"Daniel Verite" <daniel@manitou-mail.org> writes:
> Tom Lane wrote:
>> Why? If we agree that that's the way forward, we could certainly
>> stick some collversion other than "1" into pg_c_utf8's pg_collation
>> entry. There's already been one v17 catversion bump since beta2
>> (716bd12d2), so another one is basically free.
> pg_collation.collversion has been used so far for the sort part
> of the collations.
Hmm, we haven't particularly drawn a distinction between sort-related
and not-sort-related aspects of collation versions AFAIK. Perhaps
it'd be appropriate to do so, and I agree that there's not time to
design such a thing for v17. But pg_c_utf8 might be the only case
where we could do anything other than advance those versions in
lockstep. I doubt we have enough insight into the behaviors of
other providers to say confidently that an update affects only
one side of their behavior.
regards, tom lane