bug#23743: 24.5; Opening output file: invalid argument, .../foo-<-4-n.el

bug#23743: 24.5; Opening output file: invalid argument, .../foo-<-4-n.el

[Drew Adams]

emacs -Q

C-x C-f foo-<-4-n.el

Type some text in that file buffer.

C-x C-s  ; try to save it.

Saving file d:/usr/drew/drews-lisp-20/foo-<-4-n.el...
basic-save-buffer-2: Opening output file: invalid argument, d:/.../foo-<-4-n.el

Apparently it doesn't like `<' in a file name?  If so, and if that is
"normal", then the user should get a much better error message than
this.

(Same problem in the latest Emacs 25 build I have.)



In GNU Emacs 24.5.1 (i686-pc-mingw32)
 of 2015-04-11 on LEG570
Windowing system distributor `Microsoft Corp.', version 6.1.7601
Configured using:
 `configure --prefix=/c/usr --host=i686-pc-mingw32'

bug#23743: 24.5; Opening output file: invalid argument, .../foo-<-4-n.el

[Eli Zaretskii]

> Date: Fri, 10 Jun 2016 10:35:54 -0700 (PDT)
> From: Drew Adams <drew.adams@oracle.com>
> 
> emacs -Q
> 
> C-x C-f foo-<-4-n.el
> 
> Type some text in that file buffer.
> 
> C-x C-s  ; try to save it.
> 
> Saving file d:/usr/drew/drews-lisp-20/foo-<-4-n.el...
> basic-save-buffer-2: Opening output file: invalid argument, d:/.../foo-<-4-n.el
> 
> Apparently it doesn't like `<' in a file name?

Yes, Windows doesn't allow that.

> If so, and if that is "normal", then the user should get a much
> better error message than this.

That's the error that comes from the filesystem, Emacs doesn't know
any better.

bug#23743: 24.5; Opening output file: invalid argument, .../foo-<-4-n.el

[Noam Postavsky]

On Fri, Jun 10, 2016 at 3:04 PM, Eli Zaretskii <eliz@gnu.org> wrote:
>> Date: Fri, 10 Jun 2016 10:35:54 -0700 (PDT)
>> From: Drew Adams <drew.adams@oracle.com>
>>
>> Apparently it doesn't like `<' in a file name?
>
> Yes, Windows doesn't allow that.
>
>> If so, and if that is "normal", then the user should get a much
>> better error message than this.
>
> That's the error that comes from the filesystem, Emacs doesn't know
> any better.

Right, cf https://support.microsoft.com/en-us/kb/177506

Is it worth blocking this sort of file name in read-file-name, or
should we just mark this wontfix?

bug#23743: 24.5; Opening output file: invalid argument, .../foo-<-4-n.el

[Eli Zaretskii]

> From: Noam Postavsky <npostavs@users.sourceforge.net>
> Date: Mon, 20 Jun 2016 23:15:07 -0400
> Cc: Drew Adams <drew.adams@oracle.com>, 23743@debbugs.gnu.org
> 
> Right, cf https://support.microsoft.com/en-us/kb/177506
> 
> Is it worth blocking this sort of file name in read-file-name

No, I don't think so, because MS-Windows includes a pluggable
filesystem design, whereby any 3rd party filesystem device driver
could be added, which supports these characters in file names.  We
simply cannot know up front which filenames are invalid.

> or should we just mark this wontfix?

I think so, yes.

Thanks.