From mboxrd@z Thu Jan  1 00:00:00 1970
From: Jason at zx2c4.com (Jason A. Donenfeld)
Date: Sun, 7 Apr 2013 13:29:47 +0200
Subject: [PATCH 06/19] scan-tree: use struct strbuf instead of static
 buffers
In-Reply-To: <def4c3318fb819a4713844befce838a10a3a99e1.1365326321.git.john@keeping.me.uk>
References: <cover.1365326321.git.john@keeping.me.uk>
 <def4c3318fb819a4713844befce838a10a3a99e1.1365326321.git.john@keeping.me.uk>
Message-ID: <CAHmME9pM-zZvhPzgdbzcAh3isWQsX0RF+NpDs9kP=We+kkg5dQ@mail.gmail.com>

On Sun, Apr 7, 2013 at 11:29 AM, John Keeping <john at keeping.me.uk> wrote:
>  static int is_git_dir(const char *path)
>  {
>         struct stat st;
> -       static char buf[MAX_PATH];
> +       struct strbuf pathbuf = STRBUF_INIT;
>
> -       if (snprintf(buf, MAX_PATH, "%s/objects", path) >= MAX_PATH) {
> -               fprintf(stderr, "Insanely long path: %s\n", path);
> -               return 0;
> -       }
> -       if (stat(buf, &st)) {
> +       strbuf_addf(&pathbuf, "%s/objects", path);
> +       if (stat(pathbuf.buf, &st)) {
>                 if (errno != ENOENT)
>                         fprintf(stderr, "Error checking path %s: %s (%d)\n",
>                                 path, strerror(errno), errno);

strbuf_release is never called. Does this leak?

Are there other cases of this elsewhere?