i18n question.

[tridge at samba.org]

Kenichi,

 >                       FS CHARSET ARE NOT UTF-8.
 > 	         WE CAN NOT FIX FS CHARSET TO UTF-8.
 >         SO SAMBA HAVE TO TREAT FS CHARSET THAT ARE NOT UTF-8.

yes, and we do that now. We recommend UTF-8, but we don't require
it. If code in Samba starts to assume UTF-8 then that is a bug, and
you can report it as a bug.

We know that it is not possible to convert to UTF-8 easily, so we know
that we will need to support other filesystem charsets for a long time
to come.

 > 	!WE WANT SAMBA INTERNAL CODE TO BE FIXED TO UNICODE!
 > 	        !NO MATTER WHAT UNIX CHARSET MAY BE!

and this is where we disagree. You have not given me any convincing
arguments as to why we cannot have "unix charset" == "internal
charset". You have said that performance will be worse, but I don't
believe you. In fact, I am quite certain that allowing these two
string representations to be different will hurt performance a lot and
will greatly increase the complexity of the code.

If you provide convincing evidence as to why there will be a really
massive performance benefit in separating these, or a really
convincing argument as to why it will allow more correct operation
then I will listen to you. Shouting won't get what you want.

 > I thought you are understanding this point...

I understand what you want. I disagree. You have to convince me to
change my position. 

Cheers, Tridge