Horton

Re: Schema variables - new implementation for Postgres 15

Julien Rouhaud <rjuju123@gmail.com>

View thread
From: Julien Rouhaud <rjuju123@gmail.com>
To: Pavel Stehule <pavel.stehule@gmail.com>
Cc: Dean Rasheed <dean.a.rasheed@gmail.com>, Joel Jacobson <joel@compiler.org>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-03-25T04:18:42Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Allow underscores in integer and numeric constants.

  2. Remove special outfuncs/readfuncs handling of RangeVar.catalogname.

  3. Remove extra space from dumped ALTER DEFAULT PRIVILEGES.

  4. Create FKs properly when attaching table as partition

  5. psql: improve tab-complete's handling of variant SQL names.

Hi,

On Wed, Mar 23, 2022 at 09:58:59PM +0100, Pavel Stehule wrote:
> 
> A bit more work seems to be needed for deparsing session variables:
> >
> > # create variable myvar text;
> > CREATE VARIABLE
> >
> > # create view myview as select myvar;
> > CREATE VIEW
> >
> > # \d+ myview
> >                           View "public.myview"
> >  Column | Type | Collation | Nullable | Default | Storage  | Description
> > --------+------+-----------+----------+---------+----------+-------------
> >  myvar  | text |           |          |         | extended |
> > View definition:
> >  SELECT myvar AS myvar;
> >
> > There shouldn't be an explicit alias I think.
> >
> 
> I check this issue, and I afraid so it is not fixable. The target list
> entry related to session variable has not some magic value like ?column?
> that can be used for check if tle->resname is implicit or explicit
> 
> And in this time I cannot to use FigureColname because it doesn't work with
> transformed nodes. More - the Param node can be nested in SubscriptingRef
> or FieldSelect. It doesn't work perfectly now. See following example:
> 
> create type xt as (a int, b int);
> create view b as select (10, ((random()*100)::int)::xt).b;
> \d+ b
> SELECT (ROW(10, (random() * 100::double precision)::integer)::xt).b AS b;

Fair enough.  Since there is other code that already behaves the same I agree
that it's better to not add special cases in ruleutils.c and have an explicit
alias in the deparsed view, which isn't incorrect.