| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Fernando Nasser <fnasser(at)redhat(dot)com> |
| Cc: | Zeugswetter Andreas SB SD <ZeugswetterA(at)spardat(dot)at>, pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: Schemas vs. PostQUEL: resolving qualified identifiers |
| Date: | 2002-01-23 19:00:00 |
| Message-ID: | 4503.1011812400@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Fernando Nasser <fnasser(at)redhat(dot)com> writes:
> Tom Lane wrote:
>> Okay, but then how will you refer unambiguously to the rowtype object?
> What about casting with the keyord ROW?
> func(ROW table)
> always refers to the row-type of table "table" even if there is
> a column called "table".
Strikes me as gratuituously different from the way everything else is
done. We have .* and %ROWTYPE and so forth, and they're all suffixes.
The closest analogy to your ROW syntax is CAST, but it doesn't alter the
initial interpretation of its argument.
I was toying with the notion of inventing some new notation like
table.**
I don't like double-asterisk much, but maybe there's some other symbol
we could use here?
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Bruce Momjian | 2002-01-23 19:10:27 | Re: LWLock contention: I think I understand the problem |
| Previous Message | Bruce Momjian | 2002-01-23 18:19:05 | Savepoints |