Re: NOT ENFORCED constraint feature

On Thu, Mar 6, 2025 at 9:37 PM Alexandra Wang
<alexandra(dot)wang(dot)oss(at)gmail(dot)com> wrote:
>
> Hi Amul,
>
> On Thu, Feb 27, 2025 at 12:57 AM Amul Sul <sulamul(at)gmail(dot)com> wrote:
>>
>> Attached is the rebased patch set against the latest master head,
>> which also includes a *new* refactoring patch (0001). In this patch,
>> I’ve re-added ATExecAlterChildConstr(), which is required for the main
>> feature patch (0008) to handle recursion from different places while
>> altering enforceability.
>
>
> Thanks for the patches!
>
> I reviewed and ran “make check” on each patch. I appreciate how the
> patches are organized; separating the refactors from the
> implementations made the review process very straightforward.

Thank you for the feedback and the review !

> Overall, LGTM, and I have minor comments below:
>
> 0008
> Since we are added "convalidated" in some of the constraints tests,
> should we also add a "convalidated" field in the "table_constraints"
> system view defined in src/backend/catalog/information_schema.sql? If
> we do that, we'd also need to update the documentation for this view.
>

I am not sure why we don't already have "convalidated" in the
table_constraints, but if we need it, we can add it separately.

> 0009
> Comment on top of the function ATExecAlterConstrEnforceability():
> s/ATExecAlterConstrRecurse/ATExecAlterConstraintInternal/g
>
> Typo in tablecmds.c: s/droping/dropping, s/ke/key
> /* We should be droping trigger related to foreign ke constraint */
>

Thanks, fixed in the attached version.

Regards,
Amul