bug#5447: 23.1.91; load-file fails for C:/the-file.el.gz

bug#5447: 23.1.91; load-file fails for C:/the-file.el.gz

[Eli Zaretskii]

This bug report will be sent to the Free Software Foundation,
not to your local site managers!
Please write in English if possible, because the Emacs maintainers
usually do not have translators to read other languages for them.

Your bug report will be posted to the bug-gnu-emacs@gnu.org mailing list,
and to the gnu.emacs.bug news group.

Please describe exactly what actions triggered the bug
and the precise symptoms of the bug.  If you can, give
a recipe starting from `emacs -Q':

 emacs -Q
 M-x load-file RET C:/the-file.el.gz RET

This signals an error: Wrong type argument: consp, nil.  The traceback
is as follows:

  Debugger entered--Lisp error: (wrong-type-argument consp nil)
    jka-compr-load("c:/the-file.el.gz" nil nil t)
    apply(jka-compr-load ("c:/the-file.el.gz" nil nil t))
    jka-compr-handler(load "c:/the-file.el.gz" nil nil t)
    load("c:/the-file.el.gz" nil nil t)
    load-file("c:/the-file.el.gz")
    call-interactively(load-file t nil)
    execute-extended-command(nil)
    call-interactively(execute-extended-command nil nil)

This problem does not happen for files not in the root directory of a
Windows drive, e.g., C:/Temp/the-file.el.gz

If Emacs crashed, and you have the Emacs process in the gdb debugger,
please include the output from the following gdb commands:
    `bt full' and `xbacktrace'.
For information about debugging Emacs, please read the file
d:/gnu/bzr/emacs/trunk/etc/DEBUG.


In GNU Emacs 23.1.91.1 (i386-mingw-nt5.1.2600)
 of 2010-01-22 on HOME-C4E4A596F7
Windowing system distributor `Microsoft Corp.', version 5.1.2600
configured using `configure --with-gcc (3.4)'

Important settings:
  value of $LC_ALL: nil
  value of $LC_COLLATE: nil
  value of $LC_CTYPE: nil
  value of $LC_MESSAGES: nil
  value of $LC_MONETARY: nil
  value of $LC_NUMERIC: nil
  value of $LC_TIME: nil
  value of $LANG: ENU
  value of $XMODIFIERS: nil
  locale-coding-system: cp1255
  default enable-multibyte-characters: t

Major mode: Lisp Interaction

Minor modes in effect:
  shell-dirtrack-mode: t
  tooltip-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-encryption-mode: t
  auto-compression-mode: t
  line-number-mode: t
  transient-mark-mode: t

Recent input:
M-x l o a C-g M-x s e t - v a r <tab> <return> d e 
b <tab> o <tab> e <tab> <return> t <return> M-x l o 
a d - f i <tab> <return> C : / t h e - f i l e . e 
l . g z <return> C-x o M-x r e p o r t - e m <tab> 
<return>

Recent messages:
For information about GNU Emacs and the GNU system, type C-h C-a.
Quit
Loading tramp...done
uncompressing the-file.el.gz...done
Loading c:/the-file.el.gz...done.
Entering debugger...

Load-path shadows:
None found.

Features:
(shadow sort mail-extr emacsbug help-mode view debug jka-compr ange-ftp
tramp-imap epa derived epg epg-config imap-hash imap message smtpmail
sendmail ecomplete rfc822 mml mml-sec mm-decode mm-bodies mm-encode
mailcap mail-parse rfc2231 rfc2047 rfc2045 qp ietf-drums mailabbrev
nnheader mm-util mail-prsvr gmm-utils mailheader canlock sha1 hex-util
hashcash mail-utils assoc tramp-gw tramp-fish tramp-cache tramp-ftp
tramp-cmds tramp regexp-opt auth-source gnus-util netrc time-date advice
advice-preload shell comint ring password-cache format-spec tramp-compat
trampver cus-edit easymenu wid-edit cus-start cus-load help-fns tooltip
ediff-hook vc-hooks lisp-float-type mwheel dos-w32 disp-table ls-lisp
w32-win w32-vars tool-bar dnd fontset image fringe lisp-mode register
page menu-bar rfn-eshadow timer select scroll-bar mldrag mouse jit-lock
font-lock syntax facemenu font-core frame cham georgian utf-8-lang
misc-lang vietnamese tibetan thai tai-viet lao korean japanese hebrew
greek romanian slovak czech european ethiopic indian cyrillic chinese
case-table epa-hook jka-cmpr-hook help simple abbrev button minibuffer
faces cus-face files text-properties overlay md5 base64 format env
code-pages mule custom widget hashtable-print-readable backquote
make-network-process multi-tty emacs)

bug#5447: marked as done (23.1.91; load-file fails for C:/the-file.el.gz)

[Emacs bug Tracking System]

[-- Attachment #1: Type: text/plain, Size: 845 bytes --]

Your message dated Fri, 22 Jan 2010 12:32:52 +0200
with message-id <83hbqe77l7.fsf@gnu.org>
and subject line Re: bug#5447: 23.1.91; load-file fails for C:/the-file.el.gz
has caused the Emacs bug report #5447,
regarding 23.1.91; load-file fails for C:/the-file.el.gz
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact bug-gnu-emacs@gnu.org
immediately.)


-- 
5447: http://debbugs.gnu.org/cgi/bugreport.cgi?bug=5447
Emacs Bug Tracking System
Contact bug-gnu-emacs@gnu.org with problems

[-- Attachment #2: Type: message/rfc822, Size: 7800 bytes --]

From: Eli Zaretskii <eliz@gnu.org>
To: bug-gnu-emacs@gnu.org
Subject: 23.1.91; load-file fails for C:/the-file.el.gz
Date: Fri, 22 Jan 2010 11:01:59 +0200
Message-ID: <83iqau7bso.fsf@gnu.org>

This bug report will be sent to the Free Software Foundation,
not to your local site managers!
Please write in English if possible, because the Emacs maintainers
usually do not have translators to read other languages for them.

Your bug report will be posted to the bug-gnu-emacs@gnu.org mailing list,
and to the gnu.emacs.bug news group.

Please describe exactly what actions triggered the bug
and the precise symptoms of the bug.  If you can, give
a recipe starting from `emacs -Q':

 emacs -Q
 M-x load-file RET C:/the-file.el.gz RET

This signals an error: Wrong type argument: consp, nil.  The traceback
is as follows:

  Debugger entered--Lisp error: (wrong-type-argument consp nil)
    jka-compr-load("c:/the-file.el.gz" nil nil t)
    apply(jka-compr-load ("c:/the-file.el.gz" nil nil t))
    jka-compr-handler(load "c:/the-file.el.gz" nil nil t)
    load("c:/the-file.el.gz" nil nil t)
    load-file("c:/the-file.el.gz")
    call-interactively(load-file t nil)
    execute-extended-command(nil)
    call-interactively(execute-extended-command nil nil)

This problem does not happen for files not in the root directory of a
Windows drive, e.g., C:/Temp/the-file.el.gz

If Emacs crashed, and you have the Emacs process in the gdb debugger,
please include the output from the following gdb commands:
    `bt full' and `xbacktrace'.
For information about debugging Emacs, please read the file
d:/gnu/bzr/emacs/trunk/etc/DEBUG.


In GNU Emacs 23.1.91.1 (i386-mingw-nt5.1.2600)
 of 2010-01-22 on HOME-C4E4A596F7
Windowing system distributor `Microsoft Corp.', version 5.1.2600
configured using `configure --with-gcc (3.4)'

Important settings:
  value of $LC_ALL: nil
  value of $LC_COLLATE: nil
  value of $LC_CTYPE: nil
  value of $LC_MESSAGES: nil
  value of $LC_MONETARY: nil
  value of $LC_NUMERIC: nil
  value of $LC_TIME: nil
  value of $LANG: ENU
  value of $XMODIFIERS: nil
  locale-coding-system: cp1255
  default enable-multibyte-characters: t

Major mode: Lisp Interaction

Minor modes in effect:
  shell-dirtrack-mode: t
  tooltip-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-encryption-mode: t
  auto-compression-mode: t
  line-number-mode: t
  transient-mark-mode: t

Recent input:
M-x l o a C-g M-x s e t - v a r <tab> <return> d e 
b <tab> o <tab> e <tab> <return> t <return> M-x l o 
a d - f i <tab> <return> C : / t h e - f i l e . e 
l . g z <return> C-x o M-x r e p o r t - e m <tab> 
<return>

Recent messages:
For information about GNU Emacs and the GNU system, type C-h C-a.
Quit
Loading tramp...done
uncompressing the-file.el.gz...done
Loading c:/the-file.el.gz...done.
Entering debugger...

Load-path shadows:
None found.

Features:
(shadow sort mail-extr emacsbug help-mode view debug jka-compr ange-ftp
tramp-imap epa derived epg epg-config imap-hash imap message smtpmail
sendmail ecomplete rfc822 mml mml-sec mm-decode mm-bodies mm-encode
mailcap mail-parse rfc2231 rfc2047 rfc2045 qp ietf-drums mailabbrev
nnheader mm-util mail-prsvr gmm-utils mailheader canlock sha1 hex-util
hashcash mail-utils assoc tramp-gw tramp-fish tramp-cache tramp-ftp
tramp-cmds tramp regexp-opt auth-source gnus-util netrc time-date advice
advice-preload shell comint ring password-cache format-spec tramp-compat
trampver cus-edit easymenu wid-edit cus-start cus-load help-fns tooltip
ediff-hook vc-hooks lisp-float-type mwheel dos-w32 disp-table ls-lisp
w32-win w32-vars tool-bar dnd fontset image fringe lisp-mode register
page menu-bar rfn-eshadow timer select scroll-bar mldrag mouse jit-lock
font-lock syntax facemenu font-core frame cham georgian utf-8-lang
misc-lang vietnamese tibetan thai tai-viet lao korean japanese hebrew
greek romanian slovak czech european ethiopic indian cyrillic chinese
case-table epa-hook jka-cmpr-hook help simple abbrev button minibuffer
faces cus-face files text-properties overlay md5 base64 format env
code-pages mule custom widget hashtable-print-readable backquote
make-network-process multi-tty emacs)




[-- Attachment #3: Type: message/rfc822, Size: 4195 bytes --]

From: Eli Zaretskii <eliz@gnu.org>
To: 5447-done@debbugs.gnu.org
Subject: Re: bug#5447: 23.1.91; load-file fails for C:/the-file.el.gz
Date: Fri, 22 Jan 2010 12:32:52 +0200
Message-ID: <83hbqe77l7.fsf@gnu.org>

> Date: Fri, 22 Jan 2010 11:01:59 +0200
> From: Eli Zaretskii <eliz@gnu.org>
> Cc: 
> 
>  emacs -Q
>  M-x load-file RET C:/the-file.el.gz RET
> 
> This signals an error: Wrong type argument: consp, nil.  The traceback
> is as follows:
> 
>   Debugger entered--Lisp error: (wrong-type-argument consp nil)
>     jka-compr-load("c:/the-file.el.gz" nil nil t)
>     apply(jka-compr-load ("c:/the-file.el.gz" nil nil t))
>     jka-compr-handler(load "c:/the-file.el.gz" nil nil t)
>     load("c:/the-file.el.gz" nil nil t)
>     load-file("c:/the-file.el.gz")
>     call-interactively(load-file t nil)
>     execute-extended-command(nil)
>     call-interactively(execute-extended-command nil nil)

Fixed with the patch below.  The interesting thing is that
jka-compr-load is not even called if the file is not in the root
directory of a drive.  I will try to look into that now.

Should we perhaps run $TEMP etc. through file-truename when we compute
the value of temporary-file-directory?

2010-01-22  Eli Zaretskii  <eliz@gnu.org>

	* jka-compr.el (jka-compr-load): If load-file is not in
	load-history, try its file-truename version.  (bug#5447)

=== modified file 'lisp/jka-compr.el'
--- lisp/jka-compr.el	2010-01-13 08:35:10 +0000
+++ lisp/jka-compr.el	2010-01-22 10:19:39 +0000
@@ -590,7 +590,14 @@ There should be no more than seven chara
 	  (or nomessage
 	      (message "Loading %s...done." file))
 	  ;; Fix up the load history to point at the right library.
-	  (let ((l (assoc load-file load-history)))
+	  (let ((l (or (assoc load-file load-history)
+		       ;; On MS-Windows, if load-file is in
+		       ;; temporary-file-directory, it will look like
+		       ;; "c:/DOCUME~1/USER/LOCALS~1/foo", whereas
+		       ;; readevalloop will record its truename in
+		       ;; load-history.  Therefore try truename if the
+		       ;; original name is not in load-history.
+		       (assoc (file-truename load-file) load-history))))
 	    ;; Remove .gz and .elc?.
 	    (while (file-name-extension file)
 	      (setq file (file-name-sans-extension file)))

bug#5447: closed by Eli Zaretskii <eliz@gnu.org> (Re: bug#5447: 23.1.91; load-file fails for C:/the-file.el.gz)

[Eli Zaretskii]

> From: bug-gnu-emacs@gnu.org (Emacs bug Tracking System)
> Date: Fri, 22 Jan 2010 10:34:01 +0000
> 
> The interesting thing is that jka-compr-load is not even called if
> the file is not in the root directory of a drive.  I will try to
> look into that now.

This is actually trivial: a file that is not in C:/ is not considered
a remote file by Tramp handlers.  Case closed.

Like Lennart, I think that considering X:/ files on Windows to be
remote is a bad idea.  If it didn't have any serious consequences, I'd
just shrug.  But the recent bug reports indicate that it does have
very serious consequences, and is quite an annoyance.

Can we please get rid of that misfeature?

bug#5447: closed by Eli Zaretskii <eliz@gnu.org> (Re: bug#5447: 23.1.91; load-file fails for C:/the-file.el.gz)

[Stefan Monnier]

>> The interesting thing is that jka-compr-load is not even called if
>> the file is not in the root directory of a drive.  I will try to
>> look into that now.
> This is actually trivial: a file that is not in C:/ is not considered
> a remote file by Tramp handlers.  Case closed.
> Like Lennart, I think that considering X:/ files on Windows to be
> remote is a bad idea.  If it didn't have any serious consequences, I'd
> just shrug.  But the recent bug reports indicate that it does have
> very serious consequences, and is quite an annoyance.
> Can we please get rid of that misfeature?

I generally agree, but I'd like to see a concrete patch for it first.


        Stefan

bug#5447: closed by Eli Zaretskii <eliz@gnu.org> (Re: bug#5447: 23.1.91; load-file fails for C:/the-file.el.gz)

[Lennart Borgman]

On Fri, Jan 22, 2010 at 4:41 PM, Stefan Monnier
<monnier@iro.umontreal.ca> wrote:
>>> The interesting thing is that jka-compr-load is not even called if
>>> the file is not in the root directory of a drive.  I will try to
>>> look into that now.
>> This is actually trivial: a file that is not in C:/ is not considered
>> a remote file by Tramp handlers.  Case closed.
>> Like Lennart, I think that considering X:/ files on Windows to be
>> remote is a bad idea.  If it didn't have any serious consequences, I'd
>> just shrug.  But the recent bug reports indicate that it does have
>> very serious consequences, and is quite an annoyance.
>> Can we please get rid of that misfeature?
>
> I generally agree, but I'd like to see a concrete patch for it first.


Maybe the patch is not the problem. In the discussion around bug 5448
Michael wrote as an answer to my question:


On Wed, Jan 20, 2010 at 10:00 AM, Michael Albinus
<michael.albinus@gmx.de> wrote:
> Lennart Borgman <lennart.borgman@gmail.com> writes:
>
>>>> However I wonder why those files at all are interesting for tramp. I
>>>> know little about tramp, but does not remote file names always start
>>>> with something like "/ssh:", "/ftp:", "/telnet:" etc?
>>>>
>>>> If so why look for file names starting with "c:/"?
>>>
>>> Some packages ( I don't remember which ones) do add the volume letter to
>>> file names, which haven't one. Tramp silently removes the volume letter
>>> then, and continues.
>>
>> But is not that a bug in those packages then? Does not trying to fix
>> it in Tramp introduce this new bug?
>
> In theory, you are right. But Tramp supports also GNU Emacs 21, 22, and
> XEmacs. This makes it impossible to fix all involved packages.
>
> Best regards, Michael.


I do not understand exactly what Michael means with "some packages ...
do add the volume letter". However he says there is some problem
there. I still think this is a good time to fix it.

Looking at it again I get more confused. Does not the same thing
happen on un*x like file systems? I mean does not all that stuff Eli
mentioned get loaded if you try to complete a file name in the root on
those systems too? (But perhaps you do not bump into the problem there
that often?)

At least it looks to me from the pattern that this must happen there too:

   ("\\`\\([a-zA-Z]:\\)?/[^/]*\\'" . tramp-completion-file-name-handler)

Perhaps the right thing to do is to only match "/ssh:", "/ftp:",
"/telnet:" etc and only bring in Tramp if those things matches?

bug#5447: closed by Eli Zaretskii <eliz@gnu.org> (Re: bug#5447: 23.1.91; load-file fails for C:/the-file.el.gz)

[Michael Albinus]

Lennart Borgman <lennart.borgman@gmail.com> writes:

> At least it looks to me from the pattern that this must happen there too:
>
>    ("\\`\\([a-zA-Z]:\\)?/[^/]*\\'" . tramp-completion-file-name-handler)
>
> Perhaps the right thing to do is to only match "/ssh:", "/ftp:",
> "/telnet:" etc and only bring in Tramp if those things matches?

We are speaking about the tramp-***completion***-file-name-handler. This
is responsible to expand "/s <TAB>" to "/ssh:" (and alike).

Best regards, Michael.

bug#5447: closed by Eli Zaretskii <eliz@gnu.org> (Re: bug#5447: 23.1.91; load-file fails for C:/the-file.el.gz)

[Eli Zaretskii]

> From: Stefan Monnier <monnier@IRO.UMontreal.CA>
> Cc: 5447@debbugs.gnu.org
> Date: Fri, 22 Jan 2010 10:41:21 -0500
> 
> >> The interesting thing is that jka-compr-load is not even called if
> >> the file is not in the root directory of a drive.  I will try to
> >> look into that now.
> > This is actually trivial: a file that is not in C:/ is not considered
> > a remote file by Tramp handlers.  Case closed.
> > Like Lennart, I think that considering X:/ files on Windows to be
> > remote is a bad idea.  If it didn't have any serious consequences, I'd
> > just shrug.  But the recent bug reports indicate that it does have
> > very serious consequences, and is quite an annoyance.
> > Can we please get rid of that misfeature?
> 
> I generally agree, but I'd like to see a concrete patch for it first.

I meant a Windows-specific change; is that what you meant?  AFAICS,
typing "/" on Unix does not automatically load Tramp.

Michael, how hard would it be to special-case "\`[a-z]:/" on Windows so
that matching files are never considered to be ``remote files''?

bug#5447: closed by Eli Zaretskii <eliz@gnu.org> (Re: bug#5447: 23.1.91; load-file fails for C:/the-file.el.gz)

[Michael Albinus]

Eli Zaretskii <eliz@gnu.org> writes:

> Michael, how hard would it be to special-case "\`[a-z]:/" on Windows so
> that matching files are never considered to be ``remote files''?

IIRC, there was the case that "/method:host:" was expanded to
"C:/method:host:" by I-dont-know-which-package. Exactly because of this
problem we have the `tramp-drop-volume-letter' function.

I acknowledge, that it is an error which shall be fixed where it
happens, and not in Tramp. But not easy to hunt, especially for a guy
like me who doesn't work under w32.

Best regards, Michael.

bug#5447: closed by Eli Zaretskii <eliz@gnu.org> (Re: bug#5447: 23.1.91; load-file fails for C:/the-file.el.gz)

[Eli Zaretskii]

> From: Michael Albinus <michael.albinus@gmx.de>
> Cc: Stefan Monnier <monnier@IRO.UMontreal.CA>, 5447@debbugs.gnu.org
> Date: Fri, 22 Jan 2010 22:37:24 +0100
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > Michael, how hard would it be to special-case "\`[a-z]:/" on Windows so
> > that matching files are never considered to be ``remote files''?
> 
> IIRC, there was the case that "/method:host:" was expanded to
> "C:/method:host:" by I-dont-know-which-package. Exactly because of this
> problem we have the `tramp-drop-volume-letter' function.

We could do that nevertheless, and wait for someone to holler,
couldn't we?  I think punishing 95% of innocent users for the benefit
of maybe 5% is wrong anyway, and some different solution should be
found.  I agree that we cannot find a better solution until we know
the use-case.  But leaving this code intact will never get us to that
point.