On 11/7/07, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>
> A more general objection is that causing query semantics to change in
> subtle ways based on a GUC variable has more often than not proven to be
> a bad idea.
On top of that, this is another one of those conversations that
basically are predicated on the premise that other databases have
quirks that make / encourage / allow the user to write bad SQL, and we
need to do something so that their bad SQL will run properly on
PostgreSQL.
I work with 3 Oracle DBAs, and they are all trained by Oracle (the
database, not the company) to write queries that make my brain hurt.
Case statement? nope, they use encode. And there are dozens of cases
where they use non-standard SQL, and they aren't going to stop any
time soon, because it's just what they know.
As someone who wishes we could switch case folding easily from lower
to upper for some use cases, I understand the desire of folks to want
things in pgsql to be easily switchable to fix these kinds of issues.
But I don't think most of them are worth the effort and the bugs that
could be introduced.
