This is a discussion on Re: thw rewriter and default values, again within the pgsql Hackers forums, part of the PostgreSQL category; --> Jaime Casanova <systemguards@gmail.com> writes: > create rule ins_rule as on insert to v_foo do instead > insert into foo(col1, ...
| |||||||
| FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
| ||||
| Jaime Casanova <systemguards@gmail.com> writes: > create rule ins_rule as on insert to v_foo do instead > insert into foo(col1, col2) values (new.col1, new.col2); > insert into v_foo(col2) values (1); > this give an error like: > psql:f:/views.sql:13: ERROR: null value in column "col1" violates > not-null constraint That's not a bug, and "fixing" it isn't acceptable. The correct solution to the problem you are looking at is to attach default expressions to the view itself. Adding alter table v_foo alter col1 set default nextval('public.foo_col1_seq'); to your example makes it work as you wish. regards, tom lane ---------------------------(end of broadcast)--------------------------- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq |
| |||
| On 5/28/05, Tom Lane <tgl@sss.pgh.pa.us> wrote: > Jaime Casanova <systemguards@gmail.com> writes: > > create rule ins_rule as on insert to v_foo do instead > > insert into foo(col1, col2) values (new.col1, new.col2); > > > insert into v_foo(col2) values (1); > > > this give an error like: > > psql:f:/views.sql:13: ERROR: null value in column "col1" violates > > not-null constraint > > That's not a bug, and "fixing" it isn't acceptable. > > The correct solution to the problem you are looking at is > to attach default expressions to the view itself. Adding > > alter table v_foo alter col1 set default nextval('public.foo_col1_seq'); > > to your example makes it work as you wish. > I know you're right, but -when dealing with updateable views- doing that implies to add a lot of time altering views when base table change, and of course we maybe don't want all views get that values. -- regards, Jaime Casanova (DBA: DataBase Aniquilator ---------------------------(end of broadcast)--------------------------- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to majordomo@postgresql.org so that your message can get through to the mailing list cleanly |
| |||
| Jaime Casanova wrote: > I know you're right, but -when dealing with updateable views- doing > that implies to add a lot of time altering views when base table > change, and of course we maybe don't want all views get that values. Sorry for the late posting on this thread. Might there be any way to have something like ...SET DEFAULT pg_same_as('public','foo','col1'). Hmm - you'd need pg_same_as_int/text/etc. depending on the return-type - still not ideal. -- Richard Huxton Archonet Ltd ---------------------------(end of broadcast)--------------------------- TIP 7: don't forget to increase your free space map settings |
| ||||
| On 6/6/05, Richard Huxton <dev@archonet.com> wrote: > Jaime Casanova wrote: > > I know you're right, but -when dealing with updateable views- doing > > that implies to add a lot of time altering views when base table > > change, and of course we maybe don't want all views get that values. > > Sorry for the late posting on this thread. > > Might there be any way to have something like ...SET DEFAULT > pg_same_as('public','foo','col1'). > > Hmm - you'd need pg_same_as_int/text/etc. depending on the return-type - > still not ideal. > Actually, i try to do that but i need the function to be polymorphic and because polymorphic functions needs to receive at least one polymorphic argument that teach about the return type of the function i can't go ahead with this idea. About the code i talk earler in this thread i solve the problem with the update to serial columns. -- Atentamente, Jaime Casanova (DBA: DataBase Aniquilator ---------------------------(end of broadcast)--------------------------- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq |