> Hi!
>
> Yes, you are near of the truth :). But if I use some complicated joining
> query to few large tables I think this is not too comfortable for the
> server.
>
>
Well, this is the line between mere programming and software
engineering.
The simplest technique is the most straightforward -- trade memory for
time
my $data = $dbh->selectall_arrayref( q ( select *
from table1
join table2
where complex_function() ) ,
{ Columns => {} } );
This is one of my All Time Favorite idioms .. gives you a data
structure of the form:
$data = [ { first_column => 'aubergine' ,
second_column => 'baklazhan' } ,
{ first_column => 'hello',
second_column => 'privyet' } ];
Another technique would be (if your database supports it) is push a
PART of the load off to the db server.
$dbh->do ( q ( select into temp table _results ...complex-sql.... ) );
my $sth = $dbh->prepare( q ( select * from _results ) );
if ( my $cursor = $sth->execute ) {
while (my $row = $sth->fetchrow_hashref) {
.
.
.
}
}
Of course, this just trades the KNOWN quantity of "straightforward
memory for time" for a relatively unknown tradeoff on the database
server.
-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Lawrence Statton - [EMAIL PROTECTED] s/aba/c/g
Computer software consists of only two components: ones and
zeros, in roughly equal proportions. All that is required is to
sort them into the correct order.
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
<http://learn.perl.org/> <http://learn.perl.org/first-response>
