Re: [PATCH v3] diff: make diff_free_filespec_data accept NULL

From: Johannes Schindelin <johannes.schindelin@gmx.de>
Date: Nov 10, 2020, 12:08 UTC
Message-ID: <nycvar.QRO.7.76.6.2011101257540.18437@tvgsbejvaqbjf.bet>
In-Reply-To: <a096d122-52a3-700a-3a14-30a81b099cd8@theori.io>

Hi Jinoh,

On Fri, 6 Nov 2020, Jinoh Kang wrote:

> Commit 3aef54e8b8 ("diff: munmap() file contents before running external
> diff") introduced calls to diff_free_filespec_data in
> run_external_diff, which may pass NULL pointers.

Sorry for the breakage!

Maybe the commit message could talk a bit about the circumstances when this happens? Of course, I (and every other reader...) could analyze your patch to guess what it is that triggers the bug, but it's really a good use of the commit message to describe it in plain English.

Show 5 quoted lines

>
> Fix this and prevent any such bugs in the future by making
> `diff_free_filespec_data(NULL)` a no-op.
>
> Fixes: 3aef54e8b8 ("diff: munmap() file contents before running external diff")

I am unaware of any other commit having a `Fixes:` trailer in the commit message. In any case, I would have expected `Fixes:` to mention a ticket or a bug report, not the commit that fixed _a separate_ issue (but unfortunately introduced this regression).

> Signed-off-by: Jinoh Kang <luke1337@theori.io>

Show 18 quoted lines

> ---
>  diff.c              |  3 +++
>  t/t7800-difftool.sh | 23 +++++++++++++++++++++++
>  2 files changed, 26 insertions(+)
>
> diff --git a/diff.c b/diff.c
> index d24f47df99..ace4a1d387 100644
> --- a/diff.c
> +++ b/diff.c
> @@ -4115,6 +4115,9 @@ void diff_free_filespec_blob(struct diff_filespec *s)
>
>  void diff_free_filespec_data(struct diff_filespec *s)
>  {
> +	if (!s)
> +		return;
> +
>  	diff_free_filespec_blob(s);
>  	FREE_AND_NULL(s->cnt_data);

We can write this in a more canonical way without wasting all that many lines:

	if (s) {
		diff_free_filespec_blob(s);
		FREE_AND_NULL(s->cnt_data);
	}

Show 14 quoted lines

>  }
> diff --git a/t/t7800-difftool.sh b/t/t7800-difftool.sh
> index 524f30f7dc..e9391abb54 100755
> --- a/t/t7800-difftool.sh
> +++ b/t/t7800-difftool.sh
> @@ -728,6 +728,29 @@ test_expect_success 'add -N and difftool -d' '
>  	git difftool --dir-diff --extcmd ls
>  '
>
> +test_expect_success 'difftool --cached with unmerged files' '
> +	test_when_finished git reset --hard &&
> +	echo base >file &&
> +	git add file &&
> +	git commit -m base &&

This does not advance the committer date. Let's just use the helper function we invented to make this much easier:

	test_commit base

This has also the advantage of already tagging the outcome.

Show 16 quoted lines

> +	git checkout -B conflict-a &&
> +	git checkout -B conflict-b &&
> +	git checkout conflict-a &&
> +	echo conflict-a >>file &&
> +	git add file &&
> +	git commit -m conflict-a &&
> +	git checkout conflict-b &&
> +	echo conflict-b >>file &&
> +	git add file &&
> +	git commit -m conflict-b &&
> +	git checkout master &&
> +	git merge conflict-a &&
> +	test_must_fail git merge conflict-b &&
> +	: >expect &&
> +	git difftool --cached --no-prompt >actual &&
> +	test_cmp expect actual

Shouldn't this use the `test_must_be_empty` function instead?

How about writing the test case this way:

test_expect_success 'difftool --cached with unmerged files' '
	test_when_finished git reset --hard &&

	test_commit conflicting &&
	test_commit conflict-a a conflicting.t &&
	git reset --hard conflicting &&
	test_commit conflict-b b conflicting.t &&
	test_must_fail git merge conflict-a &&

	git difftool --cached --no-prompt >out &&
	test_must_be_empty out
'

Ciao, Dscho

Show 9 quoted lines

> +'
> +
>  test_expect_success 'outside worktree' '
>  	echo 1 >1 &&
>  	echo 2 >2 &&
> --
> 2.26.2
>
>