From: Karthik Nayak <karthik.188@gmail.com>
Date: Mon, 16 Mar 2026 11:14:15 GMT
Subject: Re: [GSoC PATCH] apply: report the location of corrupt patches
Message-ID: <CAOLa=ZTpfHUySnMgCFMnvo2JcRSv8zqFP-cLFSs+Ab5Cy2zsvg@mail.gmail.com>
In-Reply-To: <20260315231538.68586-1-jerrywang183@yahoo.com>

Jialong Wang <jerrywang183@yahoo.com> writes:

> When parsing a corrupt patch, git apply reports only the line number.
> That does not tell the user which input the line number refers to.
>
> Include the patch input path in the error message so the reported
> location is easier to use.
>

Definitely a welcome change.

> Add tests for both file input and standard input.
>
> Signed-off-by: Jialong Wang <jerrywang183@yahoo.com>
> ---
>  apply.c               |  3 ++-
>  t/t4100-apply-stat.sh | 16 +++++++++++++++-
>  2 files changed, 17 insertions(+), 2 deletions(-)
>
> diff --git a/apply.c b/apply.c
> index b6dd1066a0..904d1c3e55 100644
> --- a/apply.c
> +++ b/apply.c
> @@ -1875,7 +1875,8 @@ static int parse_single_patch(struct apply_state *state,
>  		len = parse_fragment(state, line, size, patch, fragment);
>  		if (len <= 0) {
>  			free(fragment);
> -			return error(_("corrupt patch at line %d"), state->linenr);
> +			return error(_("corrupt patch at %s:%d"),
> +				     state->patch_input_file, state->linenr);

Okay so `prase_single_patch()` is called by `parse_chunk()` which is
called by `apply_patch()`, which unconditionally sets
`state->patch_input_file`. So this looks good.

>  		}
>  		fragment->patch = line;
>  		fragment->size = len;
> diff --git a/t/t4100-apply-stat.sh b/t/t4100-apply-stat.sh
> index a5664f3eb3..f99e439688 100755
> --- a/t/t4100-apply-stat.sh
> +++ b/t/t4100-apply-stat.sh
> @@ -48,7 +48,21 @@ test_expect_success 'applying a hunk header which overflows fails' '
>  	+b
>  	EOF
>  	test_must_fail git apply patch 2>err &&
> -	echo "error: corrupt patch at line 4" >expect &&
> +	echo "error: corrupt patch at patch:4" >expect &&
> +	test_cmp expect err
> +'
> +
> +test_expect_success 'applying a hunk header which overflows from stdin fails' '
> +	cat >patch <<-\EOF &&
> +	diff -u a/file b/file
> +	--- a/file
> +	+++ b/file
> +	@@ -98765432109876543210 +98765432109876543210 @@
> +	-a
> +	+b
> +	EOF
> +	test_must_fail git apply <patch 2>err &&
> +	echo "error: corrupt patch at <stdin>:4" >expect &&
>  	test_cmp expect err
>  '
>  test_done
> --
> 2.51.0

Nit: It would also be nice to see that this does work with multiple
patch inputs, where one of them is corrupted.