Data::SearchEngine - A role for search engine abstraction.


version 0.33


  package Data::SearchEngine::MySearch;
  use Moose;

  with 'Data::SearchEngine';

  sub search {
    my ($self, $query) = @_;

    # ... your internal search junk

    my $result = Data::SearchEngine::Results->new(
        query => $query,
        pager => # ... make a Data::Page

    my @hits; # Populate with hits somehow

    foreach my $hit (@hits) {
            values => {
                name => $hit->name,
                description => $hit->description
            score => $hit->score

    return $result;


There are lots of search engine libraries. Each has a different interface. The goal of Data::SearchEngine is to provide a simple, extensive set of classes and roles that you can use to wrap a search implementation. The net result will be an easily swappable backend with a common set of features.


NOTE: You should avoid adding new attributes or subclassing Data::SearchEngine classes unless otherwise noted. Doing so will break compatability with future releases and obviate the whole reason that Data::SearchEngine exists. The only exception is the Results class (step 3 below) which should only be a subclass with roles applied, no new attributes or methods.

Step 1 - Wrap a search implementation

As shown in the SYNOPSIS, use the Data::SearchEngine role in a class that wraps your search implementation. Implement a search method that takes a Data::SearchEngine::Query object and returns a Data::SearchEngine::Results object.

Step 2 - Use Other Roles

If your library includes functionality other than searching, such as indexing new documents or removing them, you may include the Data::SearchEngine::Modifiable role. If you have other suggestions for roles please drop me a line!

Step 3 - Extend the Results

The results object may not have quite enough pieces for your implementation. If not, you can extend Data::SearchEngine::Results and add some other roles:


For results that contain faceted data.


For results that contain spellcheck data.


Data::SearchEngine provides a Digestable trait that can be applied to attributes of Query. Attributes with this trait will be added to a base64 MD5 digest to produce a unique key identifying this query. You can then serialize the Result using MooseX::Storage and store it under the digest of the Query for caching.



An attribute that signals the backend should operate in a debugging mode. Please see the implementation module for specifics on how to use this.


The defaults attribute is a simple HashRef that backends may use to get default settings from the user. The implementation of search may then use these defaults when setting up instances of a search.



Method for determining if the debug attribute has been set.

get_default ($key)

Returns the value from defaults (if any) for the specified key.

set_default ($key, $value)

Sets the value in defaults.


Cory G Watson <>


This software is copyright (c) 2012 by Cold Hard Code, LLC.

This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself.