[Dropping source-changes ...]

itojun@iijlab.net writes:

>>>2) I have had a look at the BSD auth implementation proposal which you
>>>   posted in [3].  All BSD auth logic resides in libutil, and there is
>>>   no reason to have pw_dup(3) separated from it in libc; neither does
>>>   pw_dup(3) access any libc internals in which case it being
>>>   implemented in libc would be a sensible choice.  It's also worth
>>>   noting that the bulk of password database utility functions already
>>>   reside in libutil.
>>
>>	it was a mistake.  all auth*.c has to reside in libc.
>
> 	therefore pw_dup() has to be in libc.

Ok. Can you please clarify why in particular BSD auth needs to reside
libc?  A glimpse over your diff dated the same as pw_dup(3) didn't
reveal any obvious dependency on residing in libc.  I noticed the BSD
auth code actually depends on libutil code (password database update
functions, login.conf).  And beyond that, in your original proposal
[3] you wrote that "it is working very well".

(Again, I'm afraid all this doesn't help my impression of a premature
integration attempt of BSD auth.)


- Klaus

>>> [3] http://mail-index.netbsd.org/tech-userlevel/2003/09/09/0001.html