Re: [Gnumed-devel] document control

[Karsten Hilbert]

> create a ancestor table between lab_result and clin_root_item
You'll run into big trouble with deficient PG inheritance.
Can't name them off the top of my head but remember unkind
things when you go beyond one level of inheritance. One thing
is that you start wishing to write to tables somewhere
mid-hierarchy which is a no-no.


, similarly
> for lab_request, which contains the control fields, leaving the
> lab-specific stuff in the descendant tables. Then we can have other
> descendants, linking to med_doc, for other types of communications.
measurements/lab and docs are in separate databases

> The advantage is it leaves the columns of lab_result and lab_request
> unchanged, and I'm happy to revert gmBlobs.sql, so no alteration to 
> client code is required.
> 
> You can't say fairer than that ;-)
ack but IMO (unfortunately) not the solution

Karsten
-- 
GPG key ID E4071346 @ wwwkeys.pgp.net
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346