From: "Ota, Takaaki" <Takaaki.Ota@am.sony.com>
To: <ofv@wanadoo.es>
Cc: lekktu@gmail.com, 10733@debbugs.gnu.org
Subject: bug#10733: 24.0.93; w32 file truncation
Date: Mon, 6 Feb 2012 15:43:27 -0800 [thread overview]
Message-ID: <20120206.154327.522668003.Takaaki.Ota@am.sony.com> (raw)
In-Reply-To: <87fwen4k6s.fsf@wanadoo.es>
Don't you need to dispose fh properly?
-Tak
Mon, 6 Feb 2012 15:27:07 -0800: Óscar Fuentes <ofv@wanadoo.es> wrote:
> Eli Zaretskii <eliz@gnu.org> writes:
>
> [snip]
>
> > No. The `stat' Emacs uses is in w32.c. What you see on
> > lib-src/ntlib.c is just a minimal subset of what we have in w32.c.
>
> That's important. Thanks.
>
> >> Symlinks are detected and handled specially on MSVCRT's stat. In
> >> aessence, for symlinks it uses fstat.
> >
> > But fstat probably calls GetFileInformationByHandle under the hood,
> > and we already call that function in w32.c:stat. So maybe the fix is
> > not as ugly and inelegant as you thought.
>
> Yup. This patch fixes the problem:
>
> diff --git a/src/w32.c b/src/w32.c
> index f610a36..035e1f2 100644
> --- a/src/w32.c
> +++ b/src/w32.c
> @@ -3444,6 +3444,29 @@ stat (const char * path, struct stat * buf)
> }
> }
>
> + buf->st_size = 0;
> +
> + if ((wfd.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) &&
> + (wfd.dwReserved0 == IO_REPARSE_TAG_SYMLINK))
> + {
> + HANDLE fh;
> + BY_HANDLE_FILE_INFORMATION info;
> +
> + if ((fh = CreateFile (name, 0, 0, NULL, OPEN_EXISTING,
> + FILE_FLAG_BACKUP_SEMANTICS, NULL))
> + && GetFileInformationByHandle (fh, &info))
> + {
> + buf->st_size = info.nFileSizeHigh;
> + buf->st_size <<= 32;
> + buf->st_size += info.nFileSizeLow;
> + }
> + else
> + {
> + errno = ENOENT;
> + return -1;
> + }
> + }
> +
> if (!(NILP (Vw32_get_true_file_attributes)
> || (EQ (Vw32_get_true_file_attributes, Qlocal) && is_slow_fs (name)))
> /* No access rights required to get info. */
> @@ -3532,9 +3555,12 @@ stat (const char * path, struct stat * buf)
> buf->st_dev = volume_info.serialnum;
> buf->st_rdev = volume_info.serialnum;
>
> - buf->st_size = wfd.nFileSizeHigh;
> - buf->st_size <<= 32;
> - buf->st_size += wfd.nFileSizeLow;
> + if (! buf->st_size)
> + {
> + buf->st_size = wfd.nFileSizeHigh;
> + buf->st_size <<= 32;
> + buf->st_size += wfd.nFileSizeLow;
> + }
>
> /* Convert timestamps to Unix format. */
> buf->st_mtime = convert_time (wfd.ftLastWriteTime);
>
>
> Maybe it can be integrated in the
>
> if (!(NILP(Vw32_get_true_file_attributes) ...
>
> hence reusing the calls to CreateFile and GetFileInformationByHandle and
> shortening the patch, but as I don't know what
> Vw32_get_true_file_attributes does, preferread to follow the safe way.
>
> However, we still have another implementation on ntlib.c.
>
> And the fix is just for the size. Don't know if there are other
> attributes suffer from the same problem. Possibly the right thing is to
> do what MSVCRT does:
>
> if is-symlink?
> use fstat
> fi
>
next prev parent reply other threads:[~2012-02-06 23:43 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-02-05 22:34 bug#10733: 24.0.93; w32 file truncation Ota, Takaaki
2012-02-05 23:51 ` Juanma Barranquero
2012-02-06 0:16 ` Ota, Takaaki
2012-02-06 4:05 ` Eli Zaretskii
2012-02-06 5:25 ` Óscar Fuentes
2012-02-06 16:16 ` Óscar Fuentes
2012-02-06 17:21 ` Eli Zaretskii
2012-02-06 17:58 ` Óscar Fuentes
2012-02-06 18:37 ` Eli Zaretskii
2012-02-06 20:19 ` Ota, Takaaki
2012-02-06 20:23 ` Lennart Borgman
2012-02-06 21:11 ` Eli Zaretskii
2012-02-06 21:20 ` Lennart Borgman
2012-02-06 21:31 ` Eli Zaretskii
2012-02-06 21:35 ` Lennart Borgman
2012-02-06 20:24 ` Lars Ingebrigtsen
2012-02-06 21:09 ` Eli Zaretskii
2012-02-06 21:08 ` Eli Zaretskii
2012-02-06 23:27 ` Óscar Fuentes
2012-02-06 23:43 ` Ota, Takaaki [this message]
2012-02-07 0:07 ` Óscar Fuentes
2012-02-07 4:02 ` Eli Zaretskii
2012-02-07 5:22 ` Óscar Fuentes
2012-02-07 17:35 ` Eli Zaretskii
2012-08-03 10:59 ` Eli Zaretskii
2012-02-06 16:55 ` Eli Zaretskii
2012-02-06 0:20 ` Óscar Fuentes
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20120206.154327.522668003.Takaaki.Ota@am.sony.com \
--to=takaaki.ota@am.sony.com \
--cc=10733@debbugs.gnu.org \
--cc=lekktu@gmail.com \
--cc=ofv@wanadoo.es \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this external index
https://git.savannah.gnu.org/cgit/emacs.git
https://git.savannah.gnu.org/cgit/emacs/org-mode.git
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.