bug#28959: 25.3; Tramp not honoring `tramp-backup-directory-alist'

bug#28959: 25.3; Tramp not honoring `tramp-backup-directory-alist'

[Matz, Daniel A. (JSC-EG511)]

[-- Attachment #1: Type: text/plain, Size: 3976 bytes --]

Hello,

Tramp seems to have recently stopped honoring `tramp-backup-directory-alist’.  I have it configured to put all backup files in a single directory:

    (setq tramp-backup-directory-alist '(("." . "/Users/dmatz/.emacs.d/var/backup/“)))

But I’ve started noticing backup files being created on the remote systems, in the same directory as the file I am visiting.

I just tested this with emacs -Q as well, and got the same behavior.

Thanks!

Daniel


In GNU Emacs 25.3.1 (x86_64-apple-darwin15.6.0, NS appkit-1404.47 Version 10.11.6 (Build 15G1611))
of 2017-09-13 built on JSSLA17030925
Windowing system distributor 'Apple', version 10.3.1404
Configured using:
'configure --disable-dependency-tracking --disable-silent-rules
--enable-locallisppath=/usr/local/share/emacs/site-lisp
--infodir=/usr/local/Cellar/emacs/25.3/share/info/emacs
--prefix=/usr/local/Cellar/emacs/25.3 --without-x --with-xml2
--without-dbus --without-gnutls --with-imagemagick --with-ns
--disable-ns-self-contained'

Configured features:
JPEG IMAGEMAGICK NOTIFY ACL LIBXML2 ZLIB TOOLKIT_SCROLL_BARS NS

Important settings:
  value of $LANG: en_US.UTF-8
  locale-coding-system: utf-8-unix

Major mode: Lisp Interaction

Minor modes in effect:
  shell-dirtrack-mode: t
  tooltip-mode: t
  global-eldoc-mode: t
  electric-indent-mode: t
  mouse-wheel-mode: t
  tool-bar-mode: t
  menu-bar-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  blink-cursor-mode: t
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t
  line-number-mode: t
  transient-mark-mode: t

Recent messages:
Tramp: Decoding remote file ‘/scpx:ngc09:/home/dmatz/test_file’ using ‘base64 -d -i >%s’...done
Wrote /scpx:ngc09:/home/dmatz/test_file
Making completion list...
Quit [2 times]
Making completion list... [2 times]
command-execute: Command attempted to use minibuffer while in minibuffer
Quit [4 times]
Undo!
Quit [2 times]
Type C-x 1 to delete the help window.

Load-path shadows:
None found.

Features:
(shadow sort mail-extr pp emacsbug message rfc822 mml mml-sec epg
epg-config mm-decode mm-bodies mm-encode mail-parse rfc2231 mailabbrev
gmm-utils mailheader sendmail rfc2047 rfc2045 ietf-drums mail-utils
thingatpt misearch multi-isearch dired-aux tramp-cmds dired tramp-cache
tramp-sh tramp tramp-compat auth-source cl-seq eieio byte-opt bytecomp
byte-compile cl-extra cconv eieio-core cl-macs gv gnus-util mm-util
help-fns help-mode easymenu cl-loaddefs pcase cl-lib mail-prsvr
password-cache tramp-loaddefs trampver shell pcomplete comint ansi-color
ring format-spec advice time-date mule-util tooltip eldoc electric
uniquify ediff-hook vc-hooks lisp-float-type mwheel ns-win ucs-normalize
term/common-win tool-bar dnd fontset image regexp-opt fringe
tabulated-list newcomment elisp-mode lisp-mode prog-mode register page
menu-bar rfn-eshadow timer select scroll-bar mouse jit-lock font-lock
syntax facemenu font-core frame cl-generic cham georgian utf-8-lang
misc-lang vietnamese tibetan thai tai-viet lao korean japanese eucjp-ms
cp51932 hebrew greek romanian slovak czech european ethiopic indian
cyrillic chinese charscript case-table epa-hook jka-cmpr-hook help
simple abbrev minibuffer cl-preloaded nadvice loaddefs button faces
cus-face macroexp files text-properties overlay sha1 md5 base64 format
env code-pages mule custom widget hashtable-print-readable backquote
kqueue cocoa ns multi-tty make-network-process emacs)

Memory information:
((conses 16 220937 14198)
(symbols 48 22348 0)
(miscs 40 70 316)
(strings 32 24744 6349)
(string-bytes 1 791176)
(vectors 16 37242)
(vector-slots 8 707102 8716)
(floats 8 274 420)
(intervals 56 924 7)
(buffers 976 22))


--

Daniel Matz
EG5 - Flight Mechanics and Trajectory Design
NASA Johnson Space Center
daniel.a.matz@nasa.gov<mailto:daniel.a.matz@nasa.gov>
(281) 244-0263


[-- Attachment #2: Type: text/html, Size: 10104 bytes --]

bug#28959: 25.3; Tramp not honoring `tramp-backup-directory-alist'

[Michael Albinus]

"Matz, Daniel A. (JSC-EG511)" <daniel.a.matz@nasa.gov> writes:

> Hello, 

Hi Daniel,

> Tramp seems to have recently stopped honoring
> `tramp-backup-directory-alist’.  I have it configured to put all
> backup files in a single directory:
>
>     (setq tramp-backup-directory-alist '(("." .
> "/Users/dmatz/.emacs.d/var/backup/“)))
>
> But I’ve started noticing backup files being created on the remote
> systems, in the same directory as the file I am visiting.
>
> I just tested this with emacs -Q as well, and got the same behavior.

I can reproduce the problem with Emacs 26.0.90, but not with Emacs 25.3
as the report indicates. Will check what's up.

> Thanks!
>
> Daniel

Best regards, Michael.

bug#28959: 25.3; Tramp not honoring `tramp-backup-directory-alist'

[Michael Albinus]

"Matz, Daniel A. (JSC-EG511)" <daniel.a.matz@nasa.gov> writes:

> Hello, 

Hi Daniel,

> Tramp seems to have recently stopped honoring
> `tramp-backup-directory-alist’.  I have it configured to put all
> backup files in a single directory:
>
>     (setq tramp-backup-directory-alist '(("." .
> "/Users/dmatz/.emacs.d/var/backup/“)))
>
> But I’ve started noticing backup files being created on the remote
> systems, in the same directory as the file I am visiting.
>
> I just tested this with emacs -Q as well, and got the same behavior.

I've committed a fix to the emacs-26 branch of Emacs' git repository. Do
you have a chance to check this?

> Thanks!
>
> Daniel

Best regards, Michael.

bug#28959: 25.3; Tramp not honoring `tramp-backup-directory-alist'

[Michael Albinus]

"Matz, Daniel A. (JSC-EG511)" <daniel.a.matz@nasa.gov> writes:

> Hi, Michael.

Hi Daniel,

[Pls keep 28959@debbugs.gnu.org in Cc, for the archives]

> I built the emacs-26 branch and started emacs with -Q, and
> unfortunately the backup files are still being created on the remote
> system, in the same directory as the file I am visiting.
>
> Is there anything else I can do to help you debug this?

Could you pls check that your checkout is recent? It must contain

--8<---------------cut here---------------start------------->8---
~/src/emacs-26> git log
commit 646e56e150ca08978d6ce736b12867b4958a0cd8 (HEAD -> emacs-26, origin/emacs-26)
Author: Michael Albinus <michael.albinus@gmx.de>
Date:   Thu Oct 26 16:24:28 2017 +0200

    Fix Bug#28959
    
    * lisp/net/tramp.el (tramp-handle-find-backup-file-name):
    Use `tramp-tramp-file-p' rather than `tramp-file-name-p'.  Add
    hop to backup file name.  (Bug#28959)
--8<---------------cut here---------------end--------------->8---

Furthermore, you might run

--8<---------------cut here---------------start------------->8---
~/src/emacs-26> make -C test tramp-tests
--8<---------------cut here---------------end--------------->8---

There are several Tramp tests; the most interesting is

--8<---------------cut here---------------start------------->8---
   passed  43/58  tramp-test34-find-backup-file-name
--8<---------------cut here---------------end--------------->8---

Do you see this?

> Thanks!
>
> Daniel

Best regards, Michael.

bug#28959: 25.3; Tramp not honoring `tramp-backup-directory-alist'

[Matz, Daniel A. (JSC-EG511)]

Michael,

I double checked my checkout, and it is recent and includes your commit.

For the tests, 43/58 does indeed pass.  Here’s the summary that prints at the end:

--8<---------------cut here---------------start------------->8---
Ran 58 tests, 54 results as expected, 1 unexpected, 3 skipped (2017-10-26 11:33:34-0500)
1 expected failures

1 unexpected results:
   FAILED  tramp-test30-environment-variables-and-port-numbers

3 skipped results:
  SKIPPED  tramp-test36-special-characters-with-stat
  SKIPPED  tramp-test37-utf8-with-stat
  SKIPPED  tramp-test38-file-system-info
--8<---------------cut here---------------end--------------->8---

The only other failure was 11/58 tramp-test05-expand-file-name-relative, which I suppose must be the expected failure that is mentioned.

Daniel

> On Oct 26, 2017, at 11:09 AM, Michael Albinus <michael.albinus@gmx.de> wrote:
> 
> "Matz, Daniel A. (JSC-EG511)" <daniel.a.matz@nasa.gov> writes:
> 
>> Hi, Michael.
> 
> Hi Daniel,
> 
> [Pls keep 28959@debbugs.gnu.org in Cc, for the archives]
> 
>> I built the emacs-26 branch and started emacs with -Q, and
>> unfortunately the backup files are still being created on the remote
>> system, in the same directory as the file I am visiting.
>> 
>> Is there anything else I can do to help you debug this?
> 
> Could you pls check that your checkout is recent? It must contain
> 
> --8<---------------cut here---------------start------------->8---
> ~/src/emacs-26> git log
> commit 646e56e150ca08978d6ce736b12867b4958a0cd8 (HEAD -> emacs-26, origin/emacs-26)
> Author: Michael Albinus <michael.albinus@gmx.de>
> Date:   Thu Oct 26 16:24:28 2017 +0200
> 
>    Fix Bug#28959
> 
>    * lisp/net/tramp.el (tramp-handle-find-backup-file-name):
>    Use `tramp-tramp-file-p' rather than `tramp-file-name-p'.  Add
>    hop to backup file name.  (Bug#28959)
> --8<---------------cut here---------------end--------------->8---
> 
> Furthermore, you might run
> 
> --8<---------------cut here---------------start------------->8---
> ~/src/emacs-26> make -C test tramp-tests
> --8<---------------cut here---------------end--------------->8---
> 
> There are several Tramp tests; the most interesting is
> 
> --8<---------------cut here---------------start------------->8---
>   passed  43/58  tramp-test34-find-backup-file-name
> --8<---------------cut here---------------end--------------->8---
> 
> Do you see this?
> 
>> Thanks!
>> 
>> Daniel
> 
> Best regards, Michael.

bug#28959: 25.3; Tramp not honoring `tramp-backup-directory-alist'

[Michael Albinus]

[-- Attachment #1: Type: text/plain, Size: 742 bytes --]

"Matz, Daniel A. (JSC-EG511)" <daniel.a.matz@nasa.gov> writes:

> Michael,

Hi Daniel,

> I double checked my checkout, and it is recent and includes your commit.
>
> For the tests, 43/58 does indeed pass.  Here’s the summary that prints
> at the end:

This guarantees you have the latest changes. I've added
tramp-test34-find-backup-file-name earlier today, and you have it.

So we need traces. Could you pls apply the appended patch to tramp.el?
Set tramp-verbose to 6, and rerun your test. Hopefully, we'll see what
happens in the Tramp debug buffer, which you'll send to me.

If you fear there are sensible information, you could send the debug
buffer to me directly, w/o Cc.

> Daniel

Best regards, Michael.


[-- Attachment #2: Type: text/plain, Size: 1260 bytes --]

diff --git a/lisp/tramp.el b/lisp/tramp.el
index e300b3a..08a9076 100644
--- a/lisp/tramp.el
+++ b/lisp/tramp.el
@@ -3218,6 +3218,12 @@ User is always nil."
 (defun tramp-handle-find-backup-file-name (filename)
   "Like `find-backup-file-name' for Tramp files."
   (with-parsed-tramp-file-name filename nil
+    (tramp-message
+     v 5 "filename %s backup-enable-predicate %s"
+     filename backup-enable-predicate)
+    (tramp-message
+     v 5 "backup-directory-alist %s tramp-backup-directory-alist %s"
+     backup-directory-alist tramp-backup-directory-alist)
     (let ((backup-directory-alist
 	   (if tramp-backup-directory-alist
 	       (mapcar
@@ -3231,8 +3237,14 @@ User is always nil."
 			method user domain host port (cdr x) hop)
 		     (cdr x))))
 		tramp-backup-directory-alist)
-	     backup-directory-alist)))
-      (tramp-run-real-handler 'find-backup-file-name (list filename)))))
+	     backup-directory-alist))
+	  result)
+      (setq result
+	    (tramp-run-real-handler 'find-backup-file-name (list filename)))
+    (tramp-message
+     v 5 "backup-directory-alist %s result %s"
+     backup-directory-alist result)
+    result)))
 
 (defun tramp-handle-insert-directory
   (filename switches &optional wildcard full-directory-p)

bug#28959: 25.3; Tramp not honoring `tramp-backup-directory-alist'

[Michael Albinus]

Version: 26.1

"Matz, Daniel A. (JSC-EG511)" <daniel.a.matz@nasa.gov> writes:

> Michael,

Hi Daniel,

> Well that explains it then.  The configuration you suggested does
> indeed do what I want.

[For the archives: I've recommended to use `backup-directory-alist'
instead of `tramp-backup-directory-alist']

OK, so I'm closing the bug.

> I apologize for taking up your time with this when in the end it was
> just user error!

Your bug report was helpful! As you have seen with my patch, I found a
Tramp error indeed, which is fixed now.

> Thank you again for your help, and for Tramp!
>
> Daniel

Best regards, Michael.