From: Matt Armstrong <matt@rfc20.org>
To: Eli Zaretskii <eliz@gnu.org>
Cc: 46397@debbugs.gnu.org, eggert@cs.ucla.edu, craven@gmx.net
Subject: bug#46397: 27.1; Cannot delete buffer pointing to a file in a path that includes a file
Date: Fri, 05 Mar 2021 14:19:53 -0800 [thread overview]
Message-ID: <87y2f1meeu.fsf@mdeb> (raw)
In-Reply-To: <837dmq95ee.fsf@gnu.org>
[-- Attachment #1: Type: text/plain, Size: 2293 bytes --]
Eli Zaretskii <eliz@gnu.org> writes:
>> From: Matt Armstrong <matt@rfc20.org>
>> Cc: 46397@debbugs.gnu.org, eggert@cs.ucla.edu, craven@gmx.net
>> Date: Wed, 24 Feb 2021 09:37:49 -0800
>>
>> I still like my original idea of calling display-warning for all unlock
>> errors, essentially turning "unlock" into a best effort function at the
>> API level. I think display-warning is intrusive enough that users are
>> unlikely to simply not notice the problem, and there are worse things
>> than leaving lock files around.
>
> OK, you've convinced me: let's try the warning approach. Can you
> present a patch for that, please?
Thanks Eli, will do. I have made progress with that idea, but I'd
rather get the test in first. See patch attached below.
> As for the tests you posted: too many of them rely on Posix file
> modes, and thus will probably either fail or be unable to provide
> meaningful testing on MS-Windows. Can we please augment that by tests
> that create unlocking problems by, e.g., running a shell command to
> remove or rename or otherwise sabotage the lock file, so that the new
> functionality could be meaningfully tested on Windows as well?
I have not been able to come up with a way to achieve what you ask for.
I have no access to an MS-Windows system to test with, though I have
been testing with small FAT and NTFS file systems in a ramdisk under
GNU/Linux. It may be possible to take some approach that works on
MS-Windows systems, but I'm not in a good position to write that code.
Another issue is that the lock file is typicaly a symbolink link, and
operating systems typically ignore file modes on symbolic links, so it
is hard to put the lock file itself into an "invalid" state --
i.e. where simply attempting to access or delete it generates a file
system level error. This is why I resorted to modifying file modes on
the containing directory -- attemps to remove or access files in such a
directory do reliably generate errors (on POSIX systems).
What I've done is make the tests skip themselves when `set-file-mode' on
the test's temporary directory appears to not work. When I test this on
an ext2 file system the tests complete. When I test this on a FAT and
NTFS file system (set up as a ramdisk on GNU/Linux), the tests skip
themselves.
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-Add-some-test-coverage-for-src-filelock.c-Bug-46397.patch --]
[-- Type: text/x-diff, Size: 8074 bytes --]
From e05d400c177333cff0f558aff1948bcd55130c17 Mon Sep 17 00:00:00 2001
From: Matt Armstrong <matt@rfc20.org>
Date: Mon, 15 Feb 2021 12:59:08 -0800
Subject: [PATCH 1/2] Add some test coverage for src/filelock.c (Bug#46397).
* test/src/filelock-tests.el: new file
---
test/src/filelock-tests.el | 183 +++++++++++++++++++++++++++++++++++++
1 file changed, 183 insertions(+)
create mode 100644 test/src/filelock-tests.el
diff --git a/test/src/filelock-tests.el b/test/src/filelock-tests.el
new file mode 100644
index 0000000000..e568050c42
--- /dev/null
+++ b/test/src/filelock-tests.el
@@ -0,0 +1,183 @@
+;;; filelock-tests.el --- test file locking -*- lexical-binding: t; -*-
+
+;; Copyright (C) 2021 Free Software Foundation, Inc.
+
+;; This file is part of GNU Emacs.
+
+;; GNU Emacs is free software; you can redistribute it and/or modify
+;; it under the terms of the GNU General Public License as published by
+;; the Free Software Foundation, either version 3 of the License, or
+;; (at your option) any later version.
+
+;; GNU Emacs is distributed in the hope that it will be useful,
+;; but WITHOUT ANY WARRANTY; without even the implied warranty of
+;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+;; GNU General Public License for more details.
+
+;; You should have received a copy of the GNU General Public License
+;; along with this program. If not, see <https://www.gnu.org/licenses/>.
+
+;;; Commentary:
+
+;; This file tests code in src/filelock.c and, to some extent, the
+;; related code in src/fileio.c.
+
+;; Note: many of these tests set file modes to unusual values, with
+;; the aim of exercising Emacs' error handling code. These tests
+;; require POSIX file system semantics, which are not available on all
+;; file systems (e.g. FAT, NTFS). The tests skip themselves when the
+;; file system does not support the given file modes.
+
+;;; Code:
+
+(require 'cl-lib)
+(require 'ert)
+
+(defvar filelock-tests-temporary-file-directory nil
+ "The directory for writing temporary files in filelock tests.
+This is used by the function
+`filelock-tests--make-temp-directory' to override the value of
+the variable `temporary-file-directory'.")
+
+(defun filelock-tests--make-temp-directory ()
+ "Create and return the name of a temporary directory.
+The caller should delete the directory."
+ (let ((temporary-file-directory (or filelock-tests-temporary-file-directory
+ temporary-file-directory)))
+ (make-temp-file "test" t)))
+
+(defun filelock-tests--fixture (test-function)
+ "Call TEST-FUNCTION under a test fixture.
+Create a test directory and a buffer whose `buffer-file-name' and
+`buffer-file-truename' are a file within it. Call TEST-FUNCTION,
+passing it directory name. Finally, delete the buffer and the
+test directory.
+
+This function attempts to delete the test directory even if the
+test leaves the buffer locked or the test directory with strange
+permissions."
+ (let* ((temp-dir (filelock-tests--make-temp-directory))
+ (name (concat (file-name-as-directory temp-dir)
+ "file"))
+ (create-lockfiles t))
+ (unwind-protect
+ (with-temp-buffer
+ (setq buffer-file-name name)
+ (setq buffer-file-truename name)
+ (unwind-protect
+ (save-current-buffer
+ (funcall test-function temp-dir))
+
+ ;; Here begins somewhat delicate cleanup logic. It must
+ ;; be performed in this order.
+ ;;
+ ;; First, give Emacs permission to modify the test
+ ;; directory. The test may have left the diretory's modes
+ ;; in an unusual state.
+ ;;
+ ;; Second, mark the buffer unmodified. This prevents
+ ;; prompts from `kill-buffer' about saving unmodified
+ ;; buffers (when this test is run interactively).
+ ;;
+ ;; Third, delete the temp buffer (by way of
+ ;; `with-temp-buffer' above). Emacs may also delete the
+ ;; lock file from the test directory at this point.
+ ;;
+ ;; Fourth, delete the test directory itself.
+
+ (set-file-modes temp-dir (logior #o700 (file-modes temp-dir)))
+ (set-buffer-modified-p nil)))
+ (delete-directory temp-dir t nil))))
+
+(defun filelock-tests--make-file-inaccessible (filename)
+ "Make file named FILENAME inaccessible.
+Returns non-nil if the operation succeeds, otherwise nil. Note:
+not all file systems support this operation."
+ (set-file-modes filename #o000)
+ (equal #o000 (file-modes filename)))
+
+(ert-deftest filelock-tests-lock-unlock-no-errors ()
+ "Check that locking and unlocking works without error."
+ (filelock-tests--fixture
+ (lambda (_)
+ (should (not (file-locked-p (buffer-file-name))))
+ (insert "this locks the buffer's file")
+ (should (file-locked-p (buffer-file-name)))
+ (unlock-buffer)
+ (should (not (file-locked-p (buffer-file-name)))))))
+
+(ert-deftest filelock-tests-lock-buffer-permission-denied ()
+ "Check that locking a buffer in a directory with no write
+permissions does not work."
+ (filelock-tests--fixture
+ (lambda (temp-dir)
+ (should (not (file-locked-p (buffer-file-name))))
+
+ (let ((original-modes (file-modes temp-dir)))
+ (skip-unless (filelock-tests--make-file-inaccessible temp-dir))
+ ;; FIXME: file system errors when locking a file are ignored;
+ ;; should they be?
+ (insert "this locks the current buffer's file")
+ (set-file-modes temp-dir original-modes))
+
+ (should (not (file-locked-p (buffer-file-name)))))))
+
+(ert-deftest filelock-tests-file-locked-p-permission-denied ()
+ "Check that `file-locked-p' fails if the directory is inaccesible."
+ (filelock-tests--fixture
+ (lambda (temp-dir)
+ (skip-unless (filelock-tests--make-file-inaccessible temp-dir))
+
+ (let ((err (should-error (file-locked-p (buffer-file-name)))))
+ (should (equal (cl-subseq err 0 2)
+ '(file-error "Testing file lock")))))))
+
+(ert-deftest filelock-tests-unlock-permission-denied ()
+ "Check that `unlock-buffer' fails in directories that cannot be
+modified."
+ (filelock-tests--fixture
+ (lambda (temp-dir)
+ (insert "this locks the current buffer's file")
+ (should (file-locked-p (buffer-file-name)))
+
+ (skip-unless (filelock-tests--make-file-inaccessible temp-dir))
+
+ ;; FIXME: Unlocking buffers should not signal errors related to
+ ;; their lock files (bug#46397).
+ (let ((err (should-error (unlock-buffer))))
+ (should (equal (cl-subseq err 0 2)
+ '(file-error "Unlocking file")))))))
+
+(defun filelock-tests--yes (&rest _)
+ "Return t."
+ t)
+
+(ert-deftest filelock-tests-kill-buffer-permission-denied ()
+ "Check that `unlock-buffer' fails in directories that cannot be
+modified."
+ (filelock-tests--fixture
+ (lambda (temp-dir)
+ (insert "this should lock the buffer")
+ (should (file-locked-p (buffer-file-name)))
+
+ (skip-unless (filelock-tests--make-file-inaccessible temp-dir))
+
+ ;; Kill the current buffer even if it is modified. Use advice to
+ ;; fake a "yes" answer for the "Buffer modified; kill anyway?"
+ ;; prompt. Leave the buffer modified so `kill-buffer' will
+ ;; attempt to unlock the buffer's file.
+ ;;
+ ;; FIXME: Killing buffers should not signal errors related to
+ ;; their lock files (bug#46397).
+ (let ((err (unwind-protect
+ (progn
+ (advice-add 'yes-or-no-p
+ :override
+ 'filelock-tests--yes)
+ (should-error (kill-buffer)))
+ (advice-remove 'yes-or-no-p 'filelock-tests--yes))))
+ (should (equal (cl-subseq err 0 2)
+ '(file-error "Unlocking file")))))))
+
+(provide 'filelock-tests)
+;;; filelock-tests.el ends here
--
2.30.1
next prev parent reply other threads:[~2021-03-05 22:19 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-09 9:47 bug#46397: 27.1; Cannot delete buffer pointing to a file in a path that includes a file Peter
2021-02-09 23:47 ` Matt Armstrong
2021-02-10 0:23 ` Matt Armstrong
2021-02-10 15:05 ` Eli Zaretskii
2021-02-10 19:23 ` Paul Eggert
2021-02-10 19:45 ` Eli Zaretskii
2021-02-10 22:39 ` Matt Armstrong
2021-02-12 7:43 ` Eli Zaretskii
2021-02-12 9:36 ` Paul Eggert
2021-02-12 11:33 ` Eli Zaretskii
2021-02-12 23:59 ` Matt Armstrong
2021-02-13 8:07 ` Eli Zaretskii
2021-02-11 22:14 ` Matt Armstrong
2021-02-12 2:20 ` Paul Eggert
2021-02-12 7:15 ` Eli Zaretskii
2021-02-13 1:15 ` Matt Armstrong
2021-02-13 1:26 ` Paul Eggert
2021-02-13 8:21 ` Eli Zaretskii
2021-02-13 8:28 ` Eli Zaretskii
2021-02-14 0:49 ` Matt Armstrong
2021-02-14 19:22 ` Eli Zaretskii
2021-02-14 22:16 ` Matt Armstrong
2021-02-15 15:09 ` Eli Zaretskii
2021-02-16 0:49 ` Matt Armstrong
2021-02-16 1:55 ` Paul Eggert
2021-02-16 15:06 ` Eli Zaretskii
2021-02-16 11:53 ` Lars Ingebrigtsen
2021-02-22 19:24 ` bug#46397: [PATCH] " Matt Armstrong
2021-02-19 19:10 ` Matt Armstrong
2021-02-19 19:23 ` Eli Zaretskii
2021-02-19 21:46 ` Matt Armstrong
2021-02-20 9:09 ` Eli Zaretskii
2021-02-21 0:36 ` Matt Armstrong
2021-02-21 23:43 ` Mike Kupfer
2021-02-22 1:42 ` Matt Armstrong
2021-03-14 18:03 ` Bill Wohler
2021-03-17 23:36 ` Matt Armstrong
2021-02-24 17:37 ` Matt Armstrong
2021-02-24 18:50 ` Eli Zaretskii
2021-03-01 16:59 ` Eli Zaretskii
2021-03-05 22:19 ` Matt Armstrong [this message]
2021-03-06 9:36 ` Eli Zaretskii
2021-03-06 23:39 ` Matt Armstrong
2021-03-07 2:50 ` Paul Eggert
2021-03-07 5:57 ` Matt Armstrong
2021-02-19 19:45 ` Paul Eggert
2021-02-19 21:52 ` Matt Armstrong
2021-03-08 2:18 ` Matt Armstrong
2021-03-11 14:34 ` Eli Zaretskii
2021-03-17 23:49 ` Matt Armstrong
2021-03-17 23:51 ` Matt Armstrong
2021-03-20 10:43 ` Eli Zaretskii
2021-03-22 1:43 ` Matt Armstrong
2021-03-27 9:20 ` Eli Zaretskii
2021-02-10 0:26 ` Matt Armstrong
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
List information: https://www.gnu.org/software/emacs/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87y2f1meeu.fsf@mdeb \
--to=matt@rfc20.org \
--cc=46397@debbugs.gnu.org \
--cc=craven@gmx.net \
--cc=eggert@cs.ucla.edu \
--cc=eliz@gnu.org \
/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 public inbox
https://git.savannah.gnu.org/cgit/emacs.git
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).