Junio C Hamano <gitster@xxxxxxxxx> writes:
[snip]
> * kn/reflog-migration-fix (2025-01-15) 1 commit
> (merged to 'next' on 2025-01-16 at ae8f9ce9a0)
> + reftable: write correct max_update_index to header
> (this branch is used by kn/reflog-migration-fix-fix and kn/reflog-migration-fix-followup.)
>
> "git refs migrate" for migrating reflog data was broken.
>
> On hold.
> cf. <Z4mUizLNUdq_1BgY@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>
> cf. <CAOLa=ZT4nws0irdZKUuWc70Rv9RUNQuSXnGAt1SnE1O+umSReg@xxxxxxxxxxxxxx>
> source: <CAOLa=ZTL9n_DPhNr49XAd6bT838kc09oVx_AH7Pb4o8VK_xQ9w@xxxxxxxxxxxxxx>
>
There was some discussion around Coverity [1] on this series. But the
conclusion was that this (all instances of not checking the return code
of `check()` in our unit tests) would be fixed when the unit tests are
migrated to the clar framework.
Apart from that, there was a Nit suggested by Patrick [2]. Which also is
a broader cleanup and can be done as a part of a later patch as
#leftoverbits.
So all in all, this can be moved from 'on hold'.
As for a brief about the topic, perhaps:
Additional safeguards to prevent 'update_index' changes post record
additions in the reftable backend and small cleanups alongside.
[1]: https://lore.kernel.org/git/20250201022409.GA4082344@xxxxxxxxxxxxxxxxxxxxxxx/
[2]: https://lore.kernel.org/git/Z5DgxQuc2j_-5GHg@xxxxxx/
Attachment:
signature.asc
Description: PGP signature
