Re: [HACKERS] [BUGS] BUG #2907: pg_get_serial_sequence quoting

Adriaan van Os wrote:
> Tom Lane wrote:
> > Bruce Momjian <bruce(at)momjian(dot)us> writes:
> >>> I presume the reason for that is that the first paramater can be
> >>> qualified:
> >>> select pg_get_serial_sequence('"public"."FOO"', 'Ff1');
> >
> >> Would someone explain why qualification makes us lowercase the first
> >> parameter by default? I don't understand it well enough to document it.
> >
> > The point is that we have to parse the first parameter, whereas the
> > second one can be taken literally.
>
> It still looks inconsistent and ugly. I think the design mistake of pg_get_serial_sequence is that
> it takes two parameters rather than one (a fully qualified doublequoted columnname path) or three
> (optionally empty schema, tablename, columnname, all three literal).

I did my best to document the behavior of pg_get_serial_sequence().
There actually is a technical reason why we can't auto-quote the first
parameter. Patch applied to HEAD and 8.2.X.

Example of identifiers with embedded periods:

test=> CREATE TABLE "a.b" ("c.d" SERIAL);
NOTICE: CREATE TABLE will create implicit sequence "a.b_c.d_seq" for serial column "a.b.c.d"
CREATE TABLE

test=> SELECT pg_get_serial_sequence('a.b', 'c.d');
ERROR: schema "a" does not exist

test=> SELECT pg_get_serial_sequence('"a.b"', 'c.d');
pg_get_serial_sequence
------------------------
PUBLIC."a.b_c.d_seq"
(1 row)

test=> SELECT pg_get_serial_sequence('"a.b"', '"c.d"');
ERROR: column ""c.""d of relation "a.b" does not exist

--
Bruce Momjian bruce(at)momjian(dot)us
EnterpriseDB http://www.enterprisedb.com

+ If your life is a hard drive, Christ can be your backup. +