X-Git-Url: http://git.vanrenterghem.biz/git.ikiwiki.info.git/blobdiff_plain/65d9871a380be4b6622e64de62b7ba4f5f78fb94..4691a2ad39cce273231fddd9a589b4f8fdc1b063:/IkiWiki/Plugin/search.pm?ds=sidebyside

diff --git a/IkiWiki/Plugin/search.pm b/IkiWiki/Plugin/search.pm
index 75004b201..eedfa6924 100644
--- a/IkiWiki/Plugin/search.pm
+++ b/IkiWiki/Plugin/search.pm
@@ -1,30 +1,29 @@
 #!/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 { #{{{
-	hook(type => "checkconfig", id => "hyperestraier",
-		call => \&checkconfig);
-	hook(type => "pagetemplate", id => "hyperestraier",
-		call => \&pagetemplate);
-	hook(type => "delete", id => "hyperestraier",
-		call => \&delete);
-	hook(type => "change", id => "hyperestraier",
-		call => \&change);
-	hook(type => "cgi", id => "hyperestraier",
-		call => \&cgi);
+	hook(type => "checkconfig", id => "search", call => \&checkconfig);
+	hook(type => "pagetemplate", id => "search", call => \&pagetemplate);
+	hook(type => "postscan", id => "search", call => \&index);
+	hook(type => "delete", id => "search", call => \&delete);
+	hook(type => "cgi", id => "search", call => \&cgi);
 } # }}}
 
 sub checkconfig () { #{{{
 	foreach my $required (qw(url cgiurl)) {
 		if (! length $config{$required}) {
-			error("Must specify $required when using the search plugin\n");
+			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";
+	}
 } #}}}
 
 my $form;
@@ -45,83 +44,165 @@ sub pagetemplate (@) { #{{{
 	}
 } #}}}
 
-sub delete (@) { #{{{
-	debug("cleaning hyperestraier search index");
-	estcmd("purge -cl");
-	estcfg();
-} #}}}
+my $scrubber;
+my $stemmer;
+sub index (@) { #{{{
+	my %params=@_;
+
+	setupfiles();
+
+	# A unique pageterm is used to identify the document for a page.
+	my $pageterm=pageterm($params{page});
+	return $params{content} unless defined $pageterm;
+	
+	my $db=xapiandb();
+	my $doc=Search::Xapian::Document->new();
+	my $caption=IkiWiki::pagetitle($params{page});
+	my $title;
+	if (exists $pagestate{$params{page}}{meta} &&
+		exists $pagestate{$params{page}}{meta}{title}) {
+		$title=$pagestate{$params{page}}{meta}{title};
+	}
+	else {
+		$title=$caption;
+	}
 
-sub change (@) { #{{{
-	debug("updating hyperestraier search index");
-	estcmd("gather -cm -bc -cl -sd",
-		map {
-			Encode::encode_utf8($config{destdir}."/".$renderedfiles{pagename($_)})
-		} @_
+	# Remove html from text to be indexed.
+	if (! defined $scrubber) {
+		eval q{use HTML::Scrubber};
+		if (! $@) {
+			$scrubber=HTML::Scrubber->new(allow => []);
+		}
+	}
+	my $toindex = defined $scrubber ? $scrubber->scrub($params{content}) : $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, $size);
+	if (length($sample) == $size) {
+		my $max=length($toindex);
+		my $next;
+		while ($size < $max &&
+		       ($next=substr($toindex, $size++, 1)) !~ /\s/) {
+			$sample.=$next;
+		}
+	}
+	$sample=~s/\n/ /g;
+	
+	# data used by omega
+	# Decode html entities in it, since omega re-encodes them.
+	eval q{use HTML::Entities};
+	$doc->set_data(
+		"url=".urlto($params{page}, "")."\n".
+		"sample=".decode_entities($sample)."\n".
+		"caption=".decode_entities($caption)."\n".
+		"modtime=$IkiWiki::pagemtime{$params{page}}\n".
+		"size=".length($params{content})."\n"
 	);
-	estcfg();
+
+	# Index document and add terms for other metadata.
+	my $tg = Search::Xapian::TermGenerator->new();
+	if (! $stemmer) {
+		my $langcode=$ENV{LANG} || "en";
+		$langcode=~s/_.*//;
+
+		# This whitelist is here to work around a xapian bug (#486138)
+		my @whitelist=qw{da de en es fi fr hu it no pt ru ro sv tr};
+
+		if (grep { $_ eq $langcode } @whitelist) {
+			$stemmer=Search::Xapian::Stem->new($langcode);
+		}
+		else {
+			$stemmer=Search::Xapian::Stem->new("english");
+		}
+	}
+	$tg->set_stemmer($stemmer);
+	$tg->set_document($doc);
+	$tg->index_text($params{page}, 2);
+	$tg->index_text($caption, 2);
+	$tg->index_text($title, 2) if $title ne $caption;
+	$tg->index_text($toindex);
+	$tg->index_text(lc($title), 1, "S"); # for title:foo
+	foreach my $link (@{$links{$params{page}}}) {
+		$tg->index_text(lc($link), 1, "XLINK"); # for link:bar
+	}
+
+	$doc->add_term($pageterm);
+	$db->replace_document_by_term($pageterm, $doc);
+} #}}}
+
+sub delete (@) { #{{{
+	my $db=xapiandb();
+	foreach my $page (@_) {
+		my $pageterm=pageterm(pagename($page));
+		$db->delete_document_by_term($pageterm) if defined $pageterm;
+	}
 } #}}}
 
 sub cgi ($) { #{{{
 	my $cgi=shift;
 
-	if (defined $cgi->param('phrase')) {
+	if (defined $cgi->param('P')) {
 		# only works for GET requests
-		chdir("$config{wikistatedir}/hyperestraier") || error("chdir: $!");
-		exec("./".IkiWiki::basename($config{cgiurl})) || error("estseek.cgi failed");
+		chdir("$config{wikistatedir}/xapian") || error("chdir: $!");
+		$ENV{OMEGA_CONFIG_FILE}="./omega.conf";
+		$ENV{CGIURL}=$config{cgiurl},
+		IkiWiki::loadindex();
+		$ENV{HELPLINK}=htmllink("", "", "ikiwiki/searching",
+			noimageinline => 1, linktext => "Help");
+		exec($config{omega_cgi}) || error("$config{omega_cgi} failed: $!");
 	}
 } #}}}
 
-my $configured=0;
-sub estcfg () { #{{{
-	return if $configured;
-	$configured=1;
-	
-	my $estdir="$config{wikistatedir}/hyperestraier";
-	my $cgi=IkiWiki::basename($config{cgiurl});
-	$cgi=~s/\..*$//;
-	open(TEMPLATE, ">$estdir/$cgi.tmpl") ||
-		error("write $estdir/$cgi.tmpl: $!");
-	print TEMPLATE IkiWiki::misctemplate("search", 
-		"<!--ESTFORM-->\n\n<!--ESTRESULT-->\n\n<!--ESTINFO-->\n\n");
-	close TEMPLATE;
-	open(TEMPLATE, ">$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/".IkiWiki::basename($config{cgiurl});
-	unlink($cgi);
-	symlink("/usr/lib/estraier/estseek.cgi", $cgi) ||
-		error("symlink $cgi: $!");
-} # }}}
+sub pageterm ($) { #{{{
+	my $page=shift;
 
-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";
+	# 240 is the number used by omindex to decide when to hash an
+	# overlong term. This does not use a compatible hash method though.
+	if (length $page > 240) {
+		eval q{use Digest::SHA1};
+		if ($@) {
+			debug("search: ".sprintf(gettext("need Digest::SHA1 to index %s"), $page)) if $@;
+			return undef;
 		}
-		close(CHILD) || error("estcmd @params exited nonzero: $?");
+
+		# Note no colon, therefore it's guaranteed to not overlap
+		# with a page with the same name as the hash..
+		return "U".lc(Digest::SHA1::sha1_hex($page));
 	}
 	else {
-		# child
-		open(STDOUT, "/dev/null"); # shut it up (closing won't work)
-		exec("estcmd", @params) || error("can't run estcmd");
+		return "U:".$page;
+	}
+} #}}}
+
+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;
+} #}}}
+
+{
+my $setup=0;
+sub setupfiles () { #{{{
+	if (! $setup and (! -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"))));
+		$setup=1;
 	}
 } #}}}
+}
 
 1