Re: [PATCH v4 3/3] submodule--helper: introduce add-config subcommand

[Atharva Raykar <raykar.ath@gmail.com>]

On 15-Jun-2021, at 01:21, Rafael Silva <rafaeloliveira.cs@gmail.com> wrote:
> 
> 
> Atharva Raykar <raykar.ath@gmail.com> writes:
> 
>> ---
>> builtin/submodule--helper.c | 119 ++++++++++++++++++++++++++++++++++++
>> git-submodule.sh            |  28 +--------
>> 2 files changed, 120 insertions(+), 27 deletions(-)
>> 
> 
> I do not have enough expertise to judge the entire content of this
> patch. I would like, however, to propose a slight code change for the
> sake of readability.
> 
>> diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
>> index 6dffaeb6cb..c4b2aa6537 100644
>> --- a/builtin/submodule--helper.c
>> +++ b/builtin/submodule--helper.c
>> @@ -2935,6 +2935,124 @@ static int add_clone(int argc, const char **argv, const char *prefix)
>> 	return 0;
>> }
>> 
>> +static void configure_added_submodule(struct add_data *add_data)
>> +{
>> +	char *key, *submod_pathspec = NULL;
>> +	struct child_process add_submod = CHILD_PROCESS_INIT;
>> +	struct child_process add_gitmodules = CHILD_PROCESS_INIT;
>> +	int pathspec_key_exists, activate = 0;
>> +
>> +	key = xstrfmt("submodule.%s.url", add_data->sm_name);
>> +	git_config_set_gently(key, add_data->realrepo);
>> +	free(key);
>> +
>> +	add_submod.git_cmd = 1;
>> +	strvec_pushl(&add_submod.args, "add",
>> +		     "--no-warn-embedded-repo", NULL);
>> +	if (add_data->force)
>> +		strvec_push(&add_submod.args, "--force");
>> +	strvec_pushl(&add_submod.args, "--", add_data->sm_path, NULL);
>> +
>> +	if (run_command(&add_submod))
>> +		die(_("Failed to add submodule '%s'"), add_data->sm_path);
>> +
>> +	key = xstrfmt("submodule.%s.path", add_data->sm_name);
>> +	config_set_in_gitmodules_file_gently(key, add_data->sm_path);
>> +	free(key);
> 
> This above three lines of code is very similar to the two operations that
> follows (including the one inside the `if (add_data->branch)`
> condition. So [ ... ]
> 
>> +	key = xstrfmt("submodule.%s.url", add_data->sm_name);
>> +	config_set_in_gitmodules_file_gently(key, add_data->repo);
>> +	free(key);
>> +	if (add_data->branch) {
>> +		key = xstrfmt("submodule.%s.branch", add_data->sm_path);
>> +		config_set_in_gitmodules_file_gently(key, add_data->branch);
>> +		free(key);
>> +	}
>> +
> 
> [ ... ] it might be worth to write a small wrapper that will perform: (1)
> `xstrfmt()` on the specified config section, (2) set the configuration
> in the file and (3) free()'ing the variable inside the wrapper. Thus,
> most of these code will become one liners that is easier to read (given
> the function is properly named :) ).
> 
> After abstracting the code on the wrapper, this code will become
> something like:
> 
> 
>    function_properly_named("submodule.%s.path", add_data->sm_name, add_data->sm_path);
>    function_properly_named("submodule.%s.url", add_data->sm_name, add_data->repo);
>    if (add_data->branch)
>         function_properly_named("submodule.%s.branch", add_data->sm_path, add_data->branch);
> 
> 
> Just as an example, here's a diff to demonstrate the argument:
> 
> -- >8 --
> diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
> index 65f79fbd53..48ea909f51 100644
> --- a/builtin/submodule--helper.c
> +++ b/builtin/submodule--helper.c
> @@ -2934,6 +2934,14 @@ static int add_clone(int argc, const char **argv, const char *prefix)
> 	return 0;
> }
> 
> +void add_config_in_submodules_file(const char *keyfmt, const char *submodule,
> +			  const char *value)
> +{
> +	char *key = xstrfmt(keyfmt, submodule);
> +	config_set_in_gitmodules_file_gently(key, value);
> +	free(key);
> +}
> +
> static void configure_added_submodule(struct add_data *add_data)
> {
> 	char *key, *submod_pathspec = NULL;
> @@ -2955,17 +2963,10 @@ static void configure_added_submodule(struct add_data *add_data)
> 	if (run_command(&add_submod))
> 		die(_("Failed to add submodule '%s'"), add_data->sm_path);
> 
> -	key = xstrfmt("submodule.%s.path", add_data->sm_name);
> -	config_set_in_gitmodules_file_gently(key, add_data->sm_path);
> -	free(key);
> -	key = xstrfmt("submodule.%s.url", add_data->sm_name);
> -	config_set_in_gitmodules_file_gently(key, add_data->repo);
> -	free(key);
> -	if (add_data->branch) {
> -		key = xstrfmt("submodule.%s.branch", add_data->sm_path);
> -		config_set_in_gitmodules_file_gently(key, add_data->branch);
> -		free(key);
> -	}
> +	add_config_in_submodules_file("submodule.%s.path", add_data->sm_name, add_data->sm_path);
> +	add_config_in_submodules_file("submodule.%s.url", add_data->sm_name, add_data->repo);
> +	if (add_data->branch)
> +		add_config_in_submodules_file("submodule.%s.branch", add_data->sm_path, add_data->branch);
> 
> 	add_gitmodules.git_cmd = 1;
> 	strvec_pushl(&add_gitmodules.args,
> 
> -- >8 --
> 
> A proper name than "add_config_in_submodules_file" should be considered - I'm
> not very good in naming things.
> 
> These change does (should) not change the behavior of code, even though
> I believe it make the code simpler to read, I do not have strong
> opinions about it. So, take this proposal as you wish. 

I agree with you, this will make the code simpler to read. It also
made me realise one thing that I did not replicate exactly from the
shell code.

The original shell code calls 'module_config()', which does an extra
check to see if writing to '.gitmodules' is okay. I did not perform
this check, and including that in the wrapper you propose will be a
good idea.

> -- 
> Thanks
> Rafael