generic sql [wasRe: The Gnucash database?]

[Linas Vepstas]

On Fri, Jul 23, 2004 at 10:35:36AM -0400, Jean-David Beyer was heard to remark:
> Linas Vepstas wrote:
> >On Thu, Jul 22, 2004 at 10:52:05AM -0400, Jean-David Beyer was heard to 
> >remark:
> >
> >>I normally just use the embedded SQL in a C++ program. I assume the 
> >
> >I'm pursuing a ver very different strategy.  The design assumes that
> >none of the tables and none of the objects are known at compile time, 
> >these are all discovered at run time. 
> 
> Do you perceive a large benefit to be gained from this complexity?

yes, huge.

I work with other apps to, not just gnucash.  I've done the embedded
approach, and after about your 3nd app and 30th table it gets old.
There is no real need for all that crud, it should be very very easy 
map objects to tables, and not be an error-prone process.

For example: gnucash has had scheduled txns & business objects for 3-4
years now; in all that time, no one has bothered to add these to the 
existing sql backend. Why?  Too complex.  I'm trying to simplify all 
this.

--linas

-- 
pub  1024D/01045933 2001-02-01 Linas Vepstas (Labas!) <linas at linas.org>
PGP Key fingerprint = 8305 2521 6000 0B5E 8984  3F54 64A9 9A82 0104 5933