On Jun 28, 2007, at 11:10 AM, Ralf S. Engelschall wrote:

>   RPM Package Manager, CVS Repository
>   http://rpm5.org/cvs/
>    
> ______________________________________________________________________ 
> ______
>
>   Server: rpm5.org                         Name:   Ralf S. Engelschall
>   Root:   /v/rpm/cvs                       Email:  rse@rpm5.org
>   Module: rpm                              Date:   28-Jun-2007  
> 17:10:15
>   Branch: HEAD                             Handle: 2007062816101500
>
>   Modified files:
>     rpm                     CHANGES macros.in
>
>   Log:
>     Allow Berkeley-DB and SQLite specific RPM DB per-tag parameters
>     in "macros" according to the dynamically selected DB API.
>
>     No Berkeley-DB parameters are changed here! We just introduce an
>     additional macro expansion abstraction layer which allows the  
> selection
>     of the various _dbi_config[_xxxx] macros according to the  
> macros _dbapi
>     and _dbapi_rebuild.
>
>     This allows us to configure *different* sets of parameters for
>     Berkeley-DB and SQLite *at the same time* and let the right set  
> selected
>     by RPM dynamically under runtime (usually in case of a database
>     migration where both formats are used in parallel). This  
> becomes even
>     more important once we will allow RPM to be built with just  
> Berkeley-DB
>     or just SQLite support only (as SQLite doesn't understand all the
>     Berkeley-DB parameters).
>

Um, OK. A great deal of configuration complexity for the purpose  
of ... ?

If you want to be using sqlite3 instead of Berkeley DB, then please  
state directly.

There are much deeper engineering problems that need to be addressed  
if the goal
is a high performing reliable sql database. I'd rather not be reading  
endless "Don't panic ..."
nit-picky messages when the end-point of the trajectory is becoming  
increasingly clear.

73 de Jeff