I have just read this explaining about how to structure data and functions
within postgresql
http://blog.cleverelephant.ca/2010/09/postgis-back-up-restore.html
..."The "public" schema is where the PostGIS functions and system tables get
installed, so if you dump that schema you get all those definitions in your
dump. If those definitions are mixed in amongst your data, loading them into a
fresh database gets tricky: are the paths to the libraries the same? are there
function name clashes? (The utils/postgis_restore.pl script attempts manfully
to strip out PostGIS components from a dump file to allow a clean restore, but
it is hard to get 100% performance.)
If, on the other hand, all your data is neatly separated into its own schema,
you can neatly backup just that schema and avoid having PostGIS system
information mixed in with your data. That means you can easily restore your
data into any version of PostGIS and PostgreSQL that you like. So upgrades are
easy easy easy.
Remember: Store your data in a schema other than "public".
"
Basically Paul recommends saving geodata in a different schema to the postgis
functions.
My questions are...
1"if the data is located in a different schema which does not have the 800 odd
postgis functions, are the functions still available to the data?"
2"are cross schema queries allowed?"
3 "does it also make sense to seperate non-spatial tables into their own
schemas?"
thanks for any advice,
Rob
_______________________________________________
postgis-users mailing list
postgis-users@postgis.refractions.net
http://postgis.refractions.net/mailman/listinfo/postgis-users
