Hi Ted,
Ted Byers wrote:
Thanks Uwe
This is a great start. It reduces the dump from 2 MB
down to 167K, but out of 6833 lines of SQL, 5744
relate to the public schema in the DB, and I didn't
touch that. It has over a dozen types, 419 functions,
&c., that were put there by postgresql the moment I
created the database. I'd expect the same stuff to be
there the moment I issue the create database directive
on the host machine, so all I really want is the dozen
sequences, two dozen tables, and the suite of
constraints I created, all in the schema specific to
my new DB.
Well thats usually not the case unless you changed
the default database per accident. You can hope but
not be sure to find the same situation on your
server.
Is there a reason pg_dump dumps the stuff in public
even though that stuff seems to be created, and
therefore present, in every database I create on a
given server instance? Isn't that duplication a waste
of space, and it's presence in the dump a waste of CPU
cycles?
Well, at the moment you seem to waste CPU cycles, network
bandwith and storage on the mailinglist server by not
just looking at the manual of pg_dump, which has for example
goodies as:
-n schema
--schema=schema
Dump only schemas matching schema; this selects both the
...
HTH ;)
Tino
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
