> > I anticipate that we need to have a lot of back pointers to the repository > > in question, hence I think we should have the repository pointer promoted > > to not just a back pointer. > > I will probably need more time to study that commit and maybe the mail > archive for the history of this series. But if I remember correctly > some of these for_each_ api is quite a pain (perhaps it's the for_each > version of reflog?) and it's probably better to redesign it (again > talking without real understanding of the problem).
I stepped back a bit and reconsidered the point made above, and I do not think that the repository argument is any special. If you need a repository (for e.g. lookup_commit or friends), you'll have to pass it through the callback cookie, whether directly or as part of a struct tailored to your purpose. Instead we should strive to make the refs API smaller and cleaner, omitting the repository argument at all, and instead should be focussing on a ref_store argument instead. This series applies on master; when we decide to go this direction we can drop origin/sb/refs-in-repo. Thanks, Stefan Derrick Stolee (1): replace-objects: use arbitrary repositories Stefan Beller (1): refs: switch for_each_replace_ref back to use a ref_store builtin/replace.c | 4 +--- refs.c | 4 ++-- refs.h | 2 +- replace-object.c | 5 +++-- 4 files changed, 7 insertions(+), 8 deletions(-) -- 2.18.0.132.g195c49a2227
