Re: bogus error message for ALTER TABLE ALTER CONSTRAINT

On 2025-Mar-11, Amul Sul wrote:

> I was thinking of something like the attached, which includes your
> test cases from 0001. Perhaps the macro name could be improved.

FWIW I like this general idea.  I don't like the proposed names much
though, especially the abuse of ALL_CAPS; and because they operate on
the given bits themselves rather than the output of processCASbits(), I
would have these checks before doing anything else.  (Also, for nicer
code layout I would perhaps make the macros static inline functions.)

I'm going to stay away from this for a bit, as I think this is of
somewhat secondary importance.

-- 
Álvaro Herrera         PostgreSQL Developer  —  https://www.EnterpriseDB.com/
"Every machine is a smoke machine if you operate it wrong enough."
https://twitter.com/libseybieda/status/1541673325781196801