Re: Identifier case folding notes

>>> Josh Berkus <josh(at)agliodbs(dot)com> wrote:

> Actually, there are a number of *very* popular database tools,
particularly
> in the Java world (such as Netbeans and BIRT) which do mix quoted and

> unquoted identifiers. In general, users of those tools reject
PostgreSQL
> as "broken" for our nonstandard behavoir rather than trying to work
around
> it.

Do these tools expect an unquoted identifier to be treated according
to the standard? As I read it, an unquoted identifier should be
treated identically to the same identifier folded to uppercase and
wrapped in quotes, except that it will be guaranteed to be considered
an identifier, rather than possibly considered as a reserved word,
etc.

>From our perspective, we're OK with the status quo since we always
quote all identifiers. I don't think any of the suggestions would
bite us (if implemented bug-free) because we also forbid names which
differ only in capitalization. We help out our programmers by letting
them ignore quoting (except identifiers which are reserved words) and
capitalization when they write queries in our tool; we correct the
capitalization and wrap the identifiers in quotes as we generate the
Java query classes. Doing something like that in psql autocompletion
and in other PostgreSQL tools would be a nice feature, if practicable.

-Kevin