Select one random row from cycle of function with SQLite and Perl - database

Question

Hi I tried to select one random number from this:
My source:
use DBI;
use CGI;
my $file = '.\input.txt'; # Name the file
open(FILE, $file) or die("Unable to open file");
my #data = <FILE>;
foreach my $line (#data)
{
chomp $line
my $sth = $dbh->prepare("SELECT columnA FROM table WHERE columnA LIKE '%$line%'");
$sth->execute;
my $result = $sth->fetchall_arrayref;
foreach my $row ( #$result ) {
print "- ";
print "#$row\n";
print "<BR />";
}
}
How can I print only ONE RANDOM row???
I tried something like that:
my $sth = $dbh->prepare("SELECT nazov_receptu FROM recepty WHERE nazov_receptu LIKE '%$line%' AND kategoria == 'p' AND (rowid = (abs(random()) % (select max(rowid)+1 from recepty)) or rowid = (select max(rowid) from recepty)) order by rowid limit 1;");
but its not clear... i dont know why...
I am using SQLite and printing it to web interface.
You can try it when you have
input.txt:
A
C
database:
id name
1 A
2 B
3 C
4 D
5 E
OUT:
A OR C (random)

Answer 1

Why not join the file arguments into the query right away instead of looping over them? Then it is a simple matter to extract a random index in perl:
use strict;
use warnings; # Always use these two pragmas
my $file = '.\input.txt';
open my $fh, "<", $file or die "Unable to open file: $!";
chomp(my #data = <$fh>); # chomp all lines at once
my $query = "SELECT columnA FROM table WHERE ";
$query .= join " OR ", ( "columnA LIKE ?" ) x #data;
# add placeholder for each line
#data = map "%$_%", #data; # add wildcards
my $sth = $dbh->prepare($query);
$sth->execute(#data); # execute query with lines as argument
my $result = $sth->fetchall_arrayref;
my $randid = rand #$result; # find random index
my $row = $result->[ $randid ];
print "- #$row\n";
print "<BR />";
As you see, I've used placeholders, which is the proper way to use variables with queries. It also happens to be a simple way to handle an arbitrary amount of arguments. Because we include all lines in the query, we do not need a for loop.
As you see, I've also changed a few other small details, such as using three argument open with a lexical file handle, including the error variable $! in the die statement, using proper indentation, using strict and warnings (you should never code without them)
I've handled the randomization in perl because it is simplest for me. It may be as simple and more effective to handle in the SQL query. You may just tack on the ORDER BY random() LIMIT 1 to the end of it, and that might work just fine too.

Answer 2

Perhaps order by random(),
SELECT nazov_receptu FROM recepty ORDER BY RANDOM() LIMIT 1;
If you want to fetch only one random row, make sure to put this code out of the loop,
my $sth = $dbh->prepare("SELECT nazov_receptu FROM recepty ORDER BY RANDOM() LIMIT 1");
$sth->execute;
my ($nazov_receptu) = $sth->fetchrow_array;

Answer 3

Because your query is inside the foreach my $line (#data) loop, it will run once for each item in #data, getting a different random row each time. If you want it to only run once total, you need to move it outside of that loop (in addition to using "order by random() limit 1").