Re: [PATCH v2 2/4] worktree: link worktrees with relative paths

From: Eric Sunshine <sunshine@sunshineco.com>
Date: Oct 7, 2024, 04:02 UTC
Message-ID: <CAPig+cQjHCZChToM_nEhmQsXedK4nEObs7YV2Q2PzrZqcHB4fQ@mail.gmail.com>
In-Reply-To: <ZwNZRX1LHlxawJJc@ArchLinux>

On Sun, Oct 6, 2024 at 11:45 PM shejialuo <shejialuo@gmail.com> wrote:

Show 9 quoted lines

> Actually, I somehow understand why in the original code, it will use
> "xstrdup_or_null" to initialize the "backlink". Because in
> "read_gitfile_gently", we will use a static "struct strbuf" as the
> buffer.
>
> I guess the intention of the original code is that if we call again
> "read_gitfile_gently" in this function or we have another thread which
> calls this function, the content of the buffer will be changed. So we
> explicitly use "xstrdup_or_null" to create a copy here to avoid this.

That's correct. The code is being defensive in case the static strbuf inside read_gitfile_gently() gets overwritten.

> But I wonder whether we really need to consider above problem?

It's easier to reason about the code in this function if we don't have to worry about the underlying static strbuf. Also, this is not a hot path so the extra allocation and string copy is not a concern. As such, it makes sense for the code to remain robust (by creating the copy) rather than becoming potentially more fragile by eliminating the string copy.