On Thu, 2011-06-30 at 09:11 +0200, Florian Pflug wrote:
> > How would the system catalogs be initialized under that theory: surely
> > you're not going to seed (nr. of types) * (nr. of collations) * (nr. of
> > opclasses) range types in initdb?
> There's CREATE RANGE.
Right. In that respect, it's more like a record type: many possible
record types exist, but you only define the ones you want.
> By default, no range types would exists I believe.
I was planning to include _some_ by default. Probably not text ranges,
but integer and timestamp[tz] ranges. If nothing else, it makes it
easier to document.
In response to
pgsql-hackers by date
|Next:||From: Jeff Davis||Date: 2011-06-30 16:34:18|
|Subject: Re: Range Types, constructors, and the type system|
|Previous:||From: Alvaro Herrera||Date: 2011-06-30 16:00:34|
|Subject: Re: creating CHECK constraints as NOT VALID|