kinosearch1::searcher(3pm) [debian man page]
KinoSearch1::Searcher(3pm) User Contributed Perl Documentation KinoSearch1::Searcher(3pm) NAME
KinoSearch1::Searcher - execute searches SYNOPSIS
my $analyzer = KinoSearch1::Analysis::PolyAnalyzer->new( language => 'en', ); my $searcher = KinoSearch1::Searcher->new( invindex => $invindex, analyzer => $analyzer, ); my $hits = $searcher->search( query => 'foo bar' ); DESCRIPTION
Use the Searcher class to perform queries against an invindex. METHODS
new my $searcher = KinoSearch1::Searcher->new( invindex => $invindex, analyzer => $analyzer, ); Constructor. Takes two labeled parameters, both of which are required. o invindex - can be either a path to an invindex, or a KinoSearch1::Store::InvIndex object. o analyzer - An object which subclasses KinoSearch1::Analysis::Analyer, such as a PolyAnalyzer. This must be identical to the Analyzer used at index-time, or the results won't match up. search my $hits = $searcher->search( query => $query, # required filter => $filter, # default: undef (no filtering) ); Process a search and return a Hits object. search() expects labeled hash-style parameters. o query - Can be either an object which subclasses KinoSearch1::Search::Query, or a query string. If it's a query string, it will be parsed using a QueryParser and a search will be performed against all indexed fields in the invindex. For more sophisticated searching, supply Query objects, such as TermQuery and BooleanQuery. o filter - Must be a KinoSearch1::Search::QueryFilter. Search results will be limited to only those documents which pass through the filter. Caching a Searcher When a Searcher is created, a small portion of the invindex is loaded into memory. For large document collections, this startup time may become noticeable, in which case reusing the searcher is likely to speed up your search application. Caching a Searcher is especially helpful when running a high-activity app under mod_perl. Searcher objects always represent a snapshot of an invindex as it existed when the Searcher was created. If you want the search results to reflect modifications to an invindex, you must create a new Searcher after the update process completes. COPYRIGHT
Copyright 2005-2010 Marvin Humphrey LICENSE, DISCLAIMER, BUGS, etc. See KinoSearch1 version 1.00. perl v5.14.2 2011-11-15 KinoSearch1::Searcher(3pm)
Check Out this Related Man Page
KinoSearch1::Docs::Tutorial(3pm) User Contributed Perl Documentation KinoSearch1::Docs::Tutorial(3pm) NAME
KinoSearch1::Docs::Tutorial - sample indexing and search applications DESCRIPTION
The following sample code for invindexer.plx and search.cgi can be used to create a simple search engine. It requires the html presentation of the US Constitution included in the distribution for KinoSearch1, under "t/us_constitution". Note that a proper indexer for html documents would not rely on quick-n-dirty regular expressions for stripping tags, as this one does for the sake of brevity -- it would use a dedicated parsing module such as HTML::Parser. invindexer.plx #!/usr/bin/perl use strict; use warnings; use File::Spec; use KinoSearch1::InvIndexer; use KinoSearch1::Analysis::PolyAnalyzer; ### In order for invindexer.plx to work correctly, you must modify ### $source_dir, $path_to_invindex, and possibly $base_url. ### ### $source_dir must lead to the directory containing the US ### Constitution html files. ### ### $path_to_invindex is the future location of the invindex. ### ### $base_url should reflect the location of the us_constitution directory ### when accessed via a web browser. my $source_dir = ''; my $path_to_invindex = ''; my $base_url = '/us_constitution'; opendir( my $source_dh, $source_dir ) or die "Couldn't opendir '$source_dir': $!"; my @filenames = grep {/.html/} readdir $source_dh; closedir $source_dh or die "Couldn't closedir '$source_dir': $!"; ### STEP 1: Choose an Analyzer. my $analyzer = KinoSearch1::Analysis::PolyAnalyzer->new( language => 'en', ); ### STEP 2: Create a InvIndexer object. my $invindexer = KinoSearch1::InvIndexer->new( analyzer => $analyzer, invindex => $path_to_invindex, create => 1, ); ### STEP 3: Define fields. $invindexer->spec_field( name => 'title' ); $invindexer->spec_field( name => 'bodytext', vectorized => 1, ); $invindexer->spec_field( name => 'url', indexed => 0, ); foreach my $filename (@filenames) { next if $filename eq 'index.html'; my $filepath = File::Spec->catfile( $source_dir, $filename ); open( my $fh, '<', $filepath ) or die "couldn't open file '$filepath': $!"; my $content = do { local $/; <$fh> }; ### STEP 4: Start a new document. my $doc = $invindexer->new_doc; $content =~ m#<title>(.*?)</title>#s or die "couldn't isolate title in '$filepath'"; my $title = $1; $content =~ m#<div id="bodytext">(.*?)</div><!--bodytext-->#s or die "couldn't isolate bodytext in '$filepath'"; my $bodytext = $1; $bodytext =~ s/<.*?>/ /gsm; # quick and dirty tag stripping ### STEP 5: Set the value for each field. $doc->set_value( url => "$base_url/$filename" ); $doc->set_value( title => $title ); $doc->set_value( bodytext => $bodytext ); ### STEP 6 Add the document to the invindex. $invindexer->add_doc($doc); ### STEP 7 Repeat steps 3-5 for each document in the collection. } ### STEP 8 Finalize the invindex. $invindexer->finish; search.cgi #!/usr/bin/perl -T use strict; use warnings; use CGI; use List::Util qw( max min ); use POSIX qw( ceil ); use KinoSearch1::Searcher; use KinoSearch1::Analysis::PolyAnalyzer; use KinoSearch1::Highlight::Highlighter; my $cgi = CGI->new; my $q = $cgi->param('q'); my $offset = $cgi->param('offset'); my $hits_per_page = 10; $q = '' unless defined $q; $offset = 0 unless defined $offset; ### In order for search.cgi to work, $path_to_invindex must be modified so ### that it points to the invindex created by invindexer.plx, and ### $base_url may have to change to reflect where a web-browser should ### look for the us_constitution directory. my $path_to_invindex = ''; my $base_url = '/us_constitution'; ### STEP 1: Specify the same Analyzer used to create the invindex. my $analyzer = KinoSearch1::Analysis::PolyAnalyzer->new( language => 'en', ); ### STEP 2: Create a Searcher object. my $searcher = KinoSearch1::Searcher->new( invindex => $path_to_invindex, analyzer => $analyzer, ); ### STEP 3: Feed a query to the Search object. my $hits = $searcher->search($q); ### STEP 4: Arrange for highlighted excerpts to be created. my $highlighter = KinoSearch1::Highlight::Highlighter->new( excerpt_field => 'bodytext' ); $hits->create_excerpts( highlighter => $highlighter ); ### STEP 5: Process the search. $hits->seek( $offset, $hits_per_page ); ### STEP 6: Format the results however you like. # create result list my $report = ''; while ( my $hit = $hits->fetch_hit_hashref ) { my $score = sprintf( "%0.3f", $hit->{score} ); $report .= qq| <p> <a href="$hit->{url}"><strong>$hit->{title}</strong></a> <em>$score</em> <br> $hit->{excerpt} <br> <span class="excerptURL">$hit->{url}</span> </p> |; } $q = CGI::escapeHTML($q); # display info about the number of hits, paging links my $total_hits = $hits->total_hits; my $num_hits_info; if ( !length $q ) { # no query, no display $num_hits_info = ''; } elsif ( $total_hits == 0 ) { # alert the user that their search failed $num_hits_info = qq|<p>No matches for <strong>$q</strong></p>|; } else { # calculate the nums for the first and last hit to display my $last_result = min( ( $offset + $hits_per_page ), $total_hits ); my $first_result = min( ( $offset + 1 ), $last_result ); # display the result nums, start paging info $num_hits_info = qq| <p> Results <strong>$first_result-$last_result</strong> of <strong>$total_hits</strong> for <strong>$q</strong>. </p> <p> Results Page: |; # calculate first and last hits pages to display / link to my $current_page = int( $first_result / $hits_per_page ) + 1; my $last_page = ceil( $total_hits / $hits_per_page ); my $first_page = max( 1, ( $current_page - 9 ) ); $last_page = min( $last_page, ( $current_page + 10 ) ); # create a url for use in paging links my $href = $cgi->url( -relative => 1 ) . "?" . $cgi->query_string; $href .= ";offset=0" unless $href =~ /offset=/; # generate the "Prev" link; if ( $current_page > 1 ) { my $new_offset = ( $current_page - 2 ) * $hits_per_page; $href =~ s/(?<=offset=)d+/$new_offset/; $num_hits_info .= qq|<a href="$href"><= Prev</a> |; } # generate paging links for my $page_num ( $first_page .. $last_page ) { if ( $page_num == $current_page ) { $num_hits_info .= qq|$page_num |; } else { my $new_offset = ( $page_num - 1 ) * $hits_per_page; $href =~ s/(?<=offset=)d+/$new_offset/; $num_hits_info .= qq|<a href="$href">$page_num</a> |; } } # generate the "Next" link if ( $current_page != $last_page ) { my $new_offset = $current_page * $hits_per_page; $href =~ s/(?<=offset=)d+/$new_offset/; $num_hits_info .= qq|<a href="$href">Next =></a> |; } # finish paging links $num_hits_info .= "</p> "; } # blast it all out print "Content-type: text/html "; print <<END_HTML; <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd"> <html> <head> <meta http-equiv="Content-type" content="text/html;charset=ISO-8859-1"> <link rel="stylesheet" type="text/css" href="$base_url/uscon.css"> <title>KinoSearch: $q</title> </head> <body> <div id="navigation"> <form id="usconSearch" action=""> <strong> Search the <a href="$base_url/index.html">US Constitution</a>: </strong> <input type="text" name="q" id="q" value="$q"> <input type="submit" value="=>"> <input type="hidden" name="offset" value="0"> </form> </div><!--navigation--> <div id="bodytext"> $report $num_hits_info <p style="font-size: smaller; color: #666"> <em>Powered by <a href="http://www.rectangular.com/kinosearch/"> KinoSearch </a> </em> </p> </div><!--bodytext--> </body> </html> END_HTML COPYRIGHT
Copyright 2005-2010 Marvin Humphrey LICENSE, DISCLAIMER, BUGS, etc. See KinoSearch1 version 1.00. perl v5.14.2 2011-11-15 KinoSearch1::Docs::Tutorial(3pm)