Re: ENABLE/DISABLE CONSTRAINT NAME

Jeff Davis-8 wrote
> Is there any semantic difference between marking a constraint as
> DISABLED and simply dropping it? Or does it just make it easier to
> re-add it later?

I cannot answer the question but if there is none then the main concern I'd
have is capturing "meta-information" about WHY such a constraint has been
disabled instead of dropped.

I guess this whole feature extends from the trigger disable feature that
already exists.  Given we have the one adding this seems symmetrical...

I cannot really see using either feature on a production system (if
following best practices) but I can imagine where they could both be helpful
during development.  Note with this usage pattern the meta-information about
"why" becomes considerably less important.

David J.




--
View this message in context:
http://postgresql.1045698.n5.nabble.com/ENABLE-DISABLE-CONSTRAINT-NAME-tp5769136p5769337.html
Sent from the PostgreSQL - hackers mailing list archive at Nabble.com.