1 package WebPAC::Output::Estraier;
6 use base qw/WebPAC::Common/;
16 WebPAC::Output::Estraier - Create Hyper Estraier full text index
24 our $VERSION = '0.09';
28 Create full text index using Hyper Estraier index from data with
35 Connect to Hyper Estraier index using HTTP
37 my $est = new WebPAC::Output::Estraier(
38 masterurl => 'http://localhost:1978/',
42 encoding => 'iso-8859-2',
52 URI to C<estmaster> node
56 C<estmaster> user with sufficient rights
64 name of database from which data comes
68 character encoding of C<data_structure> if it's differenet than C<ISO-8859-2>
69 (and it probably is). This encoding will be converted to C<UTF-8> for
74 Name of database will be used to form URI of documents in index.
83 my $log = $self->_get_logger;
85 #$log->debug("self: ", sub { Dumper($self) });
87 foreach my $p (qw/masterurl user passwd database/) {
88 $log->logdie("need $p") unless ($self->{$p});
91 my $url = $self->{masterurl} . '/node/' . $self->{database};
95 $log->debug("nodedel $self->{database}");
96 $self->master( action => 'nodedel', name => $self->{database} );
98 $log->debug("opening index $self->{url}");
101 my $nodes = $self->master( action => 'nodelist' );
103 $log->debug("nodes found: $nodes");
105 if ($nodes !~ m/^$self->{database}\t/sm) {
106 $log->warn("creating index $url");
109 name => $self->{database},
110 label => "WebPAC $self->{database}",
111 ) || $log->logdie("can't create Hyper Estraier node $self->{database}");
114 $self->{db} = Search::Estraier::Node->new( debug => 1 );
115 $self->{db}->set_url($self->{url});
116 $self->{db}->set_auth($self->{user}, $self->{passwd});
118 my $encoding = $self->{encoding} || 'ISO-8859-2';
119 $log->info("using index $self->{url} with encoding $encoding");
121 $self->{iconv} = new Text::Iconv($encoding, 'UTF-8') or
122 $log->logdie("can't create conversion from $encoding to UTF-8");
124 $self ? return $self : return undef;
130 Adds one entry to database.
136 text => 'optional text from which snippet is created',
139 This function will create entries in index using following URI format:
141 C<file:///type/database%20name/000>
143 Each tag in C<data_structure> with specified C<type> will create one
144 attribute and corresponding hidden text (used for search).
153 my $log = $self->_get_logger;
155 my $database = $self->{'database'} || $log->logconfess('no database in $self');
156 $log->logconfess('need db in object') unless ($self->{'db'});
158 foreach my $p (qw/id ds type/) {
159 $log->logdie("need $p") unless ($args->{$p});
162 my $type = $args->{'type'};
163 my $id = $args->{'id'};
165 my $uri = "file:///$type/$database/$id";
166 $log->debug("creating $uri");
168 my $doc = Search::Estraier::Document->new;
169 $doc->add_attr('@uri', $self->{'iconv'}->convert($uri) );
171 $log->debug("ds = ", sub { Dumper($args->{'ds'}) } );
173 # filter all tags which have type defined
175 ref($args->{'ds'}->{$_}) eq 'HASH' && defined( $args->{'ds'}->{$_}->{$type} )
176 } keys %{ $args->{'ds'} };
178 $log->debug("tags = ", join(",", @tags));
180 return unless (@tags);
182 foreach my $tag (@tags) {
184 my $vals = join(" ", @{ $args->{'ds'}->{$tag}->{$type} });
186 $log->logconfess("no values for $tag/$type") unless ($vals);
188 $vals = $self->{'iconv'}->convert( $vals ) or
189 $log->logdie("can't convert '$vals' to UTF-8");
191 $doc->add_attr( $tag, $vals );
192 $doc->add_hidden_text( $vals );
195 my $text = $args->{'text'};
197 $text = $self->{'iconv'}->convert( $text ) or
198 $log->logdie("can't convert '$text' to UTF-8");
199 $doc->add_text( $text );
202 $log->debug("adding ", sub { $doc->dump_draft } );
203 $self->{'db'}->put_doc($doc) || $log->warn("can't add document $uri with draft " . $doc->dump_draft . " to node " . $self->{url} . " status: " . $self->{db}->status());
209 # REST parametars validation data
212 my $estraier_rest = {
214 userdel => [ qw/name/ ],
216 nodeadd => [ qw/name label/ ],
217 nodedel => [ qw/name/ ],
220 _set_link => [ qw/url label credit/ ],
226 Issue administrative commands to C<estmaster> process and receive response
229 my $nodelist = $est->master( action => 'nodelist' );
237 my $log = $self->_get_logger;
239 my $action = $args->{action} || $log->logconfess("no action specified");
241 $log->logdie("action '$action' isn't supported") unless ($estraier_rest->{master}->{$action});
243 $log->debug("master action: $action");
245 return $self->estcall(
246 validate => 'master',
247 rest_url => $self->{masterurl} . '/master?action=' . $action ,
267 my $log = $self->_get_logger;
269 my @labels = $self->master( action => 'nodelist' );
271 $log->debug("got labels: ", join("|", @labels));
273 @labels = grep(/^$args->{to}\t/, @labels);
274 my $label = shift @labels;
275 (undef,$label) = split(/\t/, $label) if ($label);
278 $log->warn("can't find label for $args->{to}, skipping link creaton");
282 $log->debug("using label $label for $args->{to}");
284 return $self->estcall(
286 action => '_set_link',
287 rest_url => $self->{masterurl} . '/node/' . $args->{from} . '/_set_link' ,
288 url => $self->{masterurl} . '/node/' . $args->{to},
290 credit => $args->{credit},
296 Workhourse which does actual calls to Hyper Estraier
299 rest_url => '/master?action=' . $action,
300 validate => 'master',
304 C<rest_url> is relative URL to C<estmaster> and C<validate> is entry into
305 internal hash which will check if all parametars are available before
313 my $log = $self->_get_logger;
315 $log->debug("estcall: ",Dumper($args));
317 foreach my $p (qw/rest_url validate action/) {
318 $log->die("ectcall needs $p parametar") unless ($args->{$p});
321 my $url = $args->{rest_url};
323 $del = '&' if ($url =~ m#\?#);
327 foreach my $arg (@{ $estraier_rest->{ $args->{validate} }->{ $args->{action} } }) {
328 $log->logdie("missing parametar $arg for action $args->{action}") unless ($args->{$arg});
329 $url_args .= $del . $arg . '=' . uri_escape( $args->{$arg} );
333 $url .= $url_args if ($url_args);
335 $log->debug("calling $url");
337 my $res = $self->est_ua()->get($url);
339 if ($res->is_success) {
340 #$log->debug( $res->content );
341 return split(/\n/, $res->content) if wantarray;
342 return $res->content || 0E0;
344 $log->warn("unable to call $url: " . $res->status_line);
352 This is helper function to create C<LWP::UserAgent> object with Super User
355 my $ua = $self->est_ua( user => 'admin', passwd => 'admin' );
364 return $self->{_master_ua} if ($self->{_master_ua});
367 package AdminUserAgent;
368 use base qw/LWP::UserAgent/;
370 my $self = LWP::UserAgent::new(@_);
371 $self->agent("webpac/$VERSION");
374 sub get_basic_credentials {
375 my($self, $realm, $uri) = @_;
376 return ($self->{user}, $self->{passwd});
378 sub set_basic_credentials {
379 my ($self, $user, $passwd) = @_;
380 $self->{user} = $user;
381 $self->{passwd} = $passwd;
385 $self->{_master_ua} = AdminUserAgent->new( ) || sub {
386 my $log = $self->_get_logger;
387 $log->logdie("can't create LWP::UserAgent: $!");
390 $self->{_master_ua}->set_basic_credentials($self->{user}, $self->{passwd});
392 return $self->{_master_ua};
397 Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >>
399 =head1 COPYRIGHT & LICENSE
401 Copyright 2005 Dobrica Pavlinusic, All Rights Reserved.
403 This program is free software; you can redistribute it and/or modify it
404 under the same terms as Perl itself.
408 1; # End of WebPAC::Output::Estraier