GNU bug report logs - #18836
25.0.50; `insert-directory' barfs on nonexistent file, but not for `ls-lisp-insert-directory'

Previous Next

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 18836 in the body.
You can then email your comments to 18836 AT debbugs.gnu.org in the normal way.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Drew Adams <drew.adams <at> oracle.com>
To: bug-gnu-emacs <at> gnu.org
Subject: 25.0.50; `insert-directory' barfs on nonexistent file, but not for
 `ls-lisp-insert-directory'
Date: Sat, 25 Oct 2014 12:55:05 -0700 (PDT)

It seems that at least on some platforms (e.g. GNU/Linux), if you give
an explicit list of files and dirs to `dired', to list (which you do by
passing a cons as argument DIRNAME), the command fails with an error:
(file-error "Reading directory"
            "no such file or directory"
            "/usr/foo/no-such-file.txt")

No such error is raised on platforms that use `ls-lisp'.

This non-`ls-lisp' behavior is not as friendly as it could be, and not
as friendly as the `ls-lisp-insert-directory' behavior, which simply
displays a message and then calls `ding':
(message "%s: doesn't exist or is inaccessible" file)

Would it be possible to make vanilla `insert-directory' more tolerant,
i.e., make it act like `ls-lisp-insert-directory', in this respect?
Failing to access a single file should not prevent listing the others,
if this can be handled properly.  But maybe it's not feasible; dunno.

[FWIW, I would even prefer to see the file or dir listed in the Dired
buffer, even if it does not (yet) exist.  Let user attempts to actually
access it raise an error, and let no error be raised if the file is
created after creating the Dired buffer but before the user tries to
access the file.]

In GNU Emacs 25.0.50.1 (i686-pc-mingw32)
 of 2014-10-20 on LEG570
Bzr revision: 118168 rgm <at> gnu.org-20141020195941-icp42t8ttcnud09g
Windowing system distributor `Microsoft Corp.', version 6.1.7601
Configured using:
 `configure --enable-checking=yes,glyphs CPPFLAGS=-DGLYPH_DEBUG=1'

Message #8 received at 18836 <at> debbugs.gnu.org (full text, mbox):

From: Lars Ingebrigtsen <larsi <at> gnus.org>
To: Drew Adams <drew.adams <at> oracle.com>
Cc: 18836 <at> debbugs.gnu.org
Subject: Re: bug#18836: 25.0.50;
 `insert-directory' barfs on nonexistent file, but not for
 `ls-lisp-insert-directory'
Date: Sat, 30 Apr 2016 16:56:19 +0200

Drew Adams <drew.adams <at> oracle.com> writes:

> Would it be possible to make vanilla `insert-directory' more tolerant,
> i.e., make it act like `ls-lisp-insert-directory', in this respect?

I have no opinion on that, but: What on Earth is that function (and all
its adjacent functions) doing in files.el?

(insert-directory "." "-l")
=>
drwxrwxr-x 22 larsi larsi 4096 april 30 16:49 .

This doesn't seem like a very ... useful general function.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Message #11 received at 18836 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Lars Ingebrigtsen <larsi <at> gnus.org>
Cc: 18836 <at> debbugs.gnu.org, drew.adams <at> oracle.com
Subject: Re: bug#18836: 25.0.50;
 `insert-directory' barfs on nonexistent file, but not for
 `ls-lisp-insert-directory'
Date: Sat, 30 Apr 2016 18:39:59 +0300

> From: Lars Ingebrigtsen <larsi <at> gnus.org>
> Date: Sat, 30 Apr 2016 16:56:19 +0200
> Cc: 18836 <at> debbugs.gnu.org
> 
> > Would it be possible to make vanilla `insert-directory' more tolerant,
> > i.e., make it act like `ls-lisp-insert-directory', in this respect?
> 
> I have no opinion on that, but: What on Earth is that function (and all
> its adjacent functions) doing in files.el?

Where else would you like them to be?

> (insert-directory "." "-l")
> =>
> drwxrwxr-x 22 larsi larsi 4096 april 30 16:49 .
> 
> This doesn't seem like a very ... useful general function.

Then how come dired.el calls it?

Message #14 received at 18836 <at> debbugs.gnu.org (full text, mbox):

From: Lars Ingebrigtsen <larsi <at> gnus.org>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 18836 <at> debbugs.gnu.org
Subject: Re: bug#18836: 25.0.50;
 `insert-directory' barfs on nonexistent file, but not for
 `ls-lisp-insert-directory'
Date: Sat, 30 Apr 2016 17:46:10 +0200

Eli Zaretskii <eliz <at> gnu.org> writes:

>> From: Lars Ingebrigtsen <larsi <at> gnus.org>
>> Date: Sat, 30 Apr 2016 16:56:19 +0200
>> Cc: 18836 <at> debbugs.gnu.org
>> 
>> > Would it be possible to make vanilla `insert-directory' more tolerant,
>> > i.e., make it act like `ls-lisp-insert-directory', in this respect?
>> 
>> I have no opinion on that, but: What on Earth is that function (and all
>> its adjacent functions) doing in files.el?
>
> Where else would you like them to be?

In dired.el.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Message #17 received at 18836 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Lars Ingebrigtsen <larsi <at> gnus.org>
Cc: 18836 <at> debbugs.gnu.org
Subject: Re: bug#18836: 25.0.50;
 `insert-directory' barfs on nonexistent file, but not for
 `ls-lisp-insert-directory'
Date: Sat, 30 Apr 2016 18:53:44 +0300

> From: Lars Ingebrigtsen <larsi <at> gnus.org>
> Cc: 18836 <at> debbugs.gnu.org
> Date: Sat, 30 Apr 2016 17:46:10 +0200
> 
> >> I have no opinion on that, but: What on Earth is that function (and all
> >> its adjacent functions) doing in files.el?
> >
> > Where else would you like them to be?
> 
> In dired.el.

Dired isn't preloaded.

Message #20 received at 18836 <at> debbugs.gnu.org (full text, mbox):

From: Lars Ingebrigtsen <larsi <at> gnus.org>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 18836 <at> debbugs.gnu.org
Subject: Re: bug#18836: 25.0.50;
 `insert-directory' barfs on nonexistent file, but not for
 `ls-lisp-insert-directory'
Date: Sat, 30 Apr 2016 18:05:06 +0200

Eli Zaretskii <eliz <at> gnu.org> writes:

>> From: Lars Ingebrigtsen <larsi <at> gnus.org>
>> Cc: 18836 <at> debbugs.gnu.org
>> Date: Sat, 30 Apr 2016 17:46:10 +0200
>> 
>> >> I have no opinion on that, but: What on Earth is that function (and all
>> >> its adjacent functions) doing in files.el?
>> >
>> > Where else would you like them to be?
>> 
>> In dired.el.
>
> Dired isn't preloaded.

But does anything other than dired use it?  I see that other packages
(like ange-ftp) instrument it...  it that the reason it's in files.el?

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Message #23 received at 18836 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Lars Ingebrigtsen <larsi <at> gnus.org>
Cc: 18836 <at> debbugs.gnu.org
Subject: Re: bug#18836: 25.0.50;
 `insert-directory' barfs on nonexistent file, but not for
 `ls-lisp-insert-directory'
Date: Sat, 30 Apr 2016 19:20:28 +0300

> From: Lars Ingebrigtsen <larsi <at> gnus.org>
> Cc: 18836 <at> debbugs.gnu.org
> Date: Sat, 30 Apr 2016 18:05:06 +0200
> 
> Eli Zaretskii <eliz <at> gnu.org> writes:
> 
> >> From: Lars Ingebrigtsen <larsi <at> gnus.org>
> >> Cc: 18836 <at> debbugs.gnu.org
> >> Date: Sat, 30 Apr 2016 17:46:10 +0200
> >> 
> >> >> I have no opinion on that, but: What on Earth is that function (and all
> >> >> its adjacent functions) doing in files.el?
> >> >
> >> > Where else would you like them to be?
> >> 
> >> In dired.el.
> >
> > Dired isn't preloaded.
> 
> But does anything other than dired use it?  I see that other packages
> (like ange-ftp) instrument it...  it that the reason it's in files.el?

Yes.  Also, ls-lisp.

Message #28 received at 18836 <at> debbugs.gnu.org (full text, mbox):

From: Lars Ingebrigtsen <larsi <at> gnus.org>
To: Drew Adams <drew.adams <at> oracle.com>
Cc: 18836 <at> debbugs.gnu.org
Subject: Re: bug#18836: 25.0.50; `insert-directory' barfs on nonexistent
 file, but not for `ls-lisp-insert-directory'
Date: Tue, 25 Jan 2022 15:13:53 +0100

Drew Adams <drew.adams <at> oracle.com> writes:

> It seems that at least on some platforms (e.g. GNU/Linux), if you give
> an explicit list of files and dirs to `dired', to list (which you do by
> passing a cons as argument DIRNAME), the command fails with an error:
> (file-error "Reading directory"
>             "no such file or directory"
>             "/usr/foo/no-such-file.txt")
>
> No such error is raised on platforms that use `ls-lisp'.

This has changed between reporting this bug and now -- both now signal
errors currently, so I'm closing this bug report.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Message #33 received at 18836 <at> debbugs.gnu.org (full text, mbox):

From: Drew Adams <drew.adams <at> oracle.com>
To: Lars Ingebrigtsen <larsi <at> gnus.org>
Cc: "18836 <at> debbugs.gnu.org" <18836 <at> debbugs.gnu.org>
Subject: RE: [External] : Re: bug#18836: 25.0.50; `insert-directory' barfs on
 nonexistent file, but not for `ls-lisp-insert-directory'
Date: Tue, 25 Jan 2022 16:07:02 +0000

> This has changed between reporting this bug and now -- both now signal
> errors currently, so I'm closing this bug report.

Thanks.

I've filed bug #53524 for the additional improvement.
Users should be able to create the listing, with the
missing file/dir.  An error should be raised only if
and when you try to visit that file/dir.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.