Lucene 0.13

Lucene is a Perl API to the C++ port of the Lucene search engine.

SYNOPSIS

Initialize/Empty Lucene index

my $analyzer = new Lucene::Analysis::Standard::StandardAnalyzer();
my $store = Lucene::Store::FSDirectory->getDirectory("/home/lucene", 1);

my $tmp_writer = new Lucene::Index::IndexWriter($store, $analyzer, 1);
$tmp_writer->close;
undef $tmp_writer;

Choose your Analyzer (string tokenizer)

# lowercases text and splits it at non-letter characters
my $analyzer = new Lucene::Analysis::SimpleAnalyzer();
# same as before and removes stop words
my $analyzer = new Lucene::Analysis::StopAnalyzer();
# same as before but you provide your own stop words
my $analyzer = new Lucene::Analysis::StopAnalyzer([qw/that this in or and/]);
# splits text at whitespace characters
my $analyzer = new Lucene::Analysis::WhitespaceAnalyzer();
# lowercases text, tokenized it based on a grammer that
# leaves named authorities intact (e-mails, company names,
# web hostnames, IP addresses, etc) and removed stop words
my $analyzer = new Lucene::Analysis::Standard::StandardAnalyzer();
# same as before but you provide your own stop words
my $analyzer = new Lucene::Analysis::Standard::StandardAnalyzer([qw/that this in or and/]);
# takes string as it is (only when using clucene-0.9.17 or above)
my $analyzer = new Lucene::Analysis::KeywordAnalyzer();

Create a custom Analyzer

package MyAnalyzer;

use base Lucene::Analysis::Analyzer;

# You MUST called SUPER::new if you implement new()
sub new {
my $class = shift;
my $self = $class->SUPER::new();
# ...
return $self;
}

sub tokenStream {
my ($self, $field, $reader) = @_;
my $ret = new Lucene::Analysis::StandardTokenizer($reader);
if ($field eq "MyKeywordField") {
return $ret;
}
$ret = new Lucene::Analysis::LowerCaseFilter($ret);
$ret = new Lucene::Analysis::StopFilter($ret, [qw/foo bar bax/]);
return $ret;
}
package main;
my $analyzer = new MyAnalyzer;

Choose your Store (storage engine)

# in-memory storage
my $store = new Lucene::Store::RAMDirectory();
# disk-based storage
my $store = Lucene::Store::FSDirectory->getDirectory("/home/lucene", 0);

Open and configure an IndexWriter

my $writer = new Lucene::Index::IndexWriter($store, $analyzer, 0);
# optional settings for power users
$writer->setMergeFactor(100);
$writer->setUseCompoundFile(0);
$writer->setMaxFieldLength(255);
$writer->setMinMergeDocs(10);
$writer->setMaxMergeDocs(100);

Create Documents and add Fields

my $doc = new Lucene::Document;
# field gets analyzed, indexed and stored
$doc->add(Lucene::Document::Field->Text("content", $content));
# field gets indexed and stored
$doc->add(Lucene::Document::Field->Keyword("isbn", $isbn));
# field gets just stored
$doc->add(Lucene::Document::Field->UnIndexed("sales_rank", $sales_rank));
# field gets analyzed and indexed
$doc->add(Lucene::Document::Field->UnStored("categories", $categories));