On Wed, May 07, 2025 at 09:38:52AM -0700, Junio C Hamano wrote: > Derrick Stolee <stolee@xxxxxxxxx> writes: > > > On 5/6/25 7:09 AM, Patrick Steinhardt wrote: > >> All of the external functions provided by the object database subsystem > >> don't depend on `the_repository` anymore, but some internal functions > >> still do. Refactor those cases by plumbing through the repository that > >> owns the object database. > >> This change allows us to get rid of the > >> `USE_THE_REPOSITORY_VARIABLE` > >> preprocessor define. > > > >> --- a/odb.c > >> +++ b/odb.c > >> @@ -1,5 +1,3 @@ > >> -#define USE_THE_REPOSITORY_VARIABLE > >> - > > > > Very satisfying! Thanks, > > Yes, nice outcome. I looked at the resulting odb.c and was a bit > unhappy to see that we still need to pass "struct repository *" > (instead of "struct object_database *") around. But that is not > because we assume we can get to an odb via repo, but primarily > because repo has many things like config access and hash-algo that > we rely on, so it is perfectly OK. > > At some places, where we pass odb around, we can and do go from it > to its repository (e.g. odb->repo->hash_algo), which was what I am > expecting to see more of in the future. Yup. Over time I very much have the intent to reduce the reliance on `odb->repo` even further. Ideally, the object database should have all info, or at least almost all of it, to function correctly. Patrick
