Re: [PATCH v2] clone: error specifically with --local and symlinked objects

From: Junio C Hamano <gitster@pobox.com>
Date: Apr 10, 2023, 22:27 UTC
Message-ID: <xmqqttxncr0k.fsf@gitster.g>
In-Reply-To: <pull.1488.v2.git.git.1681165130765.gitgitgadget@gmail.com>

"Glen Choo via GitGitGadget" <gitgitgadget@gmail.com> writes:

Show 9 quoted lines

>      ++		if (errno == ENOTDIR) {
>      ++			int saved_errno = errno;
>      ++			struct stat st;
>      ++			if (lstat(src->buf, &st) == 0 && S_ISLNK(st.st_mode))
>      ++				die(_("'%s' is a symlink, refusing to clone with --local"),
>      ++				    src->buf);
>      ++			errno = saved_errno;
>      ++		}
>        		die_errno(_("failed to start iterator over '%s'"), src->buf);

I doubt you need saved_errno in the code immediately after this patch gets applied, as what you are saving is guaranteed to be ENOTDIR so you can just restore it before you fall through out of the block.

It however would not hurt, though. Especially if the condition that guarding this block is likely to change in the future, we can view the apparent waste as paying insurance premium to protect from future breakages.

Will queue.  Thanks.