On 04/11, Stefan Beller wrote:
> +int has_submodules(unsigned what_to_check)
> +{
> + if (what_to_check & SUBMODULE_CHECK_ANY_CONFIG) {
> + if (submodule_config_reading == SUBMODULE_CONFIG_NOT_READ)
> + load_submodule_config();
> + if (submodule_config_reading == SUBMODULE_CONFIG_EXISTS)
> + return 1;
> + }
> +
> + if ((what_to_check & SUBMODULE_CHECK_ABSORBED_GIT_DIRS) &&
> + file_exists(git_path("modules")))
> + return 1;
> +
> + if ((what_to_check & SUBMODULE_CHECK_GITMODULES_IN_WT) &&
> + (!is_bare_repository() && file_exists(".gitmodules")))
> + return 1;
> +
> + if (what_to_check & SUBMODULE_CHECK_GITLINKS_IN_TREE) {
> + int i;
> +
> + if (read_cache() < 0)
> + die(_("index file corrupt"));
> +
> + for (i = 0; i < active_nr; i++)
> + if (S_ISGITLINK(active_cache[i]->ce_mode))
> + return 1;
> + }
> +
> + return 0;
> +}
It may be a good idea to rearrange these by order to correctness.
Correctness may not be the best way to describe it, but which is the
strongest indicator that there is a submodule or that a repo 'has a
submodule'. That way in the future we could have a #define that is
SUBMODULE_CHECK_ANY or ALL or something....Now that I'm thinking harder
about that we may not want that, and just require explicitly stating
which check you want done.
Anyways good looking patch, and I like the idea of consolidating the
checks into a single function.
> diff --git a/submodule.h b/submodule.h
> index 8a8bc49dc9..5ec72fbb16 100644
> --- a/submodule.h
> +++ b/submodule.h
> @@ -1,6 +1,12 @@
> #ifndef SUBMODULE_H
> #define SUBMODULE_H
>
> +#define SUBMODULE_CHECK_ANY_CONFIG (1<<0)
> +#define SUBMODULE_CHECK_ABSORBED_GIT_DIRS (1<<1)
> +#define SUBMODULE_CHECK_GITMODULES_IN_WT (1<<2)
> +#define SUBMODULE_CHECK_GITLINKS_IN_TREE (1<<3)
> +int has_submodules(unsigned what_to_check);
--
Brandon Williams