Re: [PATCH] notes: accept any ref for merge

[Junio C Hamano <gitster@pobox.com>]

Jeff King <peff@peff.net> writes:

> I think this change affects not just "git notes merge", but all of the
> notes lookups (including just "git notes show"). However, I'd argue
> that's a good thing, as it allows more flexibility in note storage. The
> downside is that if you have a notes ref like
> "refs/notes/refs/heads/master", you can no longer refer to it as
> "refs/heads/master" (you have to use the fully qualified name to get the
> note). But:
>
>   1. This makes the notes resolution a lot more like regular ref
>      resolution (i.e., we now allow fully qualified refs, and you can
>      store remote notes outside of refs/notes if you want to).
>
>   2. There are already a bunch of names that have the same problem. You
>      cannot refer to "refs/notes/notes/foo" as "notes/foo", nor
>      "refs/notes/refs/notes/foo" as "refs/notes/foo". Yes, these are
>      silly names, so is the example above.
>
> So it's backwards incompatible with the current behavior, but I think in
> a good way.

Yup, I agree with the analysis.

>> ---
>>  notes.c | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> I think you need to adjust t3308 (and you should probably add a new test
> exercising your case; this is exactly the sort of thing that it's easy
> to accidentally regress later).
>
> -Peff