[SAMBA4][PATCH] dd client for samba 4

[James Peach]

On Tue, 17 Jan 2006 03:08 pm, Andrew Bartlett wrote:
> On Tue, 2006-01-17 at 14:46 +1100, James Peach wrote:
> > On Tue, 17 Jan 2006 01:39 pm, Andrew Bartlett wrote:
> > > On Tue, 2006-01-17 at 13:16 +1100, James Peach wrote:
> > > > Hi all,
> > > >
> > > > The attached patch implements a dd clone that can perform IO to/from
> > > > CIFS servers. Samba 4 folks, could you please review this before I
> > > > check it in?
> > >
> > > How critical is standard DD syntax?  It seems a real pity to avoid the
> > > standard popt we provide for the rest of Samba4.
> >
> > Without dd syntax it wouldn't be a dd clone, which would defeat (some of)
> > the purpose from my point of view.
>
> Even for all the 'extra' stuff like authentication?  I would be very sad
> to see this go in without hooking on the common credentials, version and
> config handling.
>
> I suppose I just don't see the benefit in being a perfect dd clone, over
> the costs of being different.
>
> > > This would avoid your credentials problem. (Supporting the SMB url
> > > would be nice, but it sucks for conveying username, domain and password
> > > info).
> > >
> > > Also, it ensures you correctly setup debug, loadparm and many other
> > > systems.  We hook our 'must be done by every samba program' code into
> > > popt, because it's called at the 'right time'.
> >
> > Is there something I missed?
>
> I'm not sure, but if you saw quite how much pain went into ensuring that
> every Samba3 binary called load_case_tables(), I can see this utility
> suffering from bitrot.
>
> You should certainly ensure it features in 'make test'.

OK. I'd like to do this is a subsequent checkin, however.

> > > On coding style, it just doesn't seem to match that in which the rest
> > > of Samba4 is written.  Now, this hasn't been a hard and fast rule in
> > > Samba3, but in Samba4's prog_guide.txt the Linux kernel style has been
> > > specified.  (I mostly noticed the type-before-function stuff).
> >
> > OK, I've tidied up the function definitions. Anything else?
>
> Mostly it just looked 'different', but I couldn't pin it down to
> anything else.

Well, I'm happy to fix anything once you tell me what it is :)

-- 
James Peach | jpeach at sgi.com | SGI Australian Software Group
I don't speak for SGI.