Re: Cluster wide option to control symbol case folding

"Lewis, Ian \(Microstar Laboratories\)" <ilewis(at)mstarlabs(dot)com> writes:
> Is there any chance that the PostgreSQL developers would accept a new
> cluster wide configuration option to control how the system handles
> symbol case folding?

No. This has been looked into repeatedly in the past, and we simply
don't want to deal with it. Quite aside from the impact on the server
(which would be extensive), it would break every nontrivial application,
and force them all to try to deal with each possible folding behavior.
The more behaviors there are, the worse that gets. Pretty soon, every
application is double-quoting every identifier out of sheer paranoia.
Which is exactly the behavior you say you'd rather avoid --- but not
only have you not avoided it, you've forced the entire Postgres
ecosystem into it.

The closest precedent that ever actually got into the server was the
transaction-autocommit option that existed for awhile circa PG 7.3.
Once we realized just how much complexity we were forcing on every
application, we took that out again. If somehow we forgot that
episode and agreed to take a case-folding behavioral change, I'm
pretty sure the same dynamics would play out again.

There's a lot of material on this in the archives. The latest
substantive discussion I can find, which includes links to several
previous investigations, is here:
https://www.postgresql.org/message-id/flat/200807081925.40467.peter_e%40gmx.net

regards, tom lane