"Ken Winter" <[EMAIL PROTECTED]> writes:
> My question is: Why didn't the chunk of "gen_sequences" code that consults
> pg_catalog find a record of "e_mail_address_invisible_id_seq", and thereby
> refrain from trying to create it again?
I added a few "raise notice" commands to your function, and got this:
NOTICE: sub_idcol = invisible_id
NOTICE: default_exp = nextval('public.e_mail_address_invisible_id_seq'::text)
NOTICE: sequence_name = public.e_mail_address_invisible_id_seq
NOTICE: not found
NOTICE: sub_idcol = pop_id
NOTICE: default_exp = nextval('pop_seq'::text)
NOTICE: sequence_name = pop_seq
NOTICE: found
ERROR: relation "e_mail_address_invisible_id_seq" already exists
CONTEXT: SQL statement "CREATE SEQUENCE
public.e_mail_address_invisible_id_seq;"
PL/pgSQL function "gen_sequences" line 51 at execute statement
The problem seems to be that you're not accounting for a schema name
possibly appearing in nextval's argument.
regards, tom lane
---------------------------(end of broadcast)---------------------------
TIP 2: Don't 'kill -9' the postmaster
