in the docket

[Alphare (Raphaël Gomès)]

This revision is now accepted and ready to land.
Alphare added inline comments.
Alphare accepted this revision.

INLINE COMMENTS

> docket.py:144
> +        self._older_sidedata_uuids = older_sidedata_uuids
> +        assert not set(older_index_uuids) & set(older_data_uuids)
> +        assert not set(older_data_uuids) & set(older_sidedata_uuids)

Aren't those asserts kind of overkill? I don't mind keeping this, but I'm wondering if there is a better chance of this happening than I imagine

REPOSITORY
  rHG Mercurial

BRANCH
  default

CHANGES SINCE LAST ACTION
  https://phab.mercurial-scm.org/D10866/new/

REVISION DETAIL
  https://phab.mercurial-scm.org/D10866

To: marmoute, #hg-reviewers, Alphare
Cc: Alphare, mercurial-patches
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mercurial-scm.org/pipermail/mercurial-patches/attachments/20210616/0ad73d66/attachment-0002.html>