From: | Jaime Casanova <jaime(at)2ndquadrant(dot)com> |
---|---|
To: | Darren Duncan <darren(at)darrenduncan(dot)net> |
Cc: | PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: DOMAINs and CASTs |
Date: | 2011-05-15 07:26:16 |
Message-ID: | BANLkTik46G=KRKgdVYru+VxZaViC0cZacQ@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Sun, May 15, 2011 at 2:13 AM, Darren Duncan <darren(at)darrenduncan(dot)net> wrote:
>
> ('1800-01-01 00:00:00'::timestamp)::int
>
> Now, since all values of a DOMAIN are also values of the base type the
> DOMAIN is defined as being a subset of, then the sub-expression within the
> parenthesis denotes a value that is both a timestamp and a datetime at the
> same time.
>
> So, if a generic "CAST(timestamp as int)" is already defined, and you define
> a "CAST(datetime as int)", then what should the above code (correct for
> misspelling) do, or should it fail?
>
Obviously it should run the cast from timestamp to int, why it will
run a cast from a domain?
the other way should be allowed, though... a cast from datetime to int
should first look for cast function using the domain and if it don't
find it then with base type
--
Jaime Casanova www.2ndQuadrant.com
Professional PostgreSQL: Soporte y capacitación de PostgreSQL
From | Date | Subject | |
---|---|---|---|
Next Message | Robert Haas | 2011-05-15 18:12:11 | Re: Reducing overhead of frequent table locks |
Previous Message | Darren Duncan | 2011-05-15 07:13:16 | Re: DOMAINs and CASTs |