X-Git-Url: https://sipb.mit.edu/gitweb.cgi/ikiwiki.git/blobdiff_plain/dea23a1031b55dbc408e9f99c761fd667331cccd..3ba723d79ad10420666cc7e9e1fc1537c59f6be5:/IkiWiki/Plugin/search.pm

diff --git a/IkiWiki/Plugin/search.pm b/IkiWiki/Plugin/search.pm
index c79d40469..dd44855bd 100644
--- a/IkiWiki/Plugin/search.pm
+++ b/IkiWiki/Plugin/search.pm
@@ -1,32 +1,44 @@
 #!/usr/bin/perl
-# hyperestraier search engine plugin
+# xapian-omega search engine plugin
 package IkiWiki::Plugin::search;
 
 use warnings;
 use strict;
-use IkiWiki;
+use IkiWiki 2.00;
 
 sub import { #{{{
-	IkiWiki::hook(type => "checkconfig", id => "hyperestraier",
-		call => \&checkconfig);
-	IkiWiki::hook(type => "pagetemplate", id => "hyperestraier",
-		call => \&pagetemplate);
-	IkiWiki::hook(type => "delete", id => "hyperestraier",
-		call => \&delete);
-	IkiWiki::hook(type => "change", id => "hyperestraier",
-		call => \&change);
-	IkiWiki::hook(type => "cgi", id => "hyperestraier",
-		call => \&cgi);
+	hook(type => "checkconfig", id => "search", call => \&checkconfig);
+	hook(type => "pagetemplate", id => "search", call => \&pagetemplate);
+	# run last so other needsbuild hooks can modify the list
+	hook(type => "needsbuild", id => "search", call => \&needsbuild,
+		last => 1);
+	hook(type => "filter", id => "search", call => \&filter);
+	hook(type => "delete", id => "search", call => \&delete);
+	hook(type => "cgi", id => "search", call => \&cgi);
 } # }}}
 
 sub checkconfig () { #{{{
 	foreach my $required (qw(url cgiurl)) {
-		if (! length $IkiWiki::config{$required}) {
-			IkiWiki::error("Must specify $required when using the search plugin\n");
+		if (! length $config{$required}) {
+			error(sprintf(gettext("Must specify %s when using the search plugin"), $required));
 		}
 	}
+
+	if (! exists $config{omega_cgi}) {
+		$config{omega_cgi}="/usr/lib/cgi-bin/omega/omega";
+	}
+	
+	if (! -e $config{wikistatedir}."/xapian" || $config{rebuild}) {
+		writefile("omega.conf", $config{wikistatedir}."/xapian",
+			"database_dir .\n".
+			"template_dir ./templates\n");
+		writefile("query", $config{wikistatedir}."/xapian/templates",
+			IkiWiki::misctemplate(gettext("search"),
+				readfile(IkiWiki::template_file("searchquery.tmpl"))));
+	}
 } #}}}
 
+my $form;
 sub pagetemplate (@) { #{{{
 	my %params=@_;
 	my $page=$params{page};
@@ -34,98 +46,122 @@ sub pagetemplate (@) { #{{{
 
 	# Add search box to page header.
 	if ($template->query(name => "searchform")) {
-		$template->param(searchform => qq{
-<form method="get" action="$IkiWiki::config{cgiurl}" id="searchform">
-<div>
-<input type="text" name="phrase" value="" size="16" />
-<input type="hidden" name="enc" value="UTF-8" />
-<input type="hidden" name="do" value="hyperestraier" />
-</div>
-</form>
-});
+		if (! defined $form) {
+			my $searchform = template("searchform.tmpl", blind_cache => 1);
+			$searchform->param(searchaction => $config{cgiurl});
+			$form=$searchform->output;
+		}
+
+		$template->param(searchform => $form);
 	}
 } #}}}
 
-sub delete (@) { #{{{
-	IkiWiki::debug("cleaning hyperestraier search index");
-	IkiWiki::estcmd("purge -cl");
-	IkiWiki::estcfg();
+my %toindex;
+sub needsbuild ($) { #{{{
+	%toindex = map { pagename($_) => 1 } @{shift()};
 } #}}}
 
-sub change (@) { #{{{
-	IkiWiki::debug("updating hyperestraier search index");
-	IkiWiki::estcmd("gather -cm -bc -cl -sd",
-		map {
-			$IkiWiki::config{destdir}."/".$IkiWiki::renderedfiles{IkiWiki::pagename($_)}
-		} @_
-	);
-	IkiWiki::estcfg();
+my $scrubber;
+sub filter (@) { #{{{
+	my %params=@_;
+	
+	if ($params{page} eq $params{destpage} && $toindex{$params{page}}) {
+		# index page
+		my $db=xapiandb();
+		my $doc=Search::Xapian::Document->new();
+		my $title;
+		if (exists $pagestate{$params{page}}{meta} &&
+		    exists $pagestate{$params{page}}{meta}{title}) {
+			$title=$pagestate{$params{page}}{meta}{title};
+		}
+		else {
+			$title=IkiWiki::pagetitle($params{page});
+		}
+
+		# Remove any html from text to be indexed.
+		# TODO: This removes html that is in eg, a markdown pre,
+		# which should not be removed.
+		if (! defined $scrubber) {
+			eval q{use HTML::Scrubber};
+			error($@) if $@;
+			$scrubber=HTML::Scrubber->new(allow => []);
+		}
+		my $toindex=$scrubber->scrub($params{content});
+
+		# Take 512 characters for a sample, then extend it out
+		# if it stopped in the middle of a word.
+		my $size=512;
+		my ($sample)=substr($toindex, 0, $size);
+		my $next=substr($toindex, $size++, 1);
+		while (defined $next && $next !~ /\s/) {
+			$sample.=$next;
+			$next=substr($toindex, $size++, 1);
+		}
+		$sample=~s/\n/ /g;
+
+		# data used by omega
+		$doc->set_data(
+			"url=".urlto($params{page}, "")."\n".
+			"sample=$sample\n".
+			"caption=$title\n".
+			"modtime=$IkiWiki::pagemtime{$params{page}}\n".
+			"size=".length($params{content})."\n"
+		);
+
+		my $tg = Search::Xapian::TermGenerator->new();
+		$tg->set_stemmer(new Search::Xapian::Stem("english"));
+		$tg->set_document($doc);
+		$tg->index_text($params{page}, 2);
+		$tg->index_text($title, 2);
+		$tg->index_text($toindex);
+
+		my $pageterm=pageterm($params{page});
+		$doc->add_term($pageterm);
+		$db->replace_document_by_term($pageterm, $doc);
+	}
+
+	return $params{content};
+} #}}}
+
+sub delete (@) { #{{{
+	my $db=xapiandb();
+	foreach my $page (@_) {
+		$db->delete_document_by_term(pageterm($page));
+	}
 } #}}}
 
 sub cgi ($) { #{{{
 	my $cgi=shift;
 
-	if (defined $cgi->param('phrase')) {
+	if (defined $cgi->param('P')) {
 		# only works for GET requests
-		chdir("$IkiWiki::config{wikistatedir}/hyperestraier") || IkiWiki::error("chdir: $!");
-		exec("./".IkiWiki::basename($IkiWiki::config{cgiurl})) || IkiWiki::error("estseek.cgi failed");
+		chdir("$config{wikistatedir}/xapian") || error("chdir: $!");
+		$ENV{OMEGA_CONFIG_FILE}="./omega.conf";
+		$ENV{CGIURL}=$config{cgiurl},
+		exec($config{omega_cgi}) || error("$config{omega_cgi} failed: $!");
 	}
 } #}}}
 
-# Easier to keep these in the IkiWiki namespace.
-package IkiWiki;
+sub pageterm ($) { #{{{
+	my $page=shift;
 
-my $configured=0;
-sub estcfg () { #{{{
-	return if $configured;
-	$configured=1;
-	
-	my $estdir="$config{wikistatedir}/hyperestraier";
-	my $cgi=basename($config{cgiurl});
-	$cgi=~s/\..*$//;
-	open(TEMPLATE, ">$estdir/$cgi.tmpl") ||
-		error("write $estdir/$cgi.tmpl: $!");
-	print TEMPLATE misctemplate("search", 
-		"<!--ESTFORM-->\n\n<!--ESTRESULT-->\n\n<!--ESTINFO-->\n\n");
-	close TEMPLATE;
-	open(TEMPLATE, ">$estdir/$cgi.conf") ||
-		error("write $estdir/$cgi.conf: $!");
-	my $template=template("estseek.conf");
-	eval q{use Cwd 'abs_path'};
-	$template->param(
-		index => $estdir,
-		tmplfile => "$estdir/$cgi.tmpl",
-		destdir => abs_path($config{destdir}),
-		url => $config{url},
-	);
-	print TEMPLATE $template->output;
-	close TEMPLATE;
-	$cgi="$estdir/".basename($config{cgiurl});
-	unlink($cgi);
-	symlink("/usr/lib/estraier/estseek.cgi", $cgi) ||
-		error("symlink $cgi: $!");
-} # }}}
+	# TODO: check if > 255 char page names overflow term
+	# length; use sha1 if so?
+	return "P".$page;
+} #}}}
 
-sub estcmd ($;@) { #{{{
-	my @params=split(' ', shift);
-	push @params, "-cl", "$config{wikistatedir}/hyperestraier";
-	if (@_) {
-		push @params, "-";
-	}
-	
-	my $pid=open(CHILD, "|-");
-	if ($pid) {
-		# parent
-		foreach (@_) {
-			print CHILD "$_\n";
-		}
-		close(CHILD) || error("estcmd @params exited nonzero: $?");
-	}
-	else {
-		# child
-		open(STDOUT, "/dev/null"); # shut it up (closing won't work)
-		exec("estcmd", @params) || error("can't run estcmd");
+my $db;
+sub xapiandb () { #{{{
+	if (! defined $db) {
+		eval q{
+			use Search::Xapian;
+			use Search::Xapian::WritableDatabase;
+		};
+		error($@) if $@;
+		$db=Search::Xapian::WritableDatabase->new($config{wikistatedir}."/xapian/default",
+			Search::Xapian::DB_CREATE_OR_OPEN());
 	}
+	return $db;
 } #}}}
 
 1