[Samba] how to add unicode support to samba-2.0.6

[jack dang]

On 9/28/06, simo <idra at samba.org> wrote:
>
> On Thu, 2006-09-28 at 09:53 +0800, jack dang wrote:
> > it's a pity.
> > Japanese Version of Samba 2.2.x you mentioned does not support UNICODE
> > neither,
> > because in the source code,  all the reply functions are based on
> > single-byte string processing function such as strlen() , it cannot
> handle
> > the incoming multibyte strings.
> >
> > Should I really have to take pains to substitute all the single-byte
> string
> > processing functions with mutibyte string processing function, and to do
> a
> > ASCII-UNICODE string conversion everytime the server recieves an
> incoming
> > packet that contains an ASCII encoded string?
>
> This is what samba does today.
>
> > Folks, any ideas of any simple ways?
>
> no, it is probably better for you to try to remove functionality you do
> not need in samba3 (samnba4 ?) to reduce the size of the binary than to
> try and port an ancient version like 2.0.6 to unicode.
>
> Simo.
>
> --
> Simo Sorce
> Samba Team GPL Compliance Officer
> email: idra at samba.org
> http://samba.org
>
>

I've thought of that before, but the overall binary size (smbd + nmbd, not
to mention some basic authentication functionality I may need afterwards)is
over 50M, nearly 60M.
Do you think it is poosible to reduce it to a less-than-2M size?
I suppose it may require much more modifications anyhow because samba3 is
far more complicated than samba2, and I have to figure out the functionality
of every single component and the annoyingly complicated inter-connection
and interdependency among them to make an effective modification.
well, that's my point of view, and if you have any further advice, would you
take patience to tell me?