greek (utf-8) filenames support in Freetz- please help in English

[gcf]

Hi all,
Having trouble to share a disk with greek utf-8 filenames with AVM's SAMBA, I installed freetz.
I still have the same problem even after adding
unix charset = utf8
Can someone help me solve this with correct samba settings?

Furthermore I am concerned seeing that the disk is mounted with these options:
usbfs on /proc/bus/usb type usbfs (rw)
/dev/sda1 on /var/media/ftp/uStor01 type vfat (rw,fmask=0000,dmask=0000,codepage=cp437,iocharset=iso8859-1)

I tried mounting it with
mount -t vfat -o iocharset=utf8,codepage=cp737 /dev/sda1 /var/media/ftp/uStor
01

but I get:
failed: invalid argument.
Any help will be appreciated,
thanks in advance

 

[gismotro]

You can try this:

socket options = TCP_NODELAY IPTOS_LOWDELAY
read raw = yes
write raw = yes
oplocks = yes
max xmit = 65535
dead time = 15
getwd cache = yes
[B][COLOR="red"]unix charset = UTF-8[/COLOR][/B]

 

[RalfFriedl]

What exactly is your problem? Can't you create the files, are the characters wrong, ...?

Can you create files with greek characters from the shell?

You might have to add support for cp737 to the kernel to use vfat, but a better idea would be to use another file system.

 

[gcf]

Gismotro's settings on an NTFS formatted disk worked just fine, so thank you both!

 

[hermann72pb]

@Ralf: The native charset of the Box is iso8859-1, therefore it will be impossible to use "real" greek characters. Almost all Linux systems use today UTF-8 as default, but we have on our box only iso8859-1, since the Box is used in most cases (98%) in Germany or by Germans. But theoretically it is possible to use UTF-8 at such small busybox-based systems. For example my Xtreamer and my QNAP-NAS have both UTF-8 as default. Both systems are busybox-based.
@gismotro: I can not understand, what do you say with the SAMBA option "unix charset = UTF-8"? I think, with this option you do overwrite the standard charset of the box for SAMBA using. The file names are saved as UTF-8 with this option, but you can not handle it from the shell, since your native charset is iso8859-1. Usualy you should change the native charset for the system. But if we do it, we must adapt all our packages, shell scripts and more over...

Best regards,

 

[gismotro]

Es war ein Tip den ich hier im Forum gefunden habe. Durch diesen Eintrag werden die ä, ö, und ü korrekt angezeigt.

 

[RalfFriedl]

There is no such thing as a native charset of the box. There is just a charset or interpretation in different programs. Of course life is easier if the interpretation of all programs is consistent.

Just having any Linux file system and setting Samba unix charset to utf8 (which is the default since a long time) will allow to store filenames in UTF-8 without any other change.
Creating filenames with UTF-8 names is irrelevant to most, but also easy, just use UTF-8 names in any shell command to create or read the files. Having an UTF-8 capable terminal makes this easier, but that has nothing to do with the shell. Having a shell that understands UTF-8 will improve the line editing in the shell, but not the other functions of a shell.

 

[gcf]

I confirm that by converting the disk to ntfs, my UTF8 file names are correct not only in the Samba share, but in the shell too, which I did not expect to happen.