Author image Marvin Humphrey
and 1 contributors

NAME

KinoSearch - Search engine library.

VERSION

0.20_02

EXTRA WARNING

This is a developer's release. The new features and API changes are being auditioned and may change.

WARNING

KinoSearch 0.20 BREAKS BACKWARDS COMPATIBILITY with earlier versions. Both the API and the file format have changed. Old applications must be tweaked, and old indexes cannot be read and must be recreated -- see the Changes file for details.

KinoSearch is still officially "alpha" software -- see Backwards Compatibility Policy, below.

SYNOPSIS

First, plan out your index structure and describe it with a "schema".

    # ./MySchema.pm

    package MySchema;
    use base qw( KinoSearch::Schema );
    use KinoSearch::Analysis::PolyAnalyzer;
    
    our %FIELDS = (
        title   => 'KinoSearch::Schema::FieldSpec',
        content => 'KinoSearch::Schema::FieldSpec',
    );

    sub analyzer { 
        return KinoSearch::Analysis::PolyAnalyzer->new( language => 'en' );
    }

Next, create the index and add documents to it.

    use KinoSearch::InvIndexer;
    use MySchema;
    
    my $invindexer = KinoSearch::InvIndexer->new(
        invindex => MySchema->clobber('/path/to/invindex'),
    );
    
    while ( my ( $title, $content ) = each %source_docs ) {
        $invindexer->add_doc({
            title   => $title,
            content => $content,
        });
    }
    
    $invindexer->finish;

Finally, search the index:

    use KinoSearch::Searcher;
    
    my $searcher = KinoSearch::Searcher->new(
        invindex => MySchema->open('/path/to/invindex'),
    );
    
    my $hits = $searcher->search( query => "foo bar" );
    while ( my $hit = $hits->fetch_hit_hashref ) {
        print "$hit->{title}\n";
    }

DESCRIPTION

KinoSearch is a loose port of the Java search engine library Apache Lucene, written in C and Perl. The archetypal application is website search, but it can be put to many different uses.

Features

  • Extremely fast. A single machine can handle millions of documents.

  • Scalable to multiple machines.

  • Incremental indexing (addition/deletion of documents to/from an existing index).

  • UTF-8 support.

  • Support for boolean operators AND, OR, and AND NOT; parenthetical groupings, and prepended +plus and -minus

  • Algorithmic selection of relevant excerpts and highlighting of search terms within excerpts.

  • Highly customizable query and indexing APIs.

  • Customizable Sorting.

  • Phrase matching.

  • Stemming.

  • Stoplists.

Getting Started

KinoSearch::Simple provides a stripped down API which may suffice for many tasks.

KinoSearch::Docs::Tutorial demonstrates how to build a basic CGI search application. Most people cut-and-paste the sample code from it and get right down to business, referring back to the class documentation only as needed.

The tutorial spends most of its time on these five classes:

Supported Languages and Encodings

As of version 0.20, KinoSearch supports Unicode in addition to Latin-1. All output strings use Perl's internal Unicode encoding. For use of KinoSearch with non-Latin-1 material, see Encode.

KinoSearch provides "native support" for 12 languages, meaning that a stemmer and a stoplist are available, and PolyAnalyzer supports them.

  • Danish

  • Dutch

  • English

  • Finnish

  • French

  • German

  • Italian

  • Norwegian

  • Portuguese

  • Russian

  • Spanish

  • Swedish

KinoSearch can also be extended to support other languages if you write your own subclass of KinoSearch::Analysis::Analyzer.

Delving Deeper

For creating complex queries, see KinoSearch::Search::Query and its subclasses BooleanQuery, TermQuery, and PhraseQuery, plus KinoSearch::QueryParser::QueryParser and KinoSearch::Search::QueryFilter.

If PolyAnalyzer doesn't meet your needs, see the base class KinoSearch::Analysis::Analyzer for how to write and integrate your own Analyzer subclass.

For distributed searching, see KinoSearch::Search::SearchServer, KinoSearch::Search::SearchClient, and KinoSearch::Search::MultiSearcher.

If you'd like a peek under the hood, see KinoSearch::Docs::FileFormat for an overview of the invindex file format, and KinoSearch::Docs::DevGuide for hacking/debugging tips.

Backwards Compatibility Policy

Until version 1.0 is released, KinoSearch's API and file format are subject to change without relation to the version number. Such changes are not undertaken lightly and hopefully none will be needed after the disruptions of 0.20.

Starting with 1.0, the following policy will be put in place:

    Search is a rapidly advancing field.  To stay current, KinoSearch
    has adopted a policy of "continuity" rather than backwards
    compatibility in perpetuity:

    Starting with version 1.0, KinoSearch will support obsolete
    features and files for one "extra" major revision.  API features
    which are supported in 1.0 and deprecated in 1.x will be removed
    no sooner than 3.0.  Indexes which are modified at least once
    using 2.x will be readable at least until 4.0.

    Rapid-fire incrementing of major version numbers is not
    anticipated.  With luck, someone might even solve Perl5/CPAN's
    versioning problem before the release of KinoSearch 2.0.

SEE ALSO

The KinoSearch homepage, where you'll find links to the mailing list and so on, is http://www.rectangular.com/kinosearch.

The Lucene homepage is http://lucene.apache.org.

History

Search::Kinosearch 0.02x, now dead and removed from CPAN, was this suite's forerunner. Plucene is a pure-Perl port of Lucene 1.3. KinoSearch is a from-scratch project which attempts to draws on the lessons of both.

KinoSearch is named for Kino, the main character in John Steinbeck's novella, "The Pearl".

SUPPORT

Please direct support questions to the kinosearch mailing list: subscription information at http://www.rectangular.com/kinosearch.

AUTHORS

Marvin Humphrey, <marvin at rectangular dot com>

Apache Lucene by Doug Cutting et al.

BUGS

Not thread-safe.

Some exceptions leak memory.

Won't work on esoteric architectures where a char is more than one byte, or where floats don't conform to IEEE 754.

Please report any other bugs or feature requests to bug-kinosearch@rt.cpan.org, or through the web interface at http://rt.cpan.org/NoAuth/ReportBug.html?Queue=KinoSearch.

COPYRIGHT & LICENSE

Copyright 2005-2007 Marvin Humphrey

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

Terms of usage for Apache Lucene, from which portions of KinoSearch are derived, are spelled out in the Apache License: see the file "ApacheLicense2.0.txt".

DISCLAIMER OF WARRANTY

BECAUSE THIS SOFTWARE IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE SOFTWARE, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE SOFTWARE "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE SOFTWARE IS WITH YOU. SHOULD THE SOFTWARE PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR, OR CORRECTION.

IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE SOFTWARE AS PERMITTED BY THE ABOVE LICENCE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE SOFTWARE (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE SOFTWARE TO OPERATE WITH ANY OTHER SOFTWARE), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.