--- trunk/lib/WebPAC/Output/Estraier.pm	2005/06/25 20:23:23	1
+++ trunk/lib/WebPAC/Output/Estraier.pm	2005/11/23 00:14:05	113
@@ -3,9 +3,15 @@
 use warnings;
 use strict;
 
+use base qw/WebPAC::Common/;
+
+use HyperEstraier;
+use Text::Iconv;
+use Data::Dumper;
+
 =head1 NAME
 
-WebPAC::Output::Estraier - The great new WebPAC::Output::Estraier!
+WebPAC::Output::Estraier - Create Hyper Estraier full text index
 
 =head1 VERSION
 
@@ -17,49 +23,165 @@
 
 =head1 SYNOPSIS
 
-Quick summary of what the module does.
+Create full text index using Hyper Estraier index from data with
+type C<search>.
 
-Perhaps a little code snippet.
+=head1 FUNCTIONS
 
-    use WebPAC::Output::Estraier;
+=head2 new
 
-    my $foo = WebPAC::Output::Estraier->new();
-    ...
+Connect to Hyper Estraier index using HTTP
 
-=head1 EXPORT
+ my $est = new WebPAC::Output::Estraier(
+ 	url => 'http://localhost:1978/node/webpac2',
+	user => 'admin',
+	passwd => 'admin',
+	database => 'demo',
+	encoding => 'iso-8859-2',
+ );
 
-A list of functions that can be exported.  You can delete this section
-if you don't export anything, such as for a purely object-oriented module.
+Options are:
 
-=head1 FUNCTIONS
+=over 4
+
+=item url
+
+URI to C<estmaster> node
+
+=item user
+
+C<estmaster> user with sufficient rights
+
+=item passwd
+
+password for user
+
+=item database
+
+name of database from which data comes
+
+=item encoding
 
-=head2 function1
+character encoding of C<data_structure> if it's differenet than C<ISO-8859-2>
+(and it probably is). This encoding will be converted to C<UTF-8> for
+Hyper Estraier.
+
+=back
+
+Name of database will be used to form URI of documents in index.
 
 =cut
 
-sub function1 {
+sub new {
+	my $class = shift;
+        my $self = {@_};
+        bless($self, $class);
+
+	my $log = $self->_get_logger;
+
+	foreach my $p (qw/url user passwd/) {
+		$log->logdie("need $p") unless ($self->{$p});
+	}
+
+	$log->info("opening Hyper Estraier index $self->{'url'}");
+
+	$self->{'db'} = HyperEstraier::Node->new($self->{'url'});
+	$self->{'db'}->set_auth($self->{'user'}, $self->{'passwd'});
+
+	my $encoding = $self->{'encoding'} || 'ISO-8859-2';
+	$log->info("using encoding $encoding");
+
+	$self->{'iconv'} = new Text::Iconv($encoding, 'UTF-8') or
+		$log->logdie("can't create conversion from $encoding to UTF-8");
+
+	$self ? return $self : return undef;
 }
 
-=head2 function2
+
+=head2 add
+
+Adds one entry to database.
+
+  $est->add(
+  	id => 42,
+	ds => $ds,
+	type => 'display',
+	url_prefix => 'database name',
+	text => 'optional text from which snippet is created',
+  );
+
+This function will create  entries in index using following URI format:
+
+  C<file:///database%20name/000>
+
+Each tag in C<data_structure> with specified C<type> will create one
+attribute and corresponding hidden text (used for search).
 
 =cut
 
-sub function2 {
-}
+sub add {
+	my $self = shift;
 
-=head1 AUTHOR
+        my $args = {@_};
 
-Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >>
+	my $log = $self->_get_logger;
+
+	my $database = $self->{'database'} || $log->logconfess('no database in $self');
+	$log->logconfess('need db in object') unless ($self->{'db'});
+
+	foreach my $p (qw/id ds type/) {
+		$log->logdie("need $p") unless ($args->{$p});
+	}
+
+	my $type = $args->{'type'};
+	my $mfn = $args->{'id'};
+
+	my $uri = "file:///$type/$database/$mfn";
+	$log->debug("creating $uri");
+
+	my $doc = HyperEstraier::Document->new;
+	$doc->add_attr('@uri', $self->{'iconv'}->convert($uri) );
+
+	$log->debug("ds = ", sub { Dumper($args->{'ds'}) } );
+
+	# filter all tags which have type defined
+	my @tags = grep {
+		ref($args->{'ds'}->{$_}) eq 'HASH' && defined( $args->{'ds'}->{$_}->{$type} )
+	} keys %{ $args->{'ds'} };
 
-=head1 BUGS
+	$log->debug("tags = ", join(",", @tags));
 
-Please report any bugs or feature requests to
-C<bug-webpac-output-estraier@rt.cpan.org>, or through the web interface at
-L<http://rt.cpan.org/NoAuth/ReportBug.html?Queue=WebPAC>.
-I will be notified, and then you'll automatically be notified of progress on
-your bug as I make changes.
+	return unless (@tags);
 
-=head1 ACKNOWLEDGEMENTS
+	foreach my $tag (@tags) {
+
+		my $vals = join(" ", @{ $args->{'ds'}->{$tag}->{$type} });
+
+		$log->logconfess("no values for $tag/$type") unless ($vals);
+
+		$vals = $self->{'iconv'}->convert( $vals ) or
+			$log->logdie("can't convert '$vals' to UTF-8");
+
+		$doc->add_attr( $tag, $vals );
+		$doc->add_hidden_text( $vals );
+	}
+
+	my $text = $args->{'text'};
+	if ( $text ) {
+		$text = $self->{'iconv'}->convert( $text ) or
+			$log->logdie("can't convert '$text' to UTF-8");
+		$doc->add_text( $text );
+	}
+
+	$log->debug("adding ", sub { $doc->dump_draft } );
+	$self->{'db'}->put_doc($doc) || $log->logdie("can't add document $uri to index");
+
+	return 1;
+}
+
+=head1 AUTHOR
+
+Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >>
 
 =head1 COPYRIGHT & LICENSE