On http://www.mnogosearch.org/download.html near the bottom there is something called udm-perl which appears to be a front end.

You might also try searching on the old name, udm search.

I've recently encountered this problem and after a frustrating week of trying to find the 'lazy' way of integrating mnogosearch I think I've found a way. I used the mnogo search template to construct a JSON response, rather than have it spit out content. This has two benefits, now it can be called in my perl code eg.

my $req = HTTP::Request->new(GET => 'http://localhost/cgi-bin/search.c
+gi?q=a query' );

my $res = $ua->request($req);
my $json_results;
if ($res->is_success) {
    $json_results = $json->decode($res->content);
}
else {
    die $res->status_line, "\n";
}
[download]

but it can also be called from other places! It's essentially a web service. Cheers, Adam T.

Hi Adam, I am curently trying to do something similar. Is it possible for you to send me the json template you created? Many thanks and best regards, Michael

Thanks for the response.

What you found is a Perl front end, but was written for the 3.1.x version of mnoGoSearch (udmSearch). It performs SQL selects on mnoGo's database to obtain search results. The PHP front end, as well as the older Perl front end that I referenced, used mnoGo's libraries to obtain search results. I believe that to create a Perl module to function like this would require the use of h2xs and some knowledge of C.

Maybe I need to learn h2xs and develop a new Perl module if there isn't one available. The other option is to update the front end that you pointed me to by modifying the SQL to use the new mnoGo table structure.