Hi,
i've mounted a share of a WinXP machine on a linux machine. The german
umlauts in the filenames we're not shown correctly, so i looked into the
docs and found out that i have to specify the charset and the codepage.
Well, i have a small problem with that. I have to set the charset to the
value my console uses so that i'll see all the pathnames correctly.
Where do i need the codepage for?
WinXP and NTFS support Unicode, so i guess the SMB protocol does too.
All pathnames could be transferred over the network using Unicode. No
codepage would be needed, cause only one conversion would be needed on
my linux machine:
unicode->charset
and the charset could be utf8 for example.
I guess that it is currenly that way:
- Unicode filenames are read from my NTFS filesystem
- are converted into the specified codepage
- are tranferred over the net
- the filenames gets converted from the codepage into the charset that i
specified
That doesn't satisfy me, cause even if i would use a utf8 console on my
linux machine and would mount my samba-shared with charset utf8, i
wouldn't be abled to create files with unicode-chars in it.
The codepage would stay the "bottle-neck" cause all need to be
converted
to that codepage.
Well, another point would be to stay compatible with Win9x cause i don't
guess that it does support unicode-filenames properly - and perhaps it
doesn't even support to use unicode with the SMB protocl.
Well, how's the status? Is this a known problem? Was there any progress
in Samba3.0 to "solve" that?
Thx
Sven
