On Tuesday, 31. July 2007 at 3:04 pm, Jeff Johnson wrote:
>>> On Jul 30, 2007, at 2:58 PM, Thomas Lotterer wrote:
>>>> Decision:
>>>> - exclusively use Berkeley DB
>>>
> We should start separating implementation from architecture issues.
> 
Generally a good idea. In this particular case the architectural
decision is bound to the implementation because the existing code is
closely adhered to the implementation. Taking the abstraction route to
support BDB+alternative(s) would need to finally get rid of db_emu.h and
this seems to be an impossible mission to me.

Additional functionality with new code like the depsolver you have
brought up is a different story. Maybe my original proposal should be
more precise:

Decision:
- exclusively use Berkeley DB for rpmdb 

-- 
http://thomas.lotterer.net