} #}}}
my $scrubber;
+my $stemmer;
sub index (@) { #{{{
my %params=@_;
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});
- }
+ my $title=IkiWiki::pagetitle($params{page});
# Remove html from text to be indexed.
if (! defined $scrubber) {
}
}
$sample=~s/\n/ /g;
- # Decode html entities in the sample since omega encodes them.
- eval q{use HTML::Entities};
- $sample=decode_entities($sample);
# 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=$sample\n".
- "caption=$title\n".
+ "sample=".decode_entities($sample)."\n".
+ "caption=".decode_entities($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"));
+ if (! $stemmer) {
+ my $langcode=$ENV{LANG} || "en";
+ $langcode=~s/_.*//;
+ eval { $stemmer=Search::Xapian::Stem->new($langcode) };
+ if ($@) {
+ $stemmer=Search::Xapian::Stem->new("english");
+ }
+ }
+ $tg->set_stemmer($stemmer);
$tg->set_document($doc);
$tg->index_text($params{page}, 2);
$tg->index_text($title, 2);