I solved the problem. Still, the solution is rather puzzling.
The solution is to mount with -o iocharset=utf8,codepage=cp850.
The puzzling part is the code page. The whole idea with Unicode and
its UTF-8 implementation is to remove the need for local codepage
conversions. As long as a application/system is using Unicode there
should nerve be a need for codepages other than using Unicode.
Or have I misunderstood something?
~S
> Hello,
> I don't seem to be able to mount a WindowsXP share with smbmount and
> get the correct charset. From what I understand Win2000/XP use UTF-8
> for SMB shares, so I tried to mount with -o iocharset=utf8 but it
> did not make any difference. All extended characters are messed up
> or truncated.
> Here is what I found out using filenames with a '?':
> When I mount a Windows share with Samba the '?' becomes \206 when
I
> do ls -b.
> If I create the file on the Samba share from Windows it becomes
> \303\245 when I list it with ls -b.
> when I simply create a file on the local filesystem in Linux it
> becomes \345.
> When I browse a samba share that has two filenames, \303\245 and
> \345 they both show up correct in Windows. Only \303\245 show up
> correct if I list that file in Apache (which is set to UTF-8).
> Is not \303\ the UTF-8 prefix?
> Can someone help me shed the light on this?
>
> http://se.samba.org/samba/docs/man/Samba-HOWTO-Collection/unicode.html