Re: [PATCH v2 08/12] submodule--helper: report "submodule" as our name in "-h" output

["Ævar Arnfjörð Bjarmason" <avarab@gmail.com>]

On Tue, Jun 14 2022, Glen Choo wrote:

> Ævar Arnfjörð Bjarmason <avarab@gmail.com> writes:
>
>> diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
>> index 574d6e0a79b..c2f55779cb1 100644
>> --- a/builtin/submodule--helper.c
>> +++ b/builtin/submodule--helper.c
>> @@ -286,7 +286,7 @@ static int module_list(int argc, const char **argv, const char *prefix)
>>  	};
>>  
>>  	const char *const git_submodule_helper_usage[] = {
>> -		N_("git submodule--helper list [--prefix=<path>] [<path>...]"),
>> +		N_("git submodule list [--prefix=<path>] [<path>...]"),
>>  		NULL
>>  	};
>> @@ -444,7 +444,7 @@ static int module_foreach(int argc, const char **argv, const char *prefix)
>>  	};
>>  
>>  	const char *const git_submodule_helper_usage[] = {
>> -		N_("git submodule--helper foreach [--quiet] [--recursive] [--] <command>"),
>> +		N_("git submodule foreach [--quiet] [--recursive] [--] <command>"),
>>  		NULL
>>  	};
>>  
>> @@ -582,7 +582,7 @@ static int module_init(int argc, const char **argv, const char *prefix)
>>  	};
>>  
>>  	const char *const git_submodule_helper_usage[] = {
>> -		N_("git submodule--helper init [<options>] [<path>]"),
>> +		N_("git submodule init [<options>] [<path>]"),
>>  		NULL
>>  	};
>>  
>> @@ -786,7 +786,7 @@ static int module_name(int argc, const char **argv, const char *prefix)
>>  	const struct submodule *sub;
>>  
>>  	if (argc != 2)
>> -		usage(_("git submodule--helper name <path>"));
>> +		usage(_("git submodule name <path>"));
>>  
>>  	sub = submodule_from_path(the_repository, null_oid(), argv[1]);
>>  
>> @@ -1185,7 +1185,7 @@ static int module_summary(int argc, const char **argv, const char *prefix)
>>  	};
>>  
>>  	const char *const git_submodule_helper_usage[] = {
>> -		N_("git submodule--helper summary [<options>] [<commit>] [--] [<path>]"),
>> +		N_("git submodule summary [<options>] [<commit>] [--] [<path>]"),
>>  		NULL
>>  	};
>>  
>> @@ -1349,7 +1349,7 @@ static int module_sync(int argc, const char **argv, const char *prefix)
>>  	};
>>  
>>  	const char *const git_submodule_helper_usage[] = {
>> -		N_("git submodule--helper sync [--quiet] [--recursive] [<path>]"),
>> +		N_("git submodule sync [--quiet] [--recursive] [<path>]"),
>>  		NULL
>>  	};
>>  
>> @@ -1789,7 +1789,7 @@ static int module_clone(int argc, const char **argv, const char *prefix)
>>  	};
>>  
>>  	const char *const git_submodule_helper_usage[] = {
>> -		N_("git submodule--helper clone [--prefix=<path>] [--quiet] "
>> +		N_("git submodule clone [--prefix=<path>] [--quiet] "
>>  		   "[--reference <repository>] [--name <name>] [--depth <depth>] "
>>  		   "[--single-branch] [--filter <filter-spec>] "
>>  		   "--url <url> --path <path>"),
>> @@ -2787,7 +2787,7 @@ static int absorb_git_dirs(int argc, const char **argv, const char *prefix)
>>  	};
>>  
>>  	const char *const git_submodule_helper_usage[] = {
>> -		N_("git submodule--helper absorbgitdirs [<options>] [<path>...]"),
>> +		N_("git submodule absorbgitdirs [<options>] [<path>...]"),
>>  		NULL
>>  	};
>>  
>> @@ -2851,9 +2851,9 @@ static int module_config(int argc, const char **argv, const char *prefix)
>>  		OPT_END()
>>  	};
>>  	const char *const git_submodule_helper_usage[] = {
>> -		N_("git submodule--helper config <name> [<value>]"),
>> -		N_("git submodule--helper config --unset <name>"),
>> -		"git submodule--helper config --check-writeable",
>> +		N_("git submodule config <name> [<value>]"),
>> +		N_("git submodule config --unset <name>"),
>> +		"git submodule config --check-writeable",
>>  		NULL
>>  	};
>>  
>> @@ -2892,7 +2892,7 @@ static int module_set_url(int argc, const char **argv, const char *prefix)
>>  		OPT_END()
>>  	};
>>  	const char *const usage[] = {
>> -		N_("git submodule--helper set-url [--quiet] <path> <newurl>"),
>> +		N_("git submodule set-url [--quiet] <path> <newurl>"),
>>  		NULL
>>  	};
>>  
>> @@ -2931,8 +2931,8 @@ static int module_set_branch(int argc, const char **argv, const char *prefix)
>>  		OPT_END()
>>  	};
>>  	const char *const usage[] = {
>> -		N_("git submodule--helper set-branch [-q|--quiet] (-d|--default) <path>"),
>> -		N_("git submodule--helper set-branch [-q|--quiet] (-b|--branch) <branch> <path>"),
>> +		N_("git submodule set-branch [-q|--quiet] (-d|--default) <path>"),
>> +		N_("git submodule set-branch [-q|--quiet] (-b|--branch) <branch> <path>"),
>>  		NULL
>>  	};
>>  
>> @@ -2973,7 +2973,7 @@ static int module_create_branch(int argc, const char **argv, const char *prefix)
>>  		OPT_END()
>>  	};
>>  	const char *const usage[] = {
>> -		N_("git submodule--helper create-branch [-f|--force] [--create-reflog] [-q|--quiet] [-t|--track] [-n|--dry-run] <name> <start-oid> <start-name>"),
>> +		N_("git submodule create-branch [-f|--force] [--create-reflog] [-q|--quiet] [-t|--track] [-n|--dry-run] <name> <start-oid> <start-name>"),
>>  		NULL
>>  	};
>>  
>> @@ -3276,7 +3276,7 @@ static int module_add(int argc, const char **argv, const char *prefix)
>>  	};
>>  
>>  	const char *const usage[] = {
>> -		N_("git submodule--helper add [<options>] [--] <repository> [<path>]"),
>> +		N_("git submodule add [<options>] [--] <repository> [<path>]"),
>>  		NULL
>>  	};

First, I think this patch is premature, so let's just drop it in a
re-roll. We don't *need* to fix this particular edge case or bug now,
and can just do it as we do the final migration.

> I don't feel positive about pretending to be "git submodule" when the
> subcommand is internal-only. Such commands are only invoked as "git
> submodule--helper foo" and we will probably never turn them into
> external-facing "git submodule foo" [1]. Even if it is internal-only,
> there is a debugging benefit to emitting a usage string with the
> correct, non-user facing command name.

I think we should really be most concerned about the end users here,
nobody except git developers are going to want to invoke
"submodule--helper" directly, so its help messages should be aimed at
helping users.

As you mention in another E-Mail it's currently not a big deal, I
*think* that "absorbgitdirs" is the only case where we'll implicitly
suggest to a user that they might invoke "git submodule--helper" as we
emit the usage info.

But I think for the patches that come after this we should definitely
change this, i.e. we don't want an in-between state where we have both
of builtin/submodule{,--helper}.c because we haven't merged the two yet,
but are now telling users about the inernal-only helper.

> I admit it might be a bit confusing to have a mix of usage strings with
> "submodule--helper" vs "submodule", but as long as we clearly label
> which commands are user-facing and which aren't, it shouldn't be too
> hard to maintain, e.g. this might just be a matter of adding comments to
> the array of commands like:
>
>   static struct cmd_struct commands[] = {
>     /* User facing commands - pretend to be "git submodule" */
>     {"name", module_name},
>     {"add", module_add},
>     [...]
>     /* Internal-only commands */
>     {"clone", module_clone},
>     {"resolve-relative-url-test", resolve_relative_url_test},
>     [...]
>   };
>
> Notably, even with this patch we still _kind of_ have 'pseudo usage
> strings' that reference "git submodule--helper". These are just
> internal-only commands that don't use the parse_options() API, e.g.
>
> 	if (argc < 3)
> 		die("submodule--helper push-check requires at least 2 arguments");
>
> I think it's ok to leave them as-is, provided we also leave the other
> internal-only commands as-is.

Yes let's leave them, we'll be converting these to not invoke
sub-processes sooner than later anyway.

> [1] For convenience, here is a list of all of the subcommands changed in
> this patch and whether they are user-facing/internal-only:
>
> User-facing                 Internal-only
>
> - foreach                   - list
> - init                      - name
> - summary                   - clone
> - sync                      - config
> - absorbgitdirs             - create-branch
> - set-url
> - set-branch
> - add

But yeah, if/when we change the public-facing usage strings we should be
leaving out the "internal-only" commands, thanks. I just did a dumb
search/replace.