Jonathan Tan <jonathanta...@google.com> writes:
>> Just a note (and a request-for-sanity-check) and not meant to be a
>> request to update the code, but with a still-in-pu 4b757a40 ("Use
>> remote_odb_get_direct() and has_remote_odb()", 2018-08-02) in
>> flight, repository_format_partial_clone is now gone.
>> ...
> Thanks for the heads-up. This makes me realize that even the current
> precondition (repository_format_partial_clone) is not an exact one - I
> should only be doing this if I know that there is at least one promisor
> object (if not, in the rare case that a repo is configured to be partial
> but does not have any promisor objects, repack will generate an empty
> packfile). In the next reroll, I'll take care of this, which should
> avoid this merge issue too.
I think that it is a good change, regardless of the semantic
conflict with any other topic, to look at the .promisor packs and
not look at the partial clone extension for this "do we want to
coalesce .promisor packs into one?" application. You only want to
do so when you have two or more such packs; being in partial clone
might be a preconditon for having such packs, but it is overly loose.
In any case, remote-odb topic is turning out to be more trouble than
I originally thought; in the same change that removes the variable,
it actually removes the support for a repository configured with a
partial clone extension, and causes a hard error in t0410, breaking
test of 'pu'. I do not know how many people actually use partial
clone right now, but their repositories would be broken the same way
when they update Git, if we merge that topic in its current shape.
I am dropping it from 'pu' for today's integration cycle.
