Hi Duy,

On Thu, 27 Jun 2019, Nguyễn Thái Ngọc Duy wrote:

> Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
> ---

This commit is _awfully_ short given that...

> diff --git a/builtin/grep.c b/builtin/grep.c
> index 580fd38f41..560051784e 100644
> --- a/builtin/grep.c
> +++ b/builtin/grep.c
> @@ -458,7 +458,8 @@ static int grep_submodule(struct grep_opt *opt,
>  		object = parse_object_or_die(oid, oid_to_hex(oid));
>
>  		grep_read_lock();
> -		data = read_object_with_reference(&object->oid, tree_type,
> +		data = read_object_with_reference(&subrepo,
> +						  &object->oid, tree_type,

... this change and...

>  						  &size, NULL);
>  		grep_read_unlock();
>
> @@ -623,7 +624,8 @@ static int grep_object(struct grep_opt *opt, const struct pathspec *pathspec,
>  		int hit, len;
>
>  		grep_read_lock();
> -		data = read_object_with_reference(&obj->oid, tree_type,
> +		data = read_object_with_reference(opt->repo,
> +						  &obj->oid, tree_type,

... this change is totally not what would be intuitively the easiest: to
use `the_repository` in all built-ins.

It might take quite a lot of convincing that these changes are correct, in
particular in light of the regressions introduced by the first iteration
(to paraphrase Warren Buffet [*1*]: one slip in a patch series touching as
central parts as this one will need a lot of time to restore trust in
subsequent iterations' correctness.)

In short: with such an empty commit message, this patch is no good. It's
as if it was optimized to pass the test suite on Linux instead of a best
effort to make the conversion as correct as you can make it.

Ciao,
Johannes

Footnote *1*:
https://www.forbes.com/sites/jamesberman/2014/04/20/the-three-essential-warren-buffett-quotes-to-live-by/