4 |
use strict; |
use strict; |
5 |
use warnings; |
use warnings; |
6 |
|
|
7 |
our $VERSION = '0.01'; |
our $VERSION = '0.04_1'; |
8 |
|
|
9 |
=head1 NAME |
=head1 NAME |
10 |
|
|
12 |
|
|
13 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
14 |
|
|
15 |
use Search::Estraier; |
=head2 Simple indexer |
16 |
my $est = new Search::Estraier(); |
|
17 |
|
use Search::Estraier; |
18 |
|
|
19 |
|
# create and configure node |
20 |
|
my $node = new Search::Estraier::Node; |
21 |
|
$node->set_url("http://localhost:1978/node/test"); |
22 |
|
$node->set_auth("admin","admin"); |
23 |
|
|
24 |
|
# create document |
25 |
|
my $doc = new Search::Estraier::Document; |
26 |
|
|
27 |
|
# add attributes |
28 |
|
$doc->add_attr('@uri', "http://estraier.gov/example.txt"); |
29 |
|
$doc->add_attr('@title', "Over the Rainbow"); |
30 |
|
|
31 |
|
# add body text to document |
32 |
|
$doc->add_text("Somewhere over the rainbow. Way up high."); |
33 |
|
$doc->add_text("There's a land that I heard of once in a lullaby."); |
34 |
|
|
35 |
|
die "error: ", $node->status,"\n" unless ($node->put_doc($doc)); |
36 |
|
|
37 |
|
=head2 Simple searcher |
38 |
|
|
39 |
|
use Search::Estraier; |
40 |
|
|
41 |
|
# create and configure node |
42 |
|
my $node = new Search::Estraier::Node; |
43 |
|
$node->set_url("http://localhost:1978/node/test"); |
44 |
|
$node->set_auth("admin","admin"); |
45 |
|
|
46 |
|
# create condition |
47 |
|
my $cond = new Search::Estraier::Condition; |
48 |
|
|
49 |
|
# set search phrase |
50 |
|
$cond->set_phrase("rainbow AND lullaby"); |
51 |
|
|
52 |
|
my $nres = $node->search($cond, 0); |
53 |
|
if (defined($nres)) { |
54 |
|
# for each document in results |
55 |
|
for my $i ( 0 ... $nres->doc_num - 1 ) { |
56 |
|
# get result document |
57 |
|
my $rdoc = $nres->get_doc($i); |
58 |
|
# display attribte |
59 |
|
print "URI: ", $rdoc->attr('@uri'),"\n"; |
60 |
|
print "Title: ", $rdoc->attr('@title'),"\n"; |
61 |
|
print $rdoc->snippet,"\n"; |
62 |
|
} |
63 |
|
} else { |
64 |
|
die "error: ", $node->status,"\n"; |
65 |
|
} |
66 |
|
|
67 |
=head1 DESCRIPTION |
=head1 DESCRIPTION |
68 |
|
|
74 |
It is implemented as multiple packages which closly resamble Ruby |
It is implemented as multiple packages which closly resamble Ruby |
75 |
implementation. It also includes methods to manage nodes. |
implementation. It also includes methods to manage nodes. |
76 |
|
|
77 |
|
There are few examples in C<scripts> directory of this distribution. |
78 |
|
|
79 |
=cut |
=cut |
80 |
|
|
81 |
=head1 Inheritable common methods |
=head1 Inheritable common methods |
157 |
} elsif ($line =~ m/^$/) { |
} elsif ($line =~ m/^$/) { |
158 |
$in_text = 1; |
$in_text = 1; |
159 |
next; |
next; |
160 |
} elsif ($line =~ m/^(.+)=(.+)$/) { |
} elsif ($line =~ m/^(.+)=(.*)$/) { |
161 |
$self->{attrs}->{ $1 } = $2; |
$self->{attrs}->{ $1 } = $2; |
162 |
next; |
next; |
163 |
} |
} |
164 |
|
|
165 |
warn "draft ignored: $line\n"; |
warn "draft ignored: '$line'\n"; |
166 |
} |
} |
167 |
} |
} |
168 |
|
|
320 |
my $draft; |
my $draft; |
321 |
|
|
322 |
foreach my $attr_name (sort keys %{ $self->{attrs} }) { |
foreach my $attr_name (sort keys %{ $self->{attrs} }) { |
323 |
$draft .= $attr_name . '=' . $self->{attrs}->{$attr_name} . "\n"; |
next unless(my $v = $self->{attrs}->{$attr_name}); |
324 |
|
$draft .= $attr_name . '=' . $v . "\n"; |
325 |
} |
} |
326 |
|
|
327 |
if ($self->{kwords}) { |
if ($self->{kwords}) { |
577 |
my $self = {@_}; |
my $self = {@_}; |
578 |
bless($self, $class); |
bless($self, $class); |
579 |
|
|
580 |
foreach my $f (qw/uri attrs snippet keywords/) { |
croak "missing uri for ResultDocument" unless defined($self->{uri}); |
|
croak "missing $f for ResultDocument" unless defined($self->{$f}); |
|
|
} |
|
581 |
|
|
582 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
583 |
} |
} |
736 |
return $self->{hints}->{$key}; |
return $self->{hints}->{$key}; |
737 |
} |
} |
738 |
|
|
739 |
|
=head2 hints |
740 |
|
|
741 |
|
More perlish version of C<hint>. This one returns hash. |
742 |
|
|
743 |
|
my %hints = $rec->hints; |
744 |
|
|
745 |
|
=cut |
746 |
|
|
747 |
|
sub hints { |
748 |
|
my $self = shift; |
749 |
|
return $self->{hints}; |
750 |
|
} |
751 |
|
|
752 |
package Search::Estraier::Node; |
package Search::Estraier::Node; |
753 |
|
|
763 |
|
|
764 |
my $node = new Search::HyperEstraier::Node; |
my $node = new Search::HyperEstraier::Node; |
765 |
|
|
766 |
|
or optionally with C<url> as parametar |
767 |
|
|
768 |
|
my $node = new Search::HyperEstraier::Node( 'http://localhost:1978/node/test' ); |
769 |
|
|
770 |
|
or in more verbose form |
771 |
|
|
772 |
|
my $node = new Search::HyperEstraier::Node( |
773 |
|
url => 'http://localhost:1978/node/test', |
774 |
|
debug => 1, |
775 |
|
croak_on_error => 1 |
776 |
|
); |
777 |
|
|
778 |
|
with following arguments: |
779 |
|
|
780 |
|
=over 4 |
781 |
|
|
782 |
|
=item url |
783 |
|
|
784 |
|
URL to node |
785 |
|
|
786 |
|
=item debug |
787 |
|
|
788 |
|
dumps a B<lot> of debugging output |
789 |
|
|
790 |
|
=item croak_on_error |
791 |
|
|
792 |
|
very helpful during development. It will croak on all errors instead of |
793 |
|
silently returning C<-1> (which is convention of Hyper Estraier API in other |
794 |
|
languages). |
795 |
|
|
796 |
|
=back |
797 |
|
|
798 |
=cut |
=cut |
799 |
|
|
800 |
sub new { |
sub new { |
812 |
}; |
}; |
813 |
bless($self, $class); |
bless($self, $class); |
814 |
|
|
815 |
my $args = {@_}; |
if ($#_ == 0) { |
816 |
|
$self->{url} = shift; |
817 |
|
} else { |
818 |
|
my $args = {@_}; |
819 |
|
|
820 |
|
%$self = ( %$self, @_ ); |
821 |
|
|
822 |
$self->{debug} = $args->{debug}; |
warn "## Node debug on\n" if ($self->{debug}); |
823 |
warn "## Node debug on\n" if ($self->{debug}); |
} |
824 |
|
|
825 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
826 |
} |
} |
1276 |
|
|
1277 |
my $rv = $self->shuttle_url( $self->{url} . '/search', |
my $rv = $self->shuttle_url( $self->{url} . '/search', |
1278 |
'application/x-www-form-urlencoded', |
'application/x-www-form-urlencoded', |
1279 |
$self->cond_to_query( $cond ), |
$self->cond_to_query( $cond, $depth ), |
1280 |
\$resbody, |
\$resbody, |
1281 |
); |
); |
1282 |
return if ($rv != 200); |
return if ($rv != 200); |
1370 |
|
|
1371 |
Return URI encoded string generated from Search::Estraier::Condition |
Return URI encoded string generated from Search::Estraier::Condition |
1372 |
|
|
1373 |
my $args = $node->cond_to_query( $cond ); |
my $args = $node->cond_to_query( $cond, $depth ); |
1374 |
|
|
1375 |
=cut |
=cut |
1376 |
|
|
1379 |
|
|
1380 |
my $cond = shift || return; |
my $cond = shift || return; |
1381 |
croak "condition must be Search::Estraier::Condition, not '$cond->isa'" unless ($cond->isa('Search::Estraier::Condition')); |
croak "condition must be Search::Estraier::Condition, not '$cond->isa'" unless ($cond->isa('Search::Estraier::Condition')); |
1382 |
|
my $depth = shift; |
1383 |
|
|
1384 |
my @args; |
my @args; |
1385 |
|
|
1389 |
|
|
1390 |
if (my @attrs = $cond->attrs) { |
if (my @attrs = $cond->attrs) { |
1391 |
for my $i ( 0 .. $#attrs ) { |
for my $i ( 0 .. $#attrs ) { |
1392 |
push @args,'attr' . ($i+1) . '=' . uri_escape( $attrs[$i] ); |
push @args,'attr' . ($i+1) . '=' . uri_escape( $attrs[$i] ) if ($attrs[$i]); |
1393 |
} |
} |
1394 |
} |
} |
1395 |
|
|
1407 |
push @args, 'options=' . $options; |
push @args, 'options=' . $options; |
1408 |
} |
} |
1409 |
|
|
1410 |
push @args, 'depth=' . $self->{depth} if ($self->{depth}); |
push @args, 'depth=' . $depth if ($depth); |
1411 |
push @args, 'wwidth=' . $self->{wwidth}; |
push @args, 'wwidth=' . $self->{wwidth}; |
1412 |
push @args, 'hwidth=' . $self->{hwidth}; |
push @args, 'hwidth=' . $self->{hwidth}; |
1413 |
push @args, 'awidth=' . $self->{awidth}; |
push @args, 'awidth=' . $self->{awidth}; |
1418 |
|
|
1419 |
=head2 shuttle_url |
=head2 shuttle_url |
1420 |
|
|
1421 |
This is method which uses C<IO::Socket::INET> to communicate with Hyper Estraier node |
This is method which uses C<LWP::UserAgent> to communicate with Hyper Estraier node |
1422 |
master. |
master. |
1423 |
|
|
1424 |
my $rv = shuttle_url( $url, $content_type, $req_body, \$resbody ); |
my $rv = shuttle_url( $url, $content_type, $req_body, \$resbody ); |
1460 |
|
|
1461 |
$req->headers->header( 'Host' => $url->host . ":" . $url->port ); |
$req->headers->header( 'Host' => $url->host . ":" . $url->port ); |
1462 |
$req->headers->header( 'Connection', 'close' ); |
$req->headers->header( 'Connection', 'close' ); |
1463 |
$req->headers->header( 'Authorization', 'Basic ' . $self->{auth} ); |
$req->headers->header( 'Authorization', 'Basic ' . $self->{auth} ) if ($self->{auth}); |
1464 |
$req->content_type( $content_type ); |
$req->content_type( $content_type ); |
1465 |
|
|
1466 |
warn $req->headers->as_string,"\n" if ($self->{debug}); |
warn $req->headers->as_string,"\n" if ($self->{debug}); |
1474 |
|
|
1475 |
warn "## response status: ",$res->status_line,"\n" if ($self->{debug}); |
warn "## response status: ",$res->status_line,"\n" if ($self->{debug}); |
1476 |
|
|
|
return -1 if (! $res->is_success); |
|
|
|
|
1477 |
($self->{status}, $self->{status_message}) = split(/\s+/, $res->status_line, 2); |
($self->{status}, $self->{status_message}) = split(/\s+/, $res->status_line, 2); |
1478 |
|
|
1479 |
|
if (! $res->is_success) { |
1480 |
|
if ($self->{croak_on_error}) { |
1481 |
|
croak("can't get $url: ",$res->status_line); |
1482 |
|
} else { |
1483 |
|
return -1; |
1484 |
|
} |
1485 |
|
} |
1486 |
|
|
1487 |
$$resbody .= $res->content; |
$$resbody .= $res->content; |
1488 |
|
|
1489 |
warn "## response body:\n$$resbody\n" if ($resbody && $self->{debug}); |
warn "## response body:\n$$resbody\n" if ($resbody && $self->{debug}); |
1583 |
$reqbody .= '&credit=' . $credit if ($credit > 0); |
$reqbody .= '&credit=' . $credit if ($credit > 0); |
1584 |
|
|
1585 |
$self->shuttle_url( $self->{url} . '/_set_link', |
$self->shuttle_url( $self->{url} . '/_set_link', |
1586 |
'text/plain', |
'application/x-www-form-urlencoded', |
1587 |
$reqbody, |
$reqbody, |
1588 |
undef |
undef |
1589 |
) == 200; |
) == 200; |