On Fri, Dec 12, 2008 at 10:31 AM, Greg Stark <stark@enterprisedb.com> wrote:
> On Fri, Dec 12, 2008 at 3:11 PM, Dimitri Fontaine
> <dfontaine@hi-media.com> wrote:
>> That's why I'm preferring the common-lisp syntax of :param value, or its
>> variant param: value.
>
> FWIW there is no such common-lisp syntax. Colon is just a regular
> symbol character and :param is just a regular symbol in common-lisp.
> There is a convention that functions parse their argument lists
> looking for such tokens as indicators of what to do with the next
> argument but it's purely a convention. There's no syntactic
> significance to the colon.
Drifting off-topic and being really nit-picky, you're wrong. :) It's
more than just a "convention". Colons *are* special in symbol
names--the leading colon designates the symbol as being in the KEYWORD
package (i.e. symbol namespace; you can put symbols in other packages
by prepending a package name to the colon) and there is standard
syntax (&key) for specifying allowed keyword arguments to a function;
said keys must be symbols in the KEYWORD package.
So the proposed foo( :bar 12, :baz 'stuff' ) syntax is actually very
close to the Common Lisp syntax, though there may be very good reasons
not to use it in PG.
> A similar problem arises with using Perl as a precedent. => is just a
> regular operator in perl which quotes the lhs as a string if it's a
> simple token and otherwise behaves just like a comma. That would be
> very different from what we're talking about having it do here.
Very true, and I think the "don't break people who are using => as a
prefix operator" argument is pretty conclusive.
-Doug
