> On Mon, Feb 9, 2026 at 5:55 PM Cédric Villemain
<[email protected]> wrote:
I have some questions/comments (no change on status, I didn't test)
- FK are constraint triggers, does it makes sense to align
"ATExecAlterFKConstrEnforceability" renaming with other functions like
"AlterConstrTriggerDeferrability" ?
do you mean changing
ATExecAlterConstrDeferrability
to
ATExecAlterFKConstrDeferrability
?
I mean to keep the ConstrTrigger part in the name, in the idea that it's
the same feature (FK are "just" specialized constraints triggers IIUC).
If so, it makes sense and also improves readability, IMO.
However, since only FK supports changing deferrability—and we are not
modifying deferrability here,
the incentive for this renaming change is kind of lower.
sure.
- I also wonder if it makes sense to manage NOT NULL together with
CHECK, like in ATAddCheckNNConstraint.... ?
See ATExecAlterConstraintInternal.
We do not support changing enforceability of NOT NULL, since NOT NULL
NOT ENFORCED is not supported.
I do have a patch for NOT NULL NOT ENFORCED,
https://commitfest.postgresql.org/patch/6029
Ah great, tks.
I'll check your CF entries instead of adding noise here (I was wondering
if already had a patch to cleanup processCASbits() in gram.y).
--
Cédric Villemain +33 6 20 30 22 52
https://www.Data-Bene.io
PostgreSQL Support, Expertise, Training, R&D
