>
> Bruce Momjian writes ("Re: [HACKERS] Case sensitivity in identifiers"):
> > > Right now saying '\d FOO' says 'Couldn't find table FOO!' so somebody
> > > missed a bit of case-sensitivity code somewhere, but I'm sure you see
> >
> > I will check on this. Looks like a bug.
>
> Cheers. Incidentally I wish \d tablename wouldn't complain if I stuck
> a ; on the end, just because I get in to the habit inside psql, but I
> know the ; is only supposed to terminate SQL statements which \d isn't.
I do the same thing. Would anything break if I removed a trailing ';'
from \d commands? Anyone?
>
> > You have a good point here. I can see a value in keeping case when
> > creating fields, but allowing any case to access it. Is that the basis
> > of the argument.
>
> Yes indeed.
>
> > Right now, we lowercase arguments when the come in from the parser. We
> > could preserve case for Create-type statements, and force lowercase on
> > all comparisons, but that is a big job, and I think we questioned the
> > value of it.
>
> I wouldn't have thought it was a particularly big job; you keep these
> identifiers in postgres tables anyway: why not a new batch of char
> types and operators using strcasecmp instead of strcmp?
>
> I'm suggesting that's not a big job because if you've structured your
> code right (of course you have!) you just copy all the files, do a
> spot of search and replace, and possibly some minor tweaking. Maybe
> even just two new functions (compare ignoring case - or don't you have
> it already?) plus a suite of case-insensitive character operators to
> sort out indexing.
>
> Some end users might even find such new functionality useful...
It is not that easy. We use indexes and cached keys to look up things.
Doing such things case-insensitive is a big job.
- --
Bruce Momjian
maillist@candle.pha.pa.us
------------------------------
