On Thu, Jun 20, 2013 at 12:34 PM, Amit Langote <amitlangote09@xxxxxxxxx> wrote: > On Thu, Jun 20, 2013 at 11:10 AM, Chris Angelico <rosuav@xxxxxxxxx> wrote: >> On Thu, Jun 20, 2013 at 12:09 PM, Amit Langote <amitlangote09@xxxxxxxxx> wrote: >>> Umm, my bad! I almost forgot I could write pure SQL function bodies. >>> Although, why does following happen? (sorry, a 8.4.2 installation) : >>> >>> postgres=# create or replace function gt(n int, m int) returns boolean >>> as 'select n>m' language sql; >>> ERROR: column "n" does not exist >>> LINE 2: as 'select n>m' language sql; >> >> Hmm, no idea. I'm using 9.2.4, could well have been changes. >> > > Hmm, I guess in 8.4.2, one needs to refer to function arguments as $1, $2 ... Ah, okay. I'm not all that familiar with different versions of PostgreSQL; I used it first back in the 1990s, then didn't use it for years (was all DB2), and now picked it up again at version 9.0/9.1, moving to 9.2 when it came out. > And yes. OP can go ahead with his migration using this suggested > wrapping function idea. Absolutely! ChrisA -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
