27 |
|
|
28 |
=cut |
=cut |
29 |
|
|
30 |
|
=head1 Inheritable common methods |
31 |
|
|
32 |
|
This methods should really move somewhere else. |
33 |
|
|
34 |
=head2 _s |
=head2 _s |
35 |
|
|
36 |
Remove multiple whitespaces from string, as well as whitespaces at beginning or end |
Remove multiple whitespaces from string, as well as whitespaces at beginning or end |
60 |
This class implements Document which is collection of attributes |
This class implements Document which is collection of attributes |
61 |
(key=value), vectors (also key value) display text and hidden text. |
(key=value), vectors (also key value) display text and hidden text. |
62 |
|
|
63 |
|
|
64 |
=head2 new |
=head2 new |
65 |
|
|
66 |
Create new document, empty or from draft. |
Create new document, empty or from draft. |
180 |
push @{ $self->{htexts} }, $self->_s($text); |
push @{ $self->{htexts} }, $self->_s($text); |
181 |
} |
} |
182 |
|
|
183 |
|
|
184 |
=head2 id |
=head2 id |
185 |
|
|
186 |
Get the ID number of document. If the object has never been registred, C<-1> is returned. |
Get the ID number of document. If the object has never been registred, C<-1> is returned. |
194 |
return $self->{id}; |
return $self->{id}; |
195 |
} |
} |
196 |
|
|
197 |
|
|
198 |
=head2 attr_names |
=head2 attr_names |
199 |
|
|
200 |
Returns array with attribute names from document object. |
Returns array with attribute names from document object. |
240 |
return @{ $self->{dtexts} }; |
return @{ $self->{dtexts} }; |
241 |
} |
} |
242 |
|
|
243 |
|
|
244 |
=head2 cat_texts |
=head2 cat_texts |
245 |
|
|
246 |
Return whole text as single scalar. |
Return whole text as single scalar. |
254 |
return join(' ',@{ $self->{dtexts} }); |
return join(' ',@{ $self->{dtexts} }); |
255 |
} |
} |
256 |
|
|
257 |
|
|
258 |
=head2 dump_draft |
=head2 dump_draft |
259 |
|
|
260 |
Dump draft data from document object. |
Dump draft data from document object. |
281 |
|
|
282 |
$draft .= "\n"; |
$draft .= "\n"; |
283 |
|
|
284 |
$draft .= join("\n", @{ $self->{dtexts} }) . "\n"; |
$draft .= join("\n", @{ $self->{dtexts} }) . "\n" if ($self->{dtexts}); |
285 |
$draft .= "\t" . join("\n\t", @{ $self->{htexts} }) . "\n"; |
$draft .= "\t" . join("\n\t", @{ $self->{htexts} }) . "\n" if ($self->{htexts}); |
286 |
|
|
287 |
return $draft; |
return $draft; |
288 |
} |
} |
289 |
|
|
290 |
|
|
291 |
=head2 delete |
=head2 delete |
292 |
|
|
293 |
Empty document object |
Empty document object |
340 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
341 |
} |
} |
342 |
|
|
343 |
|
|
344 |
=head2 set_phrase |
=head2 set_phrase |
345 |
|
|
346 |
$cond->set_phrase('search phrase'); |
$cond->set_phrase('search phrase'); |
352 |
$self->{phrase} = $self->_s( shift ); |
$self->{phrase} = $self->_s( shift ); |
353 |
} |
} |
354 |
|
|
355 |
|
|
356 |
=head2 add_attr |
=head2 add_attr |
357 |
|
|
358 |
$cond->add_attr('@URI STRINC /~dpavlin/'); |
$cond->add_attr('@URI STRINC /~dpavlin/'); |
365 |
push @{ $self->{attrs} }, $self->_s( $attr ); |
push @{ $self->{attrs} }, $self->_s( $attr ); |
366 |
} |
} |
367 |
|
|
368 |
|
|
369 |
=head2 set_order |
=head2 set_order |
370 |
|
|
371 |
$cond->set_order('@mdate NUMD'); |
$cond->set_order('@mdate NUMD'); |
377 |
$self->{order} = shift; |
$self->{order} = shift; |
378 |
} |
} |
379 |
|
|
380 |
|
|
381 |
=head2 set_max |
=head2 set_max |
382 |
|
|
383 |
$cond->set_max(42); |
$cond->set_max(42); |
387 |
sub set_max { |
sub set_max { |
388 |
my $self = shift; |
my $self = shift; |
389 |
my $max = shift; |
my $max = shift; |
390 |
croak "set_max needs number" unless ($max =~ m/^\d+$/); |
croak "set_max needs number, not '$max'" unless ($max =~ m/^\d+$/); |
391 |
$self->{max} = $max; |
$self->{max} = $max; |
392 |
} |
} |
393 |
|
|
394 |
|
|
395 |
=head2 set_options |
=head2 set_options |
396 |
|
|
397 |
$cond->set_options( SURE => 1 ); |
$cond->set_options( SURE => 1 ); |
420 |
$self->{options} ||= $options->{$option}; |
$self->{options} ||= $options->{$option}; |
421 |
} |
} |
422 |
|
|
423 |
|
|
424 |
=head2 phrase |
=head2 phrase |
425 |
|
|
426 |
Return search phrase. |
Return search phrase. |
434 |
return $self->{phrase}; |
return $self->{phrase}; |
435 |
} |
} |
436 |
|
|
437 |
|
|
438 |
=head2 order |
=head2 order |
439 |
|
|
440 |
Return search result order. |
Return search result order. |
448 |
return $self->{order}; |
return $self->{order}; |
449 |
} |
} |
450 |
|
|
451 |
|
|
452 |
=head2 attrs |
=head2 attrs |
453 |
|
|
454 |
Return search result attrs. |
Return search result attrs. |
463 |
return @{ $self->{attrs} }; |
return @{ $self->{attrs} }; |
464 |
} |
} |
465 |
|
|
466 |
|
|
467 |
=head2 max |
=head2 max |
468 |
|
|
469 |
Return maximum number of results. |
Return maximum number of results. |
479 |
return $self->{max}; |
return $self->{max}; |
480 |
} |
} |
481 |
|
|
482 |
|
|
483 |
=head2 options |
=head2 options |
484 |
|
|
485 |
Return options for this condition. |
Return options for this condition. |
498 |
|
|
499 |
package Search::Estraier::ResultDocument; |
package Search::Estraier::ResultDocument; |
500 |
|
|
501 |
use Carp qw/confess croak/; |
use Carp qw/croak/; |
502 |
|
|
503 |
use Search::Estraier; |
#use Search::Estraier; |
504 |
our @ISA = qw/Search::Estraier/; |
#our @ISA = qw/Search::Estraier/; |
505 |
|
|
506 |
=head1 Search::Estraier::ResultDocument |
=head1 Search::Estraier::ResultDocument |
507 |
|
|
508 |
=head2 new |
=head2 new |
509 |
|
|
510 |
my $doc = new Search::HyperEstraier::ResultDocument( |
my $rdoc = new Search::HyperEstraier::ResultDocument( |
511 |
uri => 'http://localhost/document/uri/42', |
uri => 'http://localhost/document/uri/42', |
512 |
attrs => { |
attrs => { |
513 |
foo => 1, |
foo => 1, |
532 |
} |
} |
533 |
|
|
534 |
|
|
535 |
|
=head2 uri |
536 |
|
|
537 |
|
Return URI of result document |
538 |
|
|
539 |
|
print $rdoc->uri; |
540 |
|
|
541 |
|
=cut |
542 |
|
|
543 |
|
sub uri { |
544 |
|
my $self = shift; |
545 |
|
return $self->{uri}; |
546 |
|
} |
547 |
|
|
|
package Search::Estraier::Master; |
|
548 |
|
|
549 |
use Carp; |
=head2 attr_names |
550 |
|
|
551 |
=head1 Search::Estraier::Master |
Returns array with attribute names from result document object. |
552 |
|
|
553 |
Controll node master. This requires user with administration priviledges. |
my @attrs = $rdoc->attr_names; |
554 |
|
|
555 |
=cut |
=cut |
556 |
|
|
557 |
{ |
sub attr_names { |
558 |
package RequestAgent; |
my $self = shift; |
559 |
our @ISA = qw(LWP::UserAgent); |
croak "attr_names return array, not scalar" if (! wantarray); |
560 |
|
return sort keys %{ $self->{attrs} }; |
561 |
|
} |
562 |
|
|
|
sub new { |
|
|
my $self = LWP::UserAgent::new(@_); |
|
|
$self->agent("Search-Estraier/$Search::Estraer::VERSION"); |
|
|
$self; |
|
|
} |
|
563 |
|
|
564 |
sub get_basic_credentials { |
=head2 attr |
565 |
my($self, $realm, $uri) = @_; |
|
566 |
# return ($user, $password); |
Returns value of an attribute. |
567 |
} |
|
568 |
|
my $value = $rdoc->attr( 'attribute' ); |
569 |
|
|
570 |
|
=cut |
571 |
|
|
572 |
|
sub attr { |
573 |
|
my $self = shift; |
574 |
|
my $name = shift || return; |
575 |
|
return $self->{attrs}->{ $name }; |
576 |
} |
} |
577 |
|
|
578 |
|
|
579 |
|
=head2 snippet |
580 |
|
|
581 |
=head2 new |
Return snippet from result document |
582 |
|
|
583 |
|
print $rdoc->snippet; |
584 |
|
|
585 |
Create new connection to node master. |
=cut |
586 |
|
|
587 |
my $master = new Search::Estraier::Master( |
sub snippet { |
588 |
url => 'http://localhost:1978', |
my $self = shift; |
589 |
user => 'admin', |
return $self->{snippet}; |
590 |
passwd => 'admin', |
} |
591 |
|
|
592 |
|
|
593 |
|
=head2 keywords |
594 |
|
|
595 |
|
Return keywords from result document |
596 |
|
|
597 |
|
print $rdoc->keywords; |
598 |
|
|
599 |
|
=cut |
600 |
|
|
601 |
|
sub keywords { |
602 |
|
my $self = shift; |
603 |
|
return $self->{keywords}; |
604 |
|
} |
605 |
|
|
606 |
|
|
607 |
|
package Search::Estraier::NodeResult; |
608 |
|
|
609 |
|
use Carp qw/croak/; |
610 |
|
|
611 |
|
#use Search::Estraier; |
612 |
|
#our @ISA = qw/Search::Estraier/; |
613 |
|
|
614 |
|
=head1 Search::Estraier::NodeResult |
615 |
|
|
616 |
|
=head2 new |
617 |
|
|
618 |
|
my $res = new Search::HyperEstraier::NodeResult( |
619 |
|
docs => @array_of_rdocs, |
620 |
|
hits => %hash_with_hints, |
621 |
); |
); |
622 |
|
|
623 |
=cut |
=cut |
627 |
my $self = {@_}; |
my $self = {@_}; |
628 |
bless($self, $class); |
bless($self, $class); |
629 |
|
|
630 |
foreach my $p (qw/url user passwd/) { |
foreach my $f (qw/docs hints/) { |
631 |
croak "need $p" unless ($self->{$p}); |
croak "missing $f for ResultDocument" unless defined($self->{$f}); |
632 |
} |
} |
633 |
|
|
634 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
635 |
} |
} |
636 |
|
|
637 |
|
|
638 |
|
=head2 doc_num |
639 |
|
|
640 |
|
Return number of documents |
641 |
|
|
642 |
|
print $res->doc_num; |
643 |
|
|
644 |
|
=cut |
645 |
|
|
646 |
|
sub doc_num { |
647 |
|
my $self = shift; |
648 |
|
return $#{$self->{docs}}; |
649 |
|
} |
650 |
|
|
651 |
|
|
652 |
|
=head2 get_doc |
653 |
|
|
654 |
|
Return single document |
655 |
|
|
656 |
|
my $doc = $res->get_doc( 42 ); |
657 |
|
|
658 |
|
Returns undef if document doesn't exist. |
659 |
|
|
660 |
|
=cut |
661 |
|
|
662 |
|
sub get_doc { |
663 |
|
my $self = shift; |
664 |
|
my $num = shift; |
665 |
|
croak "expect number as argument, not '$num'" unless ($num =~ m/^\d+$/); |
666 |
|
return undef if ($num < 0 || $num > $self->{docs}); |
667 |
|
return $self->{docs}->[$num]; |
668 |
|
} |
669 |
|
|
670 |
|
|
671 |
|
=head2 hint |
672 |
|
|
673 |
|
Return specific hint from results. |
674 |
|
|
675 |
|
print $rec->hint( 'VERSION' ); |
676 |
|
|
677 |
|
Possible hints are: C<VERSION>, C<NODE>, C<HIT>, C<HINT#n>, C<DOCNUM>, C<WORDNUM>, |
678 |
|
C<TIME>, C<LINK#n>, C<VIEW>. |
679 |
|
|
680 |
|
=cut |
681 |
|
|
682 |
|
sub hint { |
683 |
|
my $self = shift; |
684 |
|
my $key = shift || return; |
685 |
|
return $self->{hints}->{$key}; |
686 |
|
} |
687 |
|
|
688 |
|
|
689 |
|
package Search::Estraier::Node; |
690 |
|
|
691 |
|
use Carp qw/carp croak confess/; |
692 |
|
use URI; |
693 |
|
use MIME::Base64; |
694 |
|
use IO::Socket::INET; |
695 |
|
use URI::Escape qw/uri_escape/; |
696 |
|
|
697 |
|
=head1 Search::Estraier::Node |
698 |
|
|
699 |
|
=head2 new |
700 |
|
|
701 |
|
my $node = new Search::HyperEstraier::Node; |
702 |
|
|
703 |
|
=cut |
704 |
|
|
705 |
|
sub new { |
706 |
|
my $class = shift; |
707 |
|
my $self = { |
708 |
|
pxport => -1, |
709 |
|
timeout => 0, # this used to be -1 |
710 |
|
dnum => -1, |
711 |
|
wnum => -1, |
712 |
|
size => -1.0, |
713 |
|
wwidth => 480, |
714 |
|
hwidth => 96, |
715 |
|
awidth => 96, |
716 |
|
status => -1, |
717 |
|
}; |
718 |
|
bless($self, $class); |
719 |
|
|
720 |
|
if (@_) { |
721 |
|
$self->{debug} = shift; |
722 |
|
warn "## Node debug on\n"; |
723 |
|
} |
724 |
|
|
725 |
|
$self ? return $self : return undef; |
726 |
|
} |
727 |
|
|
728 |
|
|
729 |
|
=head2 set_url |
730 |
|
|
731 |
|
Specify URL to node server |
732 |
|
|
733 |
|
$node->set_url('http://localhost:1978'); |
734 |
|
|
735 |
|
=cut |
736 |
|
|
737 |
|
sub set_url { |
738 |
|
my $self = shift; |
739 |
|
$self->{url} = shift; |
740 |
|
} |
741 |
|
|
742 |
|
|
743 |
|
=head2 set_proxy |
744 |
|
|
745 |
|
Specify proxy server to connect to node server |
746 |
|
|
747 |
|
$node->set_proxy('proxy.example.com', 8080); |
748 |
|
|
749 |
|
=cut |
750 |
|
|
751 |
|
sub set_proxy { |
752 |
|
my $self = shift; |
753 |
|
my ($host,$port) = @_; |
754 |
|
croak "proxy port must be number, not '$port'" unless ($port =~ m/^\d+$/); |
755 |
|
$self->{pxhost} = $host; |
756 |
|
$self->{pxport} = $port; |
757 |
|
} |
758 |
|
|
759 |
|
|
760 |
|
=head2 set_timeout |
761 |
|
|
762 |
|
Specify timeout of connection in seconds |
763 |
|
|
764 |
|
$node->set_timeout( 15 ); |
765 |
|
|
766 |
|
=cut |
767 |
|
|
768 |
|
sub set_timeout { |
769 |
|
my $self = shift; |
770 |
|
my $sec = shift; |
771 |
|
croak "timeout must be number, not '$sec'" unless ($sec =~ m/^\d+$/); |
772 |
|
$self->{timeout} = $sec; |
773 |
|
} |
774 |
|
|
775 |
|
|
776 |
|
=head2 set_auth |
777 |
|
|
778 |
|
Specify name and password for authentication to node server. |
779 |
|
|
780 |
|
$node->set_auth('clint','eastwood'); |
781 |
|
|
782 |
|
=cut |
783 |
|
|
784 |
|
sub set_auth { |
785 |
|
my $self = shift; |
786 |
|
my ($login,$passwd) = @_; |
787 |
|
my $basic_auth = encode_base64( "$login:$passwd" ); |
788 |
|
chomp($basic_auth); |
789 |
|
$self->{auth} = $basic_auth; |
790 |
|
} |
791 |
|
|
792 |
|
|
793 |
|
=head2 status |
794 |
|
|
795 |
|
Return status code of last request. |
796 |
|
|
797 |
|
print $node->status; |
798 |
|
|
799 |
|
C<-1> means connection failure. |
800 |
|
|
801 |
|
=cut |
802 |
|
|
803 |
|
sub status { |
804 |
|
my $self = shift; |
805 |
|
return $self->{status}; |
806 |
|
} |
807 |
|
|
808 |
|
|
809 |
|
=head2 put_doc |
810 |
|
|
811 |
|
Add a document |
812 |
|
|
813 |
|
$node->put_doc( $document_draft ) or die "can't add document"; |
814 |
|
|
815 |
|
Return true on success or false on failture. |
816 |
|
|
817 |
|
=cut |
818 |
|
|
819 |
|
sub put_doc { |
820 |
|
my $self = shift; |
821 |
|
my $doc = shift || return; |
822 |
|
return unless ($self->{url} && $doc->isa('Search::Estraier::Document')); |
823 |
|
$self->shuttle_url( $self->{url} . '/put_doc', |
824 |
|
'text/x-estraier-draft', |
825 |
|
$doc->dump_draft, |
826 |
|
undef |
827 |
|
) == 200; |
828 |
|
} |
829 |
|
|
830 |
|
|
831 |
|
=head2 out_doc |
832 |
|
|
833 |
|
Remove a document |
834 |
|
|
835 |
|
$node->out_doc( document_id ) or "can't remove document"; |
836 |
|
|
837 |
|
Return true on success or false on failture. |
838 |
|
|
839 |
|
=cut |
840 |
|
|
841 |
|
sub out_doc { |
842 |
|
my $self = shift; |
843 |
|
my $id = shift || return; |
844 |
|
return unless ($self->{url}); |
845 |
|
croak "id must be number, not '$id'" unless ($id =~ m/^\d+$/); |
846 |
|
$self->shuttle_url( $self->{url} . '/out_doc', |
847 |
|
'application/x-www-form-urlencoded', |
848 |
|
"id=$id", |
849 |
|
undef |
850 |
|
) == 200; |
851 |
|
} |
852 |
|
|
853 |
|
|
854 |
|
=head2 out_doc_by_uri |
855 |
|
|
856 |
|
Remove a registrated document using it's uri |
857 |
|
|
858 |
|
$node->out_doc_by_uri( 'file:///document/uri/42' ) or "can't remove document"; |
859 |
|
|
860 |
|
Return true on success or false on failture. |
861 |
|
|
862 |
|
=cut |
863 |
|
|
864 |
|
sub out_doc_by_uri { |
865 |
|
my $self = shift; |
866 |
|
my $uri = shift || return; |
867 |
|
return unless ($self->{url}); |
868 |
|
$self->shuttle_url( $self->{url} . '/out_doc', |
869 |
|
'application/x-www-form-urlencoded', |
870 |
|
"uri=$uri", |
871 |
|
undef |
872 |
|
) == 200; |
873 |
|
} |
874 |
|
|
875 |
|
|
876 |
|
=head2 edit_doc |
877 |
|
|
878 |
|
Edit attributes of a document |
879 |
|
|
880 |
|
$node->edit_doc( $document_draft ) or die "can't edit document"; |
881 |
|
|
882 |
|
Return true on success or false on failture. |
883 |
|
|
884 |
|
=cut |
885 |
|
|
886 |
|
sub edit_doc { |
887 |
|
my $self = shift; |
888 |
|
my $doc = shift || return; |
889 |
|
return unless ($self->{url} && $doc->isa('Search::Estraier::Document')); |
890 |
|
$self->shuttle_url( $self->{url} . '/edit_doc', |
891 |
|
'text/x-estraier-draft', |
892 |
|
$doc->dump_draft, |
893 |
|
undef |
894 |
|
) == 200; |
895 |
|
} |
896 |
|
|
897 |
|
|
898 |
|
=head2 get_doc |
899 |
|
|
900 |
|
Retreive document |
901 |
|
|
902 |
|
my $doc = $node->get_doc( document_id ) or die "can't get document"; |
903 |
|
|
904 |
|
Return true on success or false on failture. |
905 |
|
|
906 |
|
=cut |
907 |
|
|
908 |
|
sub get_doc { |
909 |
|
my $self = shift; |
910 |
|
my $id = shift || return; |
911 |
|
return $self->_fetch_doc( id => $id ); |
912 |
|
} |
913 |
|
|
914 |
|
|
915 |
|
=head2 get_doc_by_uri |
916 |
|
|
917 |
|
Retreive document |
918 |
|
|
919 |
|
my $doc = $node->get_doc_by_uri( 'file:///document/uri/42' ) or die "can't get document"; |
920 |
|
|
921 |
|
Return true on success or false on failture. |
922 |
|
|
923 |
|
=cut |
924 |
|
|
925 |
|
sub get_doc_by_uri { |
926 |
|
my $self = shift; |
927 |
|
my $uri = shift || return; |
928 |
|
return $self->_fetch_doc( uri => $uri ); |
929 |
|
} |
930 |
|
|
931 |
|
|
932 |
|
=head2 get_doc_attr |
933 |
|
|
934 |
|
Retrieve the value of an atribute from object |
935 |
|
|
936 |
|
my $val = $node->get_doc_attr( document_id, 'attribute_name' ) or |
937 |
|
die "can't get document attribute"; |
938 |
|
|
939 |
|
=cut |
940 |
|
|
941 |
|
sub get_doc_attr { |
942 |
|
my $self = shift; |
943 |
|
my ($id,$name) = @_; |
944 |
|
return unless ($id && $name); |
945 |
|
return $self->_fetch_doc( id => $id, attr => $name ); |
946 |
|
} |
947 |
|
|
948 |
|
|
949 |
|
=head2 get_doc_attr_by_uri |
950 |
|
|
951 |
|
Retrieve the value of an atribute from object |
952 |
|
|
953 |
|
my $val = $node->get_doc_attr_by_uri( document_id, 'attribute_name' ) or |
954 |
|
die "can't get document attribute"; |
955 |
|
|
956 |
|
=cut |
957 |
|
|
958 |
|
sub get_doc_attr_by_uri { |
959 |
|
my $self = shift; |
960 |
|
my ($uri,$name) = @_; |
961 |
|
return unless ($uri && $name); |
962 |
|
return $self->_fetch_doc( uri => $uri, attr => $name ); |
963 |
|
} |
964 |
|
|
965 |
|
|
966 |
|
=head2 etch_doc |
967 |
|
|
968 |
|
Exctract document keywords |
969 |
|
|
970 |
|
my $keywords = $node->etch_doc( document_id ) or die "can't etch document"; |
971 |
|
|
972 |
|
=cut |
973 |
|
|
974 |
|
sub etch_doc { |
975 |
|
my $self = shift; |
976 |
|
my $id = shift || return; |
977 |
|
return $self->_fetch_doc( id => $id, etch => 1 ); |
978 |
|
} |
979 |
|
|
980 |
|
=head2 etch_doc_by_uri |
981 |
|
|
982 |
|
Retreive document |
983 |
|
|
984 |
|
my $keywords = $node->etch_doc_by_uri( 'file:///document/uri/42' ) or die "can't etch document"; |
985 |
|
|
986 |
|
Return true on success or false on failture. |
987 |
|
|
988 |
|
=cut |
989 |
|
|
990 |
|
sub etch_doc_by_uri { |
991 |
|
my $self = shift; |
992 |
|
my $uri = shift || return; |
993 |
|
return $self->_fetch_doc( uri => $uri, etch => 1 ); |
994 |
|
} |
995 |
|
|
996 |
|
|
997 |
|
=head2 uri_to_id |
998 |
|
|
999 |
|
Get ID of document specified by URI |
1000 |
|
|
1001 |
|
my $id = $node->uri_to_id( 'file:///document/uri/42' ); |
1002 |
|
|
1003 |
|
=cut |
1004 |
|
|
1005 |
|
sub uri_to_id { |
1006 |
|
my $self = shift; |
1007 |
|
my $uri = shift || return; |
1008 |
|
return $self->_fetch_doc( uri => $uri, path => '/uri_to_id', chomp_resbody => 1 ); |
1009 |
|
} |
1010 |
|
|
1011 |
|
|
1012 |
|
=head2 _fetch_doc |
1013 |
|
|
1014 |
|
Private function used for implementing of C<get_doc>, C<get_doc_by_uri>, |
1015 |
|
C<etch_doc>, C<etch_doc_by_uri>. |
1016 |
|
|
1017 |
|
# this will decode received draft into Search::Estraier::Document object |
1018 |
|
my $doc = $node->_fetch_doc( id => 42 ); |
1019 |
|
my $doc = $node->_fetch_doc( uri => 'file:///document/uri/42' ); |
1020 |
|
|
1021 |
|
# to extract keywords, add etch |
1022 |
|
my $doc = $node->_fetch_doc( id => 42, etch => 1 ); |
1023 |
|
my $doc = $node->_fetch_doc( uri => 'file:///document/uri/42', etch => 1 ); |
1024 |
|
|
1025 |
|
# to get document attrubute add attr |
1026 |
|
my $doc = $node->_fetch_doc( id => 42, attr => '@mdate' ); |
1027 |
|
my $doc = $node->_fetch_doc( uri => 'file:///document/uri/42', attr => '@mdate' ); |
1028 |
|
|
1029 |
|
# more general form which allows implementation of |
1030 |
|
# uri_to_id |
1031 |
|
my $id = $node->_fetch_doc( |
1032 |
|
uri => 'file:///document/uri/42', |
1033 |
|
path => '/uri_to_id', |
1034 |
|
chomp_resbody => 1 |
1035 |
|
); |
1036 |
|
|
1037 |
|
=cut |
1038 |
|
|
1039 |
|
sub _fetch_doc { |
1040 |
|
my $self = shift; |
1041 |
|
my $a = {@_}; |
1042 |
|
return unless ( ($a->{id} || $a->{uri}) && $self->{url} ); |
1043 |
|
|
1044 |
|
my ($arg, $resbody); |
1045 |
|
|
1046 |
|
my $path = $a->{path} || '/get_doc'; |
1047 |
|
$path = '/etch_doc' if ($a->{etch}); |
1048 |
|
|
1049 |
|
if ($a->{id}) { |
1050 |
|
croak "id must be numberm not '$a->{id}'" unless ($a->{id} =~ m/^\d+$/); |
1051 |
|
$arg = 'id=' . $a->{id}; |
1052 |
|
} elsif ($a->{uri}) { |
1053 |
|
$arg = 'uri=' . $a->{uri}; |
1054 |
|
} else { |
1055 |
|
confess "unhandled argument. Need id or uri."; |
1056 |
|
} |
1057 |
|
|
1058 |
|
if ($a->{attr}) { |
1059 |
|
$path = '/get_doc_attr'; |
1060 |
|
$arg .= '&attr=' . uri_escape($a->{attr}); |
1061 |
|
$a->{chomp_resbody} = 1; |
1062 |
|
} |
1063 |
|
|
1064 |
|
my $rv = $self->shuttle_url( $self->{url} . $path, |
1065 |
|
'application/x-www-form-urlencoded', |
1066 |
|
$arg, |
1067 |
|
\$resbody, |
1068 |
|
); |
1069 |
|
|
1070 |
|
return if ($rv != 200); |
1071 |
|
|
1072 |
|
if ($a->{etch}) { |
1073 |
|
$self->{kwords} = {}; |
1074 |
|
return +{} unless ($resbody); |
1075 |
|
foreach my $l (split(/\n/, $resbody)) { |
1076 |
|
my ($k,$v) = split(/\t/, $l, 2); |
1077 |
|
$self->{kwords}->{$k} = $v if ($v); |
1078 |
|
} |
1079 |
|
return $self->{kwords}; |
1080 |
|
} elsif ($a->{chomp_resbody}) { |
1081 |
|
return unless (defined($resbody)); |
1082 |
|
chomp($resbody); |
1083 |
|
return $resbody; |
1084 |
|
} else { |
1085 |
|
return new Search::Estraier::Document($resbody); |
1086 |
|
} |
1087 |
|
} |
1088 |
|
|
1089 |
|
|
1090 |
|
=head2 name |
1091 |
|
|
1092 |
|
my $node_name = $node->name; |
1093 |
|
|
1094 |
|
=cut |
1095 |
|
|
1096 |
|
sub name { |
1097 |
|
my $self = shift; |
1098 |
|
$self->set_info unless ($self->{name}); |
1099 |
|
return $self->{name}; |
1100 |
|
} |
1101 |
|
|
1102 |
|
|
1103 |
|
=head2 label |
1104 |
|
|
1105 |
|
my $node_label = $node->label; |
1106 |
|
|
1107 |
|
=cut |
1108 |
|
|
1109 |
|
sub label { |
1110 |
|
my $self = shift; |
1111 |
|
$self->set_info unless ($self->{label}); |
1112 |
|
return $self->{label}; |
1113 |
|
} |
1114 |
|
|
1115 |
|
|
1116 |
|
=head2 doc_num |
1117 |
|
|
1118 |
|
my $documents_in_node = $node->doc_num; |
1119 |
|
|
1120 |
|
=cut |
1121 |
|
|
1122 |
|
sub doc_num { |
1123 |
|
my $self = shift; |
1124 |
|
$self->set_info if ($self->{dnum} < 0); |
1125 |
|
return $self->{dnum}; |
1126 |
|
} |
1127 |
|
|
1128 |
|
|
1129 |
|
=head2 word_num |
1130 |
|
|
1131 |
|
my $words_in_node = $node->word_num; |
1132 |
|
|
1133 |
|
=cut |
1134 |
|
|
1135 |
|
sub word_num { |
1136 |
|
my $self = shift; |
1137 |
|
$self->set_info if ($self->{wnum} < 0); |
1138 |
|
return $self->{wnum}; |
1139 |
|
} |
1140 |
|
|
1141 |
|
|
1142 |
|
=head2 size |
1143 |
|
|
1144 |
|
my $node_size = $node->size; |
1145 |
|
|
1146 |
|
=cut |
1147 |
|
|
1148 |
|
sub size { |
1149 |
|
my $self = shift; |
1150 |
|
$self->set_info if ($self->{size} < 0); |
1151 |
|
return $self->{size}; |
1152 |
|
} |
1153 |
|
|
1154 |
|
|
1155 |
|
|
1156 |
|
=head2 shuttle_url |
1157 |
|
|
1158 |
|
This is method which uses C<IO::Socket::INET> to communicate with Hyper Estraier node |
1159 |
|
master. |
1160 |
|
|
1161 |
|
my $rv = shuttle_url( $url, $content_type, \$req_body, \$resbody ); |
1162 |
|
|
1163 |
|
C<$resheads> and C<$resbody> booleans controll if response headers and/or response |
1164 |
|
body will be saved within object. |
1165 |
|
|
1166 |
|
=cut |
1167 |
|
|
1168 |
|
sub shuttle_url { |
1169 |
|
my $self = shift; |
1170 |
|
|
1171 |
|
my ($url, $content_type, $reqbody, $resbody) = @_; |
1172 |
|
|
1173 |
|
$self->{status} = -1; |
1174 |
|
|
1175 |
|
warn "## $url\n" if ($self->{debug}); |
1176 |
|
|
1177 |
|
$url = new URI($url); |
1178 |
|
if ( |
1179 |
|
!$url || !$url->scheme || !$url->scheme eq 'http' || |
1180 |
|
!$url->host || !$url->port || $url->port < 1 |
1181 |
|
) { |
1182 |
|
carp "can't parse $url\n"; |
1183 |
|
return -1; |
1184 |
|
} |
1185 |
|
|
1186 |
|
my ($host,$port,$query) = ($url->host, $url->port, $url->path); |
1187 |
|
|
1188 |
|
if ($self->{pxhost}) { |
1189 |
|
($host,$port) = ($self->{pxhost}, $self->{pxport}); |
1190 |
|
$query = "http://$host:$port/$query"; |
1191 |
|
} |
1192 |
|
|
1193 |
|
$query .= '?' . $url->query if ($url->query && ! $reqbody); |
1194 |
|
|
1195 |
|
my $headers; |
1196 |
|
|
1197 |
|
if ($reqbody) { |
1198 |
|
$headers .= "POST $query HTTP/1.0\r\n"; |
1199 |
|
} else { |
1200 |
|
$headers .= "GET $query HTTP/1.0\r\n"; |
1201 |
|
} |
1202 |
|
|
1203 |
|
$headers .= "Host: " . $url->host . ":" . $url->port . "\r\n"; |
1204 |
|
$headers .= "Connection: close\r\n"; |
1205 |
|
$headers .= "User-Agent: Search-Estraier/$Search::Estraier::VERSION\r\n"; |
1206 |
|
$headers .= "Content-Type: $content_type\r\n"; |
1207 |
|
$headers .= "Authorization: Basic $self->{auth}\r\n"; |
1208 |
|
my $len = 0; |
1209 |
|
{ |
1210 |
|
use bytes; |
1211 |
|
$len = length($reqbody) if ($reqbody); |
1212 |
|
} |
1213 |
|
$headers .= "Content-Length: $len\r\n"; |
1214 |
|
$headers .= "\r\n"; |
1215 |
|
|
1216 |
|
my $sock = IO::Socket::INET->new( |
1217 |
|
PeerAddr => $host, |
1218 |
|
PeerPort => $port, |
1219 |
|
Proto => 'tcp', |
1220 |
|
Timeout => $self->{timeout} || 90, |
1221 |
|
); |
1222 |
|
|
1223 |
|
if (! $sock) { |
1224 |
|
carp "can't open socket to $host:$port"; |
1225 |
|
return -1; |
1226 |
|
} |
1227 |
|
|
1228 |
|
warn $headers if ($self->{debug}); |
1229 |
|
|
1230 |
|
print $sock $headers or |
1231 |
|
carp "can't send headers to network:\n$headers\n" and return -1; |
1232 |
|
|
1233 |
|
if ($reqbody) { |
1234 |
|
warn "$reqbody\n" if ($self->{debug}); |
1235 |
|
print $sock $reqbody or |
1236 |
|
carp "can't send request body to network:\n$$reqbody\n" and return -1; |
1237 |
|
} |
1238 |
|
|
1239 |
|
my $line = <$sock>; |
1240 |
|
chomp($line); |
1241 |
|
my ($schema, $res_status, undef) = split(/ */, $line, 3); |
1242 |
|
return if ($schema !~ /^HTTP/ || ! $res_status); |
1243 |
|
|
1244 |
|
$self->{status} = $res_status; |
1245 |
|
warn "## response status: $res_status\n" if ($self->{debug}); |
1246 |
|
|
1247 |
|
# skip rest of headers |
1248 |
|
$line = <$sock>; |
1249 |
|
while ($line) { |
1250 |
|
$line = <$sock>; |
1251 |
|
$line =~ s/[\r\n]+$//; |
1252 |
|
warn "## ", $line || 'NULL', " ##\n" if ($self->{debug}); |
1253 |
|
}; |
1254 |
|
|
1255 |
|
# read body |
1256 |
|
$len = 0; |
1257 |
|
do { |
1258 |
|
$len = read($sock, my $buf, 8192); |
1259 |
|
$$resbody .= $buf if ($resbody); |
1260 |
|
} while ($len); |
1261 |
|
|
1262 |
|
warn "## response body:\n$$resbody\n" if ($resbody && $self->{debug}); |
1263 |
|
|
1264 |
|
return $self->{status}; |
1265 |
|
} |
1266 |
|
|
1267 |
|
|
1268 |
|
=head2 set_info |
1269 |
|
|
1270 |
|
Set information for node |
1271 |
|
|
1272 |
|
$node->set_info; |
1273 |
|
|
1274 |
|
=cut |
1275 |
|
|
1276 |
|
sub set_info { |
1277 |
|
my $self = shift; |
1278 |
|
|
1279 |
|
$self->{status} = -1; |
1280 |
|
return unless ($self->{url}); |
1281 |
|
|
1282 |
|
my $resbody; |
1283 |
|
my $rv = $self->shuttle_url( $self->{url} . '/inform', |
1284 |
|
'text/plain', |
1285 |
|
undef, |
1286 |
|
\$resbody, |
1287 |
|
); |
1288 |
|
|
1289 |
|
return if ($rv != 200 || !$resbody); |
1290 |
|
|
1291 |
|
chomp($resbody); |
1292 |
|
|
1293 |
|
( $self->{name}, $self->{label}, $self->{dnum}, $self->{wnum}, $self->{size} ) = |
1294 |
|
split(/\t/, $resbody, 5); |
1295 |
|
|
1296 |
|
} |
1297 |
|
|
1298 |
### |
### |
1299 |
|
|