Atharva Raykar [off-list ref] writes:
+static char *get_default_remote_submodule(const char *module_path)
+{
+ const char *refname;
+ const struct submodule *sub;
+ struct repository subrepo;
+
+ refname = refs_resolve_ref_unsafe(get_submodule_ref_store(module_path),
+ "HEAD", 0, NULL, NULL);
+ sub = submodule_from_path(the_repository, null_oid(), module_path);
+ repo_submodule_init(&subrepo, the_repository, sub);
+ return repo_get_default_remote(&subrepo, refname);
+}
This call (and I think there is another call in this file) to
repo_submodule_init() is affected by what Jonathan's 8eb8dcf9
(repository: support unabsorbed in repo_submodule_init, 2021-09-09)
wants to do, namely to lose "struct submodule sub" as a parameter
and instead take the path to the module and the treeish name as
parameters to repo_submodule_init().
I _think_ I resolved the conflict correctly, but please double check
the result when it is pushed out later today, both of you.
Thanks.