| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Brian Hurt <bhurt(at)janestcapital(dot)com> |
| Cc: | nhrcommu(at)rochester(dot)rr(dot)com, pgsql-novice(at)postgresql(dot)org |
| Subject: | Re: Polymorphic functions without a type |
| Date: | 2006-12-01 15:53:43 |
| Message-ID: | 16301.1164988423@sss.pgh.pa.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-novice |
Brian Hurt <bhurt(at)janestcapital(dot)com> writes:
> In addition to providing a text function, explicitly assigning a type
> also works, so:
> select equals(null :: int, null);
This must be so if you think about it: the meaning of equality depends
on what type you are talking about, so equals() cannot work in a vacuum.
You have to give the parser some clue what type you want the inputs to
be understood as.
In most real-world cases you're not going to be passing the thing
literal constants, but expressions involving table columns, which will
have well-defined types already. So I don't see this as a major problem
in practice. I'd avoid adding a separate equals(text,text) function,
as that's most likely not going to do much except risk "capturing" cases
you'd rather had not gotten interpreted as text equality.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | operationsengineer1 | 2006-12-02 06:31:54 | Re: Benchmarking PostgreSQL against others on Windows? |
| Previous Message | Tom Lane | 2006-12-01 15:35:07 | Re: Messed up Postgresql |