Re: ODBC SQLColums is very slow

From: Wolfram Roesler <wr_at_spam.la>
Date: Tue, 27 Jan 2009 17:04:05 +0000 (UTC)
Message-ID: <Xns9BA0B7CD9B7B1wrgrpde_at_news.albasani.net>

Charles Hooper <hooperc2000_at_yahoo.com> wrote in news:5f84af51-f214-4617- aef4-3af34b82d332_at_r36g2000prf.googlegroups.com:

> If the OP is able to offer a suggestion to the programmers, a trick
> that I have used in the past so that I have had no need for SQLColumns
> is to select into a recordset:
> SELECT
> *
> FROM
> MYTABLE
> WHERE
> ROWNUM=1;
> (or possible ROWNUM<1)

That looks like a nice alternative to SQLColumns. I think it would not be necessary to select anything (i. e. one could use a where clause that matches nothing). Perhaps WHERE 0=1 would even be more efficient than using ROWNUM? Would a FIRST_ROWS hint help? We are dealing with some big tables here, as well as with some views that point to DB links, so I'd like to use a query that accesses the data dictionary only without touching the actual table data.

Thanks and best regards
Wolfram Roesler Received on Tue Jan 27 2009 - 11:04:05 CST