Re: [HACKERS] generated columns

On 2019-02-26 06:30, Michael Paquier wrote:
> + if (attgenerated)
> + {
> + /*
> + * Generated column: Dropping anything that the generation expression
> + * refers to automatically drops the generated column.
> + */
> + recordDependencyOnSingleRelExpr(&colobject, expr, RelationGetRelid(rel),
> + DEPENDENCY_AUTO,
> + DEPENDENCY_AUTO, false);
> + }
> A CCI is not necessary I think here, still the recent thread about
> automatic dependencies with identity columns had a much similar
> pattern...

Yeah, worth taking another look.

>
> + else if (generated[0] == ATTRIBUTE_GENERATED_VIRTUAL)
> + default_str = psprintf("generated always as (%s)", PQgetvalue(res, i, attrdef_col));
> Nit: I would add VIRTUAL instead of relying on the default option.

I suppose we'll decide that when the virtual columns are actually added.
I see your point.

> Another thing I was thinking about: could it be possible to add a
> sanity check in sanity_check.sql so as a column more that one
> field in attidentity, attgenerated and atthasdef set at the same time?

There is something like that at the top of
src/test/regress/sql/generated.sql. I can expand that. But it only
covers the tests. For run time checks, you'd want something like
pg_catcheck.

--
Peter Eisentraut http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services