As many as you want and your hardware supports.
It is directly supported and very easy to have one database
per version on one machine, eg. v7, v8 and v9. This is done
when upgrading to a new version anyway. This can be used to
keep a "real" patient database and a next-version database
for testing the next version of GNUmed on the same machine.
Ok so how do I access the old DB? For example I want to to start working with the 0.3 client on my new DB, gnumed_v9. Say in a few months we can make it to GNUmed 1.0 with gnumed_1.0 (: see how fast I think we are going to move? :), I want to know how to access the "old" v9, say something bad happens with the bootsptrapping.
It is also possible to have several databases of the same
version but that needs manual work: They need to have
different names (for example gnumed_v8_1 and gnumed_v8_2).
One needs to make sure that database permissions,
bootstrapping scripts, config files etc are adjusted to the
corresponding database name appropriately.
My problem is that if I upgrade to DB v9 on my actual computer now with the stable 0.3 version, the CVS will likely be termed v10 ... but when I upgrade again ... say to GNUmed 1.0 ;) ... what happens to the v10 DB? Won´t it be conflicting with the one already being used by the CVS? Are they diferent? Did I make sense?
I had this discussion in the devel list a few days ago ... but I am still confused ... in CVS the local database is installed INSIDE the CVS folder? And in the .deb package where is it?
Thanks
Rogerio