Re: General question about DBI/DBD

Paul Tomblin <ptomblin(at)xcski(dot)com> writes:

[converting Pg.pm to DBD::Pg]

> Also, the big program does a whole bunch of
> $conn->exec("SELECT * FROM tablename WHERE indexed_column = '$key'");
> and things like that. Is there a way in DBD:Pg I can re-parse those
> queries with "...indexed_column = :1" and then supply a different key each
> time? I remember when I used to write C programs that accessed Oracle
> (about 8 years ago, so the details are hazy), that was a major performance
> improvement.

You can basically do:

$stmt = $dbh->prepare("SELECT * FROM tablename WHERE indexed_column = ?");

$stmt->execute($key1);
$stmt->execute($key2);

This is nice because it will do any escaping in $key that needs to be
done.

However, PG doesn't cache query plans in this situation (unlike
Oracle) so you won't see a big performance win.

(Apologies if my syntax is off; I mostly use Java now.)

-Doug
--
Let us cross over the river, and rest under the shade of the trees.
--T. J. Jackson, 1863