Tom Lane <tgl@sss.pgh.pa.us> writes:
> Greg Stark <gsstark@mit.edu> writes:
> > If it matches an SQL keyword after being downcased the old fashioned way, then
> > it's an SQL keyword. If not then the locale-aware tolower() would be
> > appropriate for tables, columns, etc.
>
> That's exactly what we do already. The complaint was that the
> locale-aware downcasing is broken (not to put it too finely) in Turkish
> locales, leading to unexpected/unwanted results for identifiers that are
> not keywords.
But the example given was "SERIAL". "serial" is an English word, not a Turkish
word. It shouldn't really be subject to Turkish locale effects at all. Perhaps
"keyword" wasn't the right word in my message.
I'm wondering if he really expects all identifiers to be subject to this ascii
downcasing. Like, if he had a GÜNAYDIN column he might be surprised to when
günaydýn (where ý is the lowercase dotless i) says column "günaydýn" doesn't
exist.
Or is the real problem simply that both styles of i really ought to match all
the time, ie, that they should really be considered the same letter for
matches? I wonder if there are other locales where that's an issue.
--
greg
