RC2 for third_party.

[Jeremy Allison]

On Wed, Aug 13, 2014 at 07:32:09AM +0200, Volker Lendecke wrote:
> On Wed, Aug 13, 2014 at 04:01:08PM +1200, Andrew Bartlett wrote:
> > On Wed, 2014-08-13 at 05:55 +0200, Volker Lendecke wrote:
> > > On Wed, Aug 13, 2014 at 03:18:17PM +1200, Andrew Bartlett wrote:
> > > > On Mon, 2014-08-11 at 09:21 -0700, Jeremy Allison wrote:
> > > > > On Sun, Aug 10, 2014 at 08:51:37AM +0200, Andreas Schneider wrote:
> > > > > > Feel free to rebase it. The code is from libssh, but it works a bit different 
> > > > > > there.
> > > > > > 
> > > > > > http://git.libssh.org/projects/libssh.git/tree/src/config.c
> > > > > 
> > > > > Ah, it's LGPLv2.1+. I was hoping for MIT/BSD...
> > > > 
> > > > Jeremy,
> > > > 
> > > > What is the issue with bringing LGPLv2.1+ code into Samba?
> > > 
> > > pam_winbind is more liberal, and that's where we need it.
> > 
> > Just after pressing send I realised that was probably what the thought
> > was, but then I checked:  pam_winbind also links against libwbclient,
> > which is LGPLv3.
> 
> Just to throw another NIH one into the mix: Attached find tini.[ch]. I'm
> not sure it covers *all* subtleties of params.c, but I tested some of
> the obvious ones and it seems to behave the same. No dependencies, fresh
> code with the pam_winbind.c license. pam_winbind should be adaptable to
> it I guess.
> 
> What do people think?

This looks *really* nice. I'm writing a tiniparser wrapper
on top of this with the same interface as iniparser so
we can just replace it !

Look for the additional patchset soon :-).