Re: [PATCH v2 3/3] builtin/add: error out when passing untracked path with -u

From: Junio C Hamano <gitster@pobox.com>
Date: Mar 29, 2024, 21:43 UTC
Message-ID: <xmqqzfugvhnf.fsf@gitster.g>
In-Reply-To: <20240329205649.1483032-5-shyamthakkar001@gmail.com>

Ghanshyam Thakkar <shyamthakkar001@gmail.com> writes:

Show 7 quoted lines

> When passing untracked path with -u option, it silently succeeds. There
> is no error message and the exit code is zero. This is inconsistent
> with other instances of git commands where the expected argument is a
> known path. In those other instances, we error out when the path is
> not known.
>
> Therefore, fix this by passing a character array to

"Therefore, fix" -> "Fix".

Show 28 quoted lines

> add_files_to_cache() to collect the pathspec matching information and
> report the error if a pathspec does not match any cache entry. Also add
> a testcase to cover this scenario.
>
> Signed-off-by: Ghanshyam Thakkar <shyamthakkar001@gmail.com>
> ---
>  builtin/add.c         | 9 ++++++++-
>  t/t2200-add-update.sh | 6 ++++++
>  2 files changed, 14 insertions(+), 1 deletion(-)
>
> diff --git a/builtin/add.c b/builtin/add.c
> index ffe5fd8d44..650432bb13 100644
> --- a/builtin/add.c
> +++ b/builtin/add.c
> @@ -370,6 +370,7 @@ int cmd_add(int argc, const char **argv, const char *prefix)
>  	int add_new_files;
>  	int require_pathspec;
>  	char *seen = NULL;
> +	char *ps_matched = NULL;
>  	struct lock_file lock_file = LOCK_INIT;
>  
>  	git_config(add_config, NULL);
> @@ -547,15 +548,20 @@ int cmd_add(int argc, const char **argv, const char *prefix)
>  		string_list_clear(&only_match_skip_worktree, 0);
>  	}
>  
> +
>  	begin_odb_transaction();

Unnecessary change.

Show 11 quoted lines

> +	ps_matched = xcalloc(pathspec.nr, 1);
>  	if (add_renormalize)
>  		exit_status |= renormalize_tracked_files(&pathspec, flags);
>  	else
>  		exit_status |= add_files_to_cache(the_repository, prefix,
> -						  &pathspec, NULL,
> +						  &pathspec, ps_matched,
>  						  include_sparse, flags);
>  
> +	if (take_worktree_changes)
> +		exit_status |= report_path_error(ps_matched, &pathspec);

Hmph, are we sure take_worktree_changes is true only when add_renormalize is false?

>  	if (add_new_files)
>  		exit_status |= add_files(&dir, flags);

If report_path_error() detected that the pathspec were faulty, should we still proceed to add new files? This is NOT a rhetorical question, as I do not know the answer myself. I do not even know offhand what add_files_to_cache() above did when pathspec elements are not all consumed---if it does not complain and does not refrain from doing any change to the index, then we should follow suite and add_files() here, too.

Show 25 quoted lines

> @@ -568,6 +574,7 @@ int cmd_add(int argc, const char **argv, const char *prefix)
>  			       COMMIT_LOCK | SKIP_IF_UNCHANGED))
>  		die(_("unable to write new index file"));
>  
> +	free(ps_matched);
>  	dir_clear(&dir);
>  	clear_pathspec(&pathspec);
>  	return exit_status;
> diff --git a/t/t2200-add-update.sh b/t/t2200-add-update.sh
> index c01492f33f..7cba325f08 100755
> --- a/t/t2200-add-update.sh
> +++ b/t/t2200-add-update.sh
> @@ -65,6 +65,12 @@ test_expect_success 'update did not touch untracked files' '
>  	test_must_be_empty out
>  '
>  
> +test_expect_success 'error out when passing untracked path' '
> +	echo content >baz &&
> +	test_must_fail git add -u baz 2>err &&
> +	test_grep -e "error: pathspec .baz. did not match any file(s) known to git" err
> +'
> +
>  test_expect_success 'cache tree has not been corrupted' '
>  
>  	git ls-files -s |