Re: Virtual generated columns

Peter Eisentraut <peter@eisentraut.org>

From: Peter Eisentraut <peter@eisentraut.org>

To: Tom Lane <tgl@sss.pgh.pa.us>

Cc: jian he <jian.universality@gmail.com>, pgsql-hackers <pgsql-hackers@postgresql.org>, Dean Rasheed <dean.a.rasheed@gmail.com>

Date: 2025-01-08T19:28:33Z

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 →

Expand virtual generated columns for ALTER COLUMN TYPE
- 5069fef1cfae 18.0 landed
Eliminate code duplication in replace_rte_variables callbacks
- 363a6e8c6fcf 18.0 landed
Expand virtual generated columns in the planner
- 1e4351af329f 18.0 landed
Virtual generated columns
- 83ea6c54025b 18.0 landed
Additional tests for stored generated columns
- 41084409f635 18.0 landed
Improve generated_stored test
- 44b61efb7928 18.0 landed
- 86749ea3b766 18.0 landed
Fix handling of CREATE DOMAIN with GENERATED constraint syntax
- 84a67725cd11 18.0 landed
Add pg_constraint rows for not-null constraints
- 14e87ffa5c54 18.0 cited
Put generated_stored test objects in a schema
- 894be11adfa6 18.0 landed
Rename regress test generated to generated_stored
- b9ed4969250d 18.0 landed
Small code simplification
- 7ff9afbbd1df 18.0 landed
Remove useless code
- e26d313bad92 18.0 landed
Remove useless initializations
- da2aeba8f533 18.0 landed
doc: Clarify that pg_attrdef also stores generation expressions
- da486d360103 18.0 landed
Clean out column-level pg_init_privs entries when dropping tables.
- 76618097a6c0 17.0 cited
Re-implement the ereport() macro using __VA_ARGS__.
- e3a87b4991cc 13.0 cited

On 08.01.25 17:38, Tom Lane wrote:
> Peter Eisentraut <peter@eisentraut.org> writes:
>> On 03.12.24 15:15, jian he wrote:
>>> SELECT attrelid, attname, attgenerated FROM pg_attribute WHERE
>>> attgenerated IN ('v') and (attnotnull or not atthasdef);
> 
>> I don't understand what the purpose of testing attnotnull is.  That is
>> independent of attgenerated, I think.
> 
> Does it make any sense to set NOT NULL on a generated column (virtual
> or otherwise, but especially virtual)?  What is the system supposed
> to do if the expression evaluates to null?  That concern generalizes
> to any constraint really.  Even if we checked it at row storage time,
> there's no real guarantee that the expression is immutable enough
> to pass the constraint later.

The generation expression is required to be immutable.  So a table 
definition like

    a int,
    b int generated always as (a * 2) virtual,
    check (b > 0)

is not very different from

    a int,
    check (a * 2 > 0)

in terms of the constraint execution.

The current patch does not support not-null constraints, but that's 
mostly because it's not implemented yet.  Maybe that's what Jian was 
thinking about.