>>>> "APA" == Adrian Aichner
<Adrian.Aichner(a)t-online.de> writes: 
>>>>> "Martin" == Martin Buchholz <martin(a)xemacs.org>
writes:
>>>> "APA" == Adrian Aichner
<Adrian.Aichner(a)t-online.de> writes: 
>>>>
"APA" == Adrian Aichner <Adrian.Aichner(a)t-online.de> writes: 
   
APA> Torgny has located the problem.
    APA> Thanks, Torgny!
    APA> What is the right approach to fix this?
    APA> Martin,
    APA> you've worked on this last.
    APA> The current code in 21.1.14 does work on Windows 2000 while the code
    APA> in 21.2-b46 does not!
    APA> I'm willing to test any suggestions you have.
    Martin> I don't know.  Both 21.1 and 21.2 has patches from me to
    Martin> fix the behavior of copy-file with non-ascii file names
    Martin> when keep-time is t.  It's quite likely I've introduced a
    Martin> bug, but someone on Windows is going to have to dig it up.
    Martin> It's also possible that keep-time never really worked on
    Martin> Windows.
    Martin> Create a file named öld, and then try evaluating:
    Martin> (copy-file "öld" "nèw" t t)
    APA> nil
    APA> is what I get in (emacs-version)
    APA> "XEmacs 21.1 (patch 14) \"Cuyahoga Valley\" [Lucid]
(i386-pc-win32) of
    APA> Wed Mar 14 2001 on D5DC120J"
    APA> and here is what dired sez:
    APA>   c:\Users\AichnerAd\tmp:
    APA>   total 79804
    APA> <lines deleted by Adrian>
    APA>   -rw-rw-rw-   1 AichnerAd 5               1 Apr 08 21:29 öld
    APA>   drwxrwxrwx   2 AichnerAd 5               0 Apr 08 21:29 ..
    APA>   -rw-rw-rw-   1 AichnerAd 5               1 Apr 08 21:29 nèw
    APA>   drwxrwxrwx   2 AichnerAd 5               0 Apr 08 21:29 .
    APA> So, this did work on Windows 2000.
    APA> I'm off to a 21.2-b46 XEmacs now ...
(copy-file "öld" "nèw" t t)
leads to this file-error:
Signaling: (file-error "I/O error" "No such file or directory"
"c:\\Users\\AichnerAd\\tmp\\nèw")
  copy-file("öld" "nèw" t t)
  eval((copy-file "öld" "nèw" t t))
  eval-interactive((copy-file "öld" "nèw" t t))
  eval-last-sexp((4))
  call-interactively(eval-last-sexp)
in (emacs-version)
"XEmacs 21.2 (beta46) \"Urania\" [Lucid] (i586-pc-win32) of Thu Mar 29
2001 on D5DC120J"
and here is what dired sez:
  c:\Users\AichnerAd\tmp:
  total 79804
<lines deleted by Adrian>
  -rw-rw-rw-   1 AichnerAd 0               1 Apr 08 21:29 öld
  drwxrwxrwx   2 AichnerAd 0               0 Apr 08 22:01 ..
  drwxrwxrwx   2 AichnerAd 0               0 Apr 08 22:01 .
  -rw-rw-rw-   1 AichnerAd 0               1 Apr 08 22:01 nèw
So, the nèw file does get created!
Smörebröd beats Smørebrød!
I'll try the solution suggested by Ben now ...
Adrian
    APA> Adrain
    Martin> This works for me, mule or non-mule, 21.1 or 21.2 - I've
    Martin> tested it.
    Martin> Ben or Andy or Adrian (or Torgny) should try to find the
    Martin> fix, using a debugger at the C level.  It shouldn't be
    Martin> hard.
    APA> -- 
    APA> Adrian Aichner
    APA>  mailto:adrian@xemacs.org
    APA>  
http://www.xemacs.org/
-- 
Adrian Aichner
 mailto:adrian@xemacs.org
 
http://www.xemacs.org/