1 |
package Search::Estraier; |
2 |
|
3 |
use 5.008; |
4 |
use strict; |
5 |
use warnings; |
6 |
|
7 |
our $VERSION = '0.07_2'; |
8 |
|
9 |
=head1 NAME |
10 |
|
11 |
Search::Estraier - pure perl module to use Hyper Estraier search engine |
12 |
|
13 |
=head1 SYNOPSIS |
14 |
|
15 |
=head2 Simple indexer |
16 |
|
17 |
use Search::Estraier; |
18 |
|
19 |
# create and configure node |
20 |
my $node = new Search::Estraier::Node( |
21 |
url => 'http://localhost:1978/node/test', |
22 |
user => 'admin', |
23 |
passwd => 'admin', |
24 |
create => 1, |
25 |
label => 'Label for node', |
26 |
croak_on_error => 1, |
27 |
); |
28 |
|
29 |
# create document |
30 |
my $doc = new Search::Estraier::Document; |
31 |
|
32 |
# add attributes |
33 |
$doc->add_attr('@uri', "http://estraier.gov/example.txt"); |
34 |
$doc->add_attr('@title', "Over the Rainbow"); |
35 |
|
36 |
# add body text to document |
37 |
$doc->add_text("Somewhere over the rainbow. Way up high."); |
38 |
$doc->add_text("There's a land that I heard of once in a lullaby."); |
39 |
|
40 |
die "error: ", $node->status,"\n" unless (eval { $node->put_doc($doc) }); |
41 |
|
42 |
=head2 Simple searcher |
43 |
|
44 |
use Search::Estraier; |
45 |
|
46 |
# create and configure node |
47 |
my $node = new Search::Estraier::Node( |
48 |
url => 'http://localhost:1978/node/test', |
49 |
user => 'admin', |
50 |
passwd => 'admin', |
51 |
croak_on_error => 1, |
52 |
); |
53 |
|
54 |
# create condition |
55 |
my $cond = new Search::Estraier::Condition; |
56 |
|
57 |
# set search phrase |
58 |
$cond->set_phrase("rainbow AND lullaby"); |
59 |
|
60 |
my $nres = $node->search($cond, 0); |
61 |
|
62 |
if (defined($nres)) { |
63 |
print "Got ", $nres->hits, " results\n"; |
64 |
|
65 |
# for each document in results |
66 |
for my $i ( 0 ... $nres->doc_num - 1 ) { |
67 |
# get result document |
68 |
my $rdoc = $nres->get_doc($i); |
69 |
# display attribte |
70 |
print "URI: ", $rdoc->attr('@uri'),"\n"; |
71 |
print "Title: ", $rdoc->attr('@title'),"\n"; |
72 |
print $rdoc->snippet,"\n"; |
73 |
} |
74 |
} else { |
75 |
die "error: ", $node->status,"\n"; |
76 |
} |
77 |
|
78 |
=head1 DESCRIPTION |
79 |
|
80 |
This module is implementation of node API of Hyper Estraier. Since it's |
81 |
perl-only module with dependencies only on standard perl modules, it will |
82 |
run on all platforms on which perl runs. It doesn't require compilation |
83 |
or Hyper Estraier development files on target machine. |
84 |
|
85 |
It is implemented as multiple packages which closly resamble Ruby |
86 |
implementation. It also includes methods to manage nodes. |
87 |
|
88 |
There are few examples in C<scripts> directory of this distribution. |
89 |
|
90 |
=cut |
91 |
|
92 |
=head1 Inheritable common methods |
93 |
|
94 |
This methods should really move somewhere else. |
95 |
|
96 |
=head2 _s |
97 |
|
98 |
Remove multiple whitespaces from string, as well as whitespaces at beginning or end |
99 |
|
100 |
my $text = $self->_s(" this is a text "); |
101 |
$text = 'this is a text'; |
102 |
|
103 |
=cut |
104 |
|
105 |
sub _s { |
106 |
my $text = $_[1]; |
107 |
return unless defined($text); |
108 |
$text =~ s/\s\s+/ /gs; |
109 |
$text =~ s/^\s+//; |
110 |
$text =~ s/\s+$//; |
111 |
return $text; |
112 |
} |
113 |
|
114 |
package Search::Estraier::Document; |
115 |
|
116 |
use Carp qw/croak confess/; |
117 |
|
118 |
use Search::Estraier; |
119 |
our @ISA = qw/Search::Estraier/; |
120 |
|
121 |
=head1 Search::Estraier::Document |
122 |
|
123 |
This class implements Document which is single item in Hyper Estraier. |
124 |
|
125 |
It's is collection of: |
126 |
|
127 |
=over 4 |
128 |
|
129 |
=item attributes |
130 |
|
131 |
C<< 'key' => 'value' >> pairs which can later be used for filtering of results |
132 |
|
133 |
You can add common filters to C<attrindex> in estmaster's C<_conf> |
134 |
file for better performance. See C<attrindex> in |
135 |
L<Hyper Estraier P2P Guide|http://hyperestraier.sourceforge.net/nguide-en.html>. |
136 |
|
137 |
=item vectors |
138 |
|
139 |
also C<< 'key' => 'value' >> pairs |
140 |
|
141 |
=item display text |
142 |
|
143 |
Text which will be used to create searchable corpus of your index and |
144 |
included in snippet output. |
145 |
|
146 |
=item hidden text |
147 |
|
148 |
Text which will be searchable, but will not be included in snippet. |
149 |
|
150 |
=back |
151 |
|
152 |
=head2 new |
153 |
|
154 |
Create new document, empty or from draft. |
155 |
|
156 |
my $doc = new Search::HyperEstraier::Document; |
157 |
my $doc2 = new Search::HyperEstraier::Document( $draft ); |
158 |
|
159 |
=cut |
160 |
|
161 |
sub new { |
162 |
my $class = shift; |
163 |
my $self = {}; |
164 |
bless($self, $class); |
165 |
|
166 |
$self->{id} = -1; |
167 |
|
168 |
my $draft = shift; |
169 |
|
170 |
if ($draft) { |
171 |
my $in_text = 0; |
172 |
foreach my $line (split(/\n/, $draft)) { |
173 |
|
174 |
if ($in_text) { |
175 |
if ($line =~ /^\t/) { |
176 |
push @{ $self->{htexts} }, substr($line, 1); |
177 |
} else { |
178 |
push @{ $self->{dtexts} }, $line; |
179 |
} |
180 |
next; |
181 |
} |
182 |
|
183 |
if ($line =~ m/^%VECTOR\t(.+)$/) { |
184 |
my @fields = split(/\t/, $1); |
185 |
for my $i ( 0 .. ($#fields - 1) ) { |
186 |
$self->{kwords}->{ $fields[ $i ] } = $fields[ $i + 1 ]; |
187 |
$i++; |
188 |
} |
189 |
next; |
190 |
} elsif ($line =~ m/^%/) { |
191 |
# What is this? comment? |
192 |
#warn "$line\n"; |
193 |
next; |
194 |
} elsif ($line =~ m/^$/) { |
195 |
$in_text = 1; |
196 |
next; |
197 |
} elsif ($line =~ m/^(.+)=(.*)$/) { |
198 |
$self->{attrs}->{ $1 } = $2; |
199 |
next; |
200 |
} |
201 |
|
202 |
warn "draft ignored: '$line'\n"; |
203 |
} |
204 |
} |
205 |
|
206 |
$self ? return $self : return undef; |
207 |
} |
208 |
|
209 |
|
210 |
=head2 add_attr |
211 |
|
212 |
Add an attribute. |
213 |
|
214 |
$doc->add_attr( name => 'value' ); |
215 |
|
216 |
Delete attribute using |
217 |
|
218 |
$doc->add_attr( name => undef ); |
219 |
|
220 |
=cut |
221 |
|
222 |
sub add_attr { |
223 |
my $self = shift; |
224 |
my $attrs = {@_}; |
225 |
|
226 |
while (my ($name, $value) = each %{ $attrs }) { |
227 |
if (! defined($value)) { |
228 |
delete( $self->{attrs}->{ $self->_s($name) } ); |
229 |
} else { |
230 |
$self->{attrs}->{ $self->_s($name) } = $self->_s($value); |
231 |
} |
232 |
} |
233 |
|
234 |
return 1; |
235 |
} |
236 |
|
237 |
|
238 |
=head2 add_text |
239 |
|
240 |
Add a sentence of text. |
241 |
|
242 |
$doc->add_text('this is example text to display'); |
243 |
|
244 |
=cut |
245 |
|
246 |
sub add_text { |
247 |
my $self = shift; |
248 |
my $text = shift; |
249 |
return unless defined($text); |
250 |
|
251 |
push @{ $self->{dtexts} }, $self->_s($text); |
252 |
} |
253 |
|
254 |
|
255 |
=head2 add_hidden_text |
256 |
|
257 |
Add a hidden sentence. |
258 |
|
259 |
$doc->add_hidden_text('this is example text just for search'); |
260 |
|
261 |
=cut |
262 |
|
263 |
sub add_hidden_text { |
264 |
my $self = shift; |
265 |
my $text = shift; |
266 |
return unless defined($text); |
267 |
|
268 |
push @{ $self->{htexts} }, $self->_s($text); |
269 |
} |
270 |
|
271 |
|
272 |
=head2 id |
273 |
|
274 |
Get the ID number of document. If the object has never been registred, C<-1> is returned. |
275 |
|
276 |
print $doc->id; |
277 |
|
278 |
=cut |
279 |
|
280 |
sub id { |
281 |
my $self = shift; |
282 |
return $self->{id}; |
283 |
} |
284 |
|
285 |
|
286 |
=head2 attr_names |
287 |
|
288 |
Returns array with attribute names from document object. |
289 |
|
290 |
my @attrs = $doc->attr_names; |
291 |
|
292 |
=cut |
293 |
|
294 |
sub attr_names { |
295 |
my $self = shift; |
296 |
return unless ($self->{attrs}); |
297 |
#croak "attr_names return array, not scalar" if (! wantarray); |
298 |
return sort keys %{ $self->{attrs} }; |
299 |
} |
300 |
|
301 |
|
302 |
=head2 attr |
303 |
|
304 |
Returns value of an attribute. |
305 |
|
306 |
my $value = $doc->attr( 'attribute' ); |
307 |
|
308 |
=cut |
309 |
|
310 |
sub attr { |
311 |
my $self = shift; |
312 |
my $name = shift; |
313 |
return unless (defined($name) && $self->{attrs}); |
314 |
return $self->{attrs}->{ $name }; |
315 |
} |
316 |
|
317 |
|
318 |
=head2 texts |
319 |
|
320 |
Returns array with text sentences. |
321 |
|
322 |
my @texts = $doc->texts; |
323 |
|
324 |
=cut |
325 |
|
326 |
sub texts { |
327 |
my $self = shift; |
328 |
#confess "texts return array, not scalar" if (! wantarray); |
329 |
return @{ $self->{dtexts} } if ($self->{dtexts}); |
330 |
} |
331 |
|
332 |
|
333 |
=head2 cat_texts |
334 |
|
335 |
Return whole text as single scalar. |
336 |
|
337 |
my $text = $doc->cat_texts; |
338 |
|
339 |
=cut |
340 |
|
341 |
sub cat_texts { |
342 |
my $self = shift; |
343 |
return join(' ',@{ $self->{dtexts} }) if ($self->{dtexts}); |
344 |
} |
345 |
|
346 |
|
347 |
=head2 dump_draft |
348 |
|
349 |
Dump draft data from document object. |
350 |
|
351 |
print $doc->dump_draft; |
352 |
|
353 |
=cut |
354 |
|
355 |
sub dump_draft { |
356 |
my $self = shift; |
357 |
my $draft; |
358 |
|
359 |
foreach my $attr_name (sort keys %{ $self->{attrs} }) { |
360 |
next unless defined(my $v = $self->{attrs}->{$attr_name}); |
361 |
$draft .= $attr_name . '=' . $v . "\n"; |
362 |
} |
363 |
|
364 |
if ($self->{kwords}) { |
365 |
$draft .= '%%VECTOR'; |
366 |
while (my ($key, $value) = each %{ $self->{kwords} }) { |
367 |
$draft .= "\t$key\t$value"; |
368 |
} |
369 |
$draft .= "\n"; |
370 |
} |
371 |
|
372 |
$draft .= "\n"; |
373 |
|
374 |
$draft .= join("\n", @{ $self->{dtexts} }) . "\n" if ($self->{dtexts}); |
375 |
$draft .= "\t" . join("\n\t", @{ $self->{htexts} }) . "\n" if ($self->{htexts}); |
376 |
|
377 |
return $draft; |
378 |
} |
379 |
|
380 |
|
381 |
=head2 delete |
382 |
|
383 |
Empty document object |
384 |
|
385 |
$doc->delete; |
386 |
|
387 |
This function is addition to original Ruby API, and since it was included in C wrappers it's here as a |
388 |
convinience. Document objects which go out of scope will be destroyed |
389 |
automatically. |
390 |
|
391 |
=cut |
392 |
|
393 |
sub delete { |
394 |
my $self = shift; |
395 |
|
396 |
foreach my $data (qw/attrs dtexts stexts kwords/) { |
397 |
delete($self->{$data}); |
398 |
} |
399 |
|
400 |
$self->{id} = -1; |
401 |
|
402 |
return 1; |
403 |
} |
404 |
|
405 |
|
406 |
|
407 |
package Search::Estraier::Condition; |
408 |
|
409 |
use Carp qw/carp confess croak/; |
410 |
|
411 |
use Search::Estraier; |
412 |
our @ISA = qw/Search::Estraier/; |
413 |
|
414 |
=head1 Search::Estraier::Condition |
415 |
|
416 |
=head2 new |
417 |
|
418 |
my $cond = new Search::HyperEstraier::Condition; |
419 |
|
420 |
=cut |
421 |
|
422 |
sub new { |
423 |
my $class = shift; |
424 |
my $self = {}; |
425 |
bless($self, $class); |
426 |
|
427 |
$self->{max} = -1; |
428 |
$self->{options} = 0; |
429 |
|
430 |
$self ? return $self : return undef; |
431 |
} |
432 |
|
433 |
|
434 |
=head2 set_phrase |
435 |
|
436 |
$cond->set_phrase('search phrase'); |
437 |
|
438 |
=cut |
439 |
|
440 |
sub set_phrase { |
441 |
my $self = shift; |
442 |
$self->{phrase} = $self->_s( shift ); |
443 |
} |
444 |
|
445 |
|
446 |
=head2 add_attr |
447 |
|
448 |
$cond->add_attr('@URI STRINC /~dpavlin/'); |
449 |
|
450 |
=cut |
451 |
|
452 |
sub add_attr { |
453 |
my $self = shift; |
454 |
my $attr = shift || return; |
455 |
push @{ $self->{attrs} }, $self->_s( $attr ); |
456 |
} |
457 |
|
458 |
|
459 |
=head2 set_order |
460 |
|
461 |
$cond->set_order('@mdate NUMD'); |
462 |
|
463 |
=cut |
464 |
|
465 |
sub set_order { |
466 |
my $self = shift; |
467 |
$self->{order} = shift; |
468 |
} |
469 |
|
470 |
|
471 |
=head2 set_max |
472 |
|
473 |
$cond->set_max(42); |
474 |
|
475 |
=cut |
476 |
|
477 |
sub set_max { |
478 |
my $self = shift; |
479 |
my $max = shift; |
480 |
croak "set_max needs number, not '$max'" unless ($max =~ m/^\d+$/); |
481 |
$self->{max} = $max; |
482 |
} |
483 |
|
484 |
|
485 |
=head2 set_options |
486 |
|
487 |
$cond->set_options( 'SURE' ); |
488 |
|
489 |
$cond->set_options( qw/AGITO NOIDF SIMPLE/ ); |
490 |
|
491 |
Possible options are: |
492 |
|
493 |
=over 8 |
494 |
|
495 |
=item SURE |
496 |
|
497 |
check every N-gram |
498 |
|
499 |
=item USUAL |
500 |
|
501 |
check every second N-gram |
502 |
|
503 |
=item FAST |
504 |
|
505 |
check every third N-gram |
506 |
|
507 |
=item AGITO |
508 |
|
509 |
check every fourth N-gram |
510 |
|
511 |
=item NOIDF |
512 |
|
513 |
don't perform TF-IDF tuning |
514 |
|
515 |
=item SIMPLE |
516 |
|
517 |
use simplified query phrase |
518 |
|
519 |
=back |
520 |
|
521 |
Skipping N-grams will speed up search, but reduce accuracy. Every call to C<set_options> will reset previous |
522 |
options; |
523 |
|
524 |
This option changed in version C<0.04> of this module. It's backwards compatibile. |
525 |
|
526 |
=cut |
527 |
|
528 |
my $options = { |
529 |
SURE => 1 << 0, |
530 |
USUAL => 1 << 1, |
531 |
FAST => 1 << 2, |
532 |
AGITO => 1 << 3, |
533 |
NOIDF => 1 << 4, |
534 |
SIMPLE => 1 << 10, |
535 |
}; |
536 |
|
537 |
sub set_options { |
538 |
my $self = shift; |
539 |
my $opt = 0; |
540 |
foreach my $option (@_) { |
541 |
my $mask; |
542 |
unless ($mask = $options->{$option}) { |
543 |
if ($option eq '1') { |
544 |
next; |
545 |
} else { |
546 |
croak "unknown option $option"; |
547 |
} |
548 |
} |
549 |
$opt += $mask; |
550 |
} |
551 |
$self->{options} = $opt; |
552 |
} |
553 |
|
554 |
|
555 |
=head2 phrase |
556 |
|
557 |
Return search phrase. |
558 |
|
559 |
print $cond->phrase; |
560 |
|
561 |
=cut |
562 |
|
563 |
sub phrase { |
564 |
my $self = shift; |
565 |
return $self->{phrase}; |
566 |
} |
567 |
|
568 |
|
569 |
=head2 order |
570 |
|
571 |
Return search result order. |
572 |
|
573 |
print $cond->order; |
574 |
|
575 |
=cut |
576 |
|
577 |
sub order { |
578 |
my $self = shift; |
579 |
return $self->{order}; |
580 |
} |
581 |
|
582 |
|
583 |
=head2 attrs |
584 |
|
585 |
Return search result attrs. |
586 |
|
587 |
my @cond_attrs = $cond->attrs; |
588 |
|
589 |
=cut |
590 |
|
591 |
sub attrs { |
592 |
my $self = shift; |
593 |
#croak "attrs return array, not scalar" if (! wantarray); |
594 |
return @{ $self->{attrs} } if ($self->{attrs}); |
595 |
} |
596 |
|
597 |
|
598 |
=head2 max |
599 |
|
600 |
Return maximum number of results. |
601 |
|
602 |
print $cond->max; |
603 |
|
604 |
C<-1> is returned for unitialized value, C<0> is unlimited. |
605 |
|
606 |
=cut |
607 |
|
608 |
sub max { |
609 |
my $self = shift; |
610 |
return $self->{max}; |
611 |
} |
612 |
|
613 |
|
614 |
=head2 options |
615 |
|
616 |
Return options for this condition. |
617 |
|
618 |
print $cond->options; |
619 |
|
620 |
Options are returned in numerical form. |
621 |
|
622 |
=cut |
623 |
|
624 |
sub options { |
625 |
my $self = shift; |
626 |
return $self->{options}; |
627 |
} |
628 |
|
629 |
|
630 |
=head2 set_skip |
631 |
|
632 |
Set number of skipped documents from beginning of results |
633 |
|
634 |
$cond->set_skip(42); |
635 |
|
636 |
Similar to C<offset> in RDBMS. |
637 |
|
638 |
=cut |
639 |
|
640 |
sub set_skip { |
641 |
my $self = shift; |
642 |
$self->{skip} = shift; |
643 |
} |
644 |
|
645 |
=head2 skip |
646 |
|
647 |
Return skip for this condition. |
648 |
|
649 |
print $cond->skip; |
650 |
|
651 |
=cut |
652 |
|
653 |
sub skip { |
654 |
my $self = shift; |
655 |
return $self->{skip}; |
656 |
} |
657 |
|
658 |
|
659 |
package Search::Estraier::ResultDocument; |
660 |
|
661 |
use Carp qw/croak/; |
662 |
|
663 |
#use Search::Estraier; |
664 |
#our @ISA = qw/Search::Estraier/; |
665 |
|
666 |
=head1 Search::Estraier::ResultDocument |
667 |
|
668 |
=head2 new |
669 |
|
670 |
my $rdoc = new Search::HyperEstraier::ResultDocument( |
671 |
uri => 'http://localhost/document/uri/42', |
672 |
attrs => { |
673 |
foo => 1, |
674 |
bar => 2, |
675 |
}, |
676 |
snippet => 'this is a text of snippet' |
677 |
keywords => 'this\tare\tkeywords' |
678 |
); |
679 |
|
680 |
=cut |
681 |
|
682 |
sub new { |
683 |
my $class = shift; |
684 |
my $self = {@_}; |
685 |
bless($self, $class); |
686 |
|
687 |
croak "missing uri for ResultDocument" unless defined($self->{uri}); |
688 |
|
689 |
$self ? return $self : return undef; |
690 |
} |
691 |
|
692 |
|
693 |
=head2 uri |
694 |
|
695 |
Return URI of result document |
696 |
|
697 |
print $rdoc->uri; |
698 |
|
699 |
=cut |
700 |
|
701 |
sub uri { |
702 |
my $self = shift; |
703 |
return $self->{uri}; |
704 |
} |
705 |
|
706 |
|
707 |
=head2 attr_names |
708 |
|
709 |
Returns array with attribute names from result document object. |
710 |
|
711 |
my @attrs = $rdoc->attr_names; |
712 |
|
713 |
=cut |
714 |
|
715 |
sub attr_names { |
716 |
my $self = shift; |
717 |
croak "attr_names return array, not scalar" if (! wantarray); |
718 |
return sort keys %{ $self->{attrs} }; |
719 |
} |
720 |
|
721 |
|
722 |
=head2 attr |
723 |
|
724 |
Returns value of an attribute. |
725 |
|
726 |
my $value = $rdoc->attr( 'attribute' ); |
727 |
|
728 |
=cut |
729 |
|
730 |
sub attr { |
731 |
my $self = shift; |
732 |
my $name = shift || return; |
733 |
return $self->{attrs}->{ $name }; |
734 |
} |
735 |
|
736 |
|
737 |
=head2 snippet |
738 |
|
739 |
Return snippet from result document |
740 |
|
741 |
print $rdoc->snippet; |
742 |
|
743 |
=cut |
744 |
|
745 |
sub snippet { |
746 |
my $self = shift; |
747 |
return $self->{snippet}; |
748 |
} |
749 |
|
750 |
|
751 |
=head2 keywords |
752 |
|
753 |
Return keywords from result document |
754 |
|
755 |
print $rdoc->keywords; |
756 |
|
757 |
=cut |
758 |
|
759 |
sub keywords { |
760 |
my $self = shift; |
761 |
return $self->{keywords}; |
762 |
} |
763 |
|
764 |
|
765 |
package Search::Estraier::NodeResult; |
766 |
|
767 |
use Carp qw/croak/; |
768 |
|
769 |
#use Search::Estraier; |
770 |
#our @ISA = qw/Search::Estraier/; |
771 |
|
772 |
=head1 Search::Estraier::NodeResult |
773 |
|
774 |
=head2 new |
775 |
|
776 |
my $res = new Search::HyperEstraier::NodeResult( |
777 |
docs => @array_of_rdocs, |
778 |
hits => %hash_with_hints, |
779 |
); |
780 |
|
781 |
=cut |
782 |
|
783 |
sub new { |
784 |
my $class = shift; |
785 |
my $self = {@_}; |
786 |
bless($self, $class); |
787 |
|
788 |
foreach my $f (qw/docs hints/) { |
789 |
croak "missing $f for ResultDocument" unless defined($self->{$f}); |
790 |
} |
791 |
|
792 |
$self ? return $self : return undef; |
793 |
} |
794 |
|
795 |
|
796 |
=head2 doc_num |
797 |
|
798 |
Return number of documents |
799 |
|
800 |
print $res->doc_num; |
801 |
|
802 |
This will return real number of documents (limited by C<max>). |
803 |
If you want to get total number of hits, see C<hits>. |
804 |
|
805 |
=cut |
806 |
|
807 |
sub doc_num { |
808 |
my $self = shift; |
809 |
return $#{$self->{docs}} + 1; |
810 |
} |
811 |
|
812 |
|
813 |
=head2 get_doc |
814 |
|
815 |
Return single document |
816 |
|
817 |
my $doc = $res->get_doc( 42 ); |
818 |
|
819 |
Returns undef if document doesn't exist. |
820 |
|
821 |
=cut |
822 |
|
823 |
sub get_doc { |
824 |
my $self = shift; |
825 |
my $num = shift; |
826 |
croak "expect number as argument, not '$num'" unless ($num =~ m/^\d+$/); |
827 |
return undef if ($num < 0 || $num > $self->{docs}); |
828 |
return $self->{docs}->[$num]; |
829 |
} |
830 |
|
831 |
|
832 |
=head2 hint |
833 |
|
834 |
Return specific hint from results. |
835 |
|
836 |
print $res->hint( 'VERSION' ); |
837 |
|
838 |
Possible hints are: C<VERSION>, C<NODE>, C<HIT>, C<HINT#n>, C<DOCNUM>, C<WORDNUM>, |
839 |
C<TIME>, C<LINK#n>, C<VIEW>. |
840 |
|
841 |
=cut |
842 |
|
843 |
sub hint { |
844 |
my $self = shift; |
845 |
my $key = shift || return; |
846 |
return $self->{hints}->{$key}; |
847 |
} |
848 |
|
849 |
=head2 hints |
850 |
|
851 |
More perlish version of C<hint>. This one returns hash. |
852 |
|
853 |
my %hints = $res->hints; |
854 |
|
855 |
=cut |
856 |
|
857 |
sub hints { |
858 |
my $self = shift; |
859 |
return $self->{hints}; |
860 |
} |
861 |
|
862 |
=head2 hits |
863 |
|
864 |
Syntaxtic sugar for total number of hits for this query |
865 |
|
866 |
print $res->hits; |
867 |
|
868 |
It's same as |
869 |
|
870 |
print $res->hint('HIT'); |
871 |
|
872 |
but shorter. |
873 |
|
874 |
=cut |
875 |
|
876 |
sub hits { |
877 |
my $self = shift; |
878 |
return $self->{hints}->{'HIT'} || 0; |
879 |
} |
880 |
|
881 |
package Search::Estraier::Node; |
882 |
|
883 |
use Carp qw/carp croak confess/; |
884 |
use URI; |
885 |
use MIME::Base64; |
886 |
use IO::Socket::INET; |
887 |
use URI::Escape qw/uri_escape/; |
888 |
|
889 |
=head1 Search::Estraier::Node |
890 |
|
891 |
=head2 new |
892 |
|
893 |
my $node = new Search::HyperEstraier::Node; |
894 |
|
895 |
or optionally with C<url> as parametar |
896 |
|
897 |
my $node = new Search::HyperEstraier::Node( 'http://localhost:1978/node/test' ); |
898 |
|
899 |
or in more verbose form |
900 |
|
901 |
my $node = new Search::HyperEstraier::Node( |
902 |
url => 'http://localhost:1978/node/test', |
903 |
user => 'admin', |
904 |
passwd => 'admin' |
905 |
create => 1, |
906 |
label => 'optional node label', |
907 |
debug => 1, |
908 |
croak_on_error => 1 |
909 |
); |
910 |
|
911 |
with following arguments: |
912 |
|
913 |
=over 4 |
914 |
|
915 |
=item url |
916 |
|
917 |
URL to node |
918 |
|
919 |
=item user |
920 |
|
921 |
specify username for node server authentication |
922 |
|
923 |
=item passwd |
924 |
|
925 |
password for authentication |
926 |
|
927 |
=item create |
928 |
|
929 |
create node if it doesn't exists |
930 |
|
931 |
=item label |
932 |
|
933 |
optional label for new node if C<create> is used |
934 |
|
935 |
=item debug |
936 |
|
937 |
dumps a B<lot> of debugging output |
938 |
|
939 |
=item croak_on_error |
940 |
|
941 |
very helpful during development. It will croak on all errors instead of |
942 |
silently returning C<-1> (which is convention of Hyper Estraier API in other |
943 |
languages). |
944 |
|
945 |
=back |
946 |
|
947 |
=cut |
948 |
|
949 |
sub new { |
950 |
my $class = shift; |
951 |
my $self = { |
952 |
pxport => -1, |
953 |
timeout => 0, # this used to be -1 |
954 |
wwidth => 480, |
955 |
hwidth => 96, |
956 |
awidth => 96, |
957 |
status => -1, |
958 |
}; |
959 |
|
960 |
bless($self, $class); |
961 |
|
962 |
if ($#_ == 0) { |
963 |
$self->{url} = shift; |
964 |
} else { |
965 |
%$self = ( %$self, @_ ); |
966 |
|
967 |
$self->set_auth( $self->{user}, $self->{passwd} ) if ($self->{user}); |
968 |
|
969 |
warn "## Node debug on\n" if ($self->{debug}); |
970 |
} |
971 |
|
972 |
$self->{inform} = { |
973 |
dnum => -1, |
974 |
wnum => -1, |
975 |
size => -1.0, |
976 |
}; |
977 |
|
978 |
if ($self->{create}) { |
979 |
if (! eval { $self->name } || $@) { |
980 |
my $name = $1 if ($self->{url} =~ m#/node/([^/]+)/*#); |
981 |
croak "can't find node name in '$self->{url}'" unless ($name); |
982 |
my $label = $self->{label} || $name; |
983 |
$self->master( |
984 |
action => 'nodeadd', |
985 |
name => $name, |
986 |
label => $label, |
987 |
) || croak "can't create node $name ($label)"; |
988 |
} |
989 |
} |
990 |
|
991 |
$self ? return $self : return undef; |
992 |
} |
993 |
|
994 |
|
995 |
=head2 set_url |
996 |
|
997 |
Specify URL to node server |
998 |
|
999 |
$node->set_url('http://localhost:1978'); |
1000 |
|
1001 |
=cut |
1002 |
|
1003 |
sub set_url { |
1004 |
my $self = shift; |
1005 |
$self->{url} = shift; |
1006 |
} |
1007 |
|
1008 |
|
1009 |
=head2 set_proxy |
1010 |
|
1011 |
Specify proxy server to connect to node server |
1012 |
|
1013 |
$node->set_proxy('proxy.example.com', 8080); |
1014 |
|
1015 |
=cut |
1016 |
|
1017 |
sub set_proxy { |
1018 |
my $self = shift; |
1019 |
my ($host,$port) = @_; |
1020 |
croak "proxy port must be number, not '$port'" unless ($port =~ m/^\d+$/); |
1021 |
$self->{pxhost} = $host; |
1022 |
$self->{pxport} = $port; |
1023 |
} |
1024 |
|
1025 |
|
1026 |
=head2 set_timeout |
1027 |
|
1028 |
Specify timeout of connection in seconds |
1029 |
|
1030 |
$node->set_timeout( 15 ); |
1031 |
|
1032 |
=cut |
1033 |
|
1034 |
sub set_timeout { |
1035 |
my $self = shift; |
1036 |
my $sec = shift; |
1037 |
croak "timeout must be number, not '$sec'" unless ($sec =~ m/^\d+$/); |
1038 |
$self->{timeout} = $sec; |
1039 |
} |
1040 |
|
1041 |
|
1042 |
=head2 set_auth |
1043 |
|
1044 |
Specify name and password for authentication to node server. |
1045 |
|
1046 |
$node->set_auth('clint','eastwood'); |
1047 |
|
1048 |
=cut |
1049 |
|
1050 |
sub set_auth { |
1051 |
my $self = shift; |
1052 |
my ($login,$passwd) = @_; |
1053 |
my $basic_auth = encode_base64( "$login:$passwd" ); |
1054 |
chomp($basic_auth); |
1055 |
$self->{auth} = $basic_auth; |
1056 |
} |
1057 |
|
1058 |
|
1059 |
=head2 status |
1060 |
|
1061 |
Return status code of last request. |
1062 |
|
1063 |
print $node->status; |
1064 |
|
1065 |
C<-1> means connection failure. |
1066 |
|
1067 |
=cut |
1068 |
|
1069 |
sub status { |
1070 |
my $self = shift; |
1071 |
return $self->{status}; |
1072 |
} |
1073 |
|
1074 |
|
1075 |
=head2 put_doc |
1076 |
|
1077 |
Add a document |
1078 |
|
1079 |
$node->put_doc( $document_draft ) or die "can't add document"; |
1080 |
|
1081 |
Return true on success or false on failure. |
1082 |
|
1083 |
=cut |
1084 |
|
1085 |
sub put_doc { |
1086 |
my $self = shift; |
1087 |
my $doc = shift || return; |
1088 |
return unless ($self->{url} && $doc->isa('Search::Estraier::Document')); |
1089 |
if ($self->shuttle_url( $self->{url} . '/put_doc', |
1090 |
'text/x-estraier-draft', |
1091 |
$doc->dump_draft, |
1092 |
undef |
1093 |
) == 200) { |
1094 |
$self->_clear_info; |
1095 |
return 1; |
1096 |
} |
1097 |
return undef; |
1098 |
} |
1099 |
|
1100 |
|
1101 |
=head2 out_doc |
1102 |
|
1103 |
Remove a document |
1104 |
|
1105 |
$node->out_doc( document_id ) or "can't remove document"; |
1106 |
|
1107 |
Return true on success or false on failture. |
1108 |
|
1109 |
=cut |
1110 |
|
1111 |
sub out_doc { |
1112 |
my $self = shift; |
1113 |
my $id = shift || return; |
1114 |
return unless ($self->{url}); |
1115 |
croak "id must be number, not '$id'" unless ($id =~ m/^\d+$/); |
1116 |
if ($self->shuttle_url( $self->{url} . '/out_doc', |
1117 |
'application/x-www-form-urlencoded', |
1118 |
"id=$id", |
1119 |
undef |
1120 |
) == 200) { |
1121 |
$self->_clear_info; |
1122 |
return 1; |
1123 |
} |
1124 |
return undef; |
1125 |
} |
1126 |
|
1127 |
|
1128 |
=head2 out_doc_by_uri |
1129 |
|
1130 |
Remove a registrated document using it's uri |
1131 |
|
1132 |
$node->out_doc_by_uri( 'file:///document/uri/42' ) or "can't remove document"; |
1133 |
|
1134 |
Return true on success or false on failture. |
1135 |
|
1136 |
=cut |
1137 |
|
1138 |
sub out_doc_by_uri { |
1139 |
my $self = shift; |
1140 |
my $uri = shift || return; |
1141 |
return unless ($self->{url}); |
1142 |
if ($self->shuttle_url( $self->{url} . '/out_doc', |
1143 |
'application/x-www-form-urlencoded', |
1144 |
"uri=" . uri_escape($uri), |
1145 |
undef |
1146 |
) == 200) { |
1147 |
$self->_clear_info; |
1148 |
return 1; |
1149 |
} |
1150 |
return undef; |
1151 |
} |
1152 |
|
1153 |
|
1154 |
=head2 edit_doc |
1155 |
|
1156 |
Edit attributes of a document |
1157 |
|
1158 |
$node->edit_doc( $document_draft ) or die "can't edit document"; |
1159 |
|
1160 |
Return true on success or false on failture. |
1161 |
|
1162 |
=cut |
1163 |
|
1164 |
sub edit_doc { |
1165 |
my $self = shift; |
1166 |
my $doc = shift || return; |
1167 |
return unless ($self->{url} && $doc->isa('Search::Estraier::Document')); |
1168 |
if ($self->shuttle_url( $self->{url} . '/edit_doc', |
1169 |
'text/x-estraier-draft', |
1170 |
$doc->dump_draft, |
1171 |
undef |
1172 |
) == 200) { |
1173 |
$self->_clear_info; |
1174 |
return 1; |
1175 |
} |
1176 |
return undef; |
1177 |
} |
1178 |
|
1179 |
|
1180 |
=head2 get_doc |
1181 |
|
1182 |
Retreive document |
1183 |
|
1184 |
my $doc = $node->get_doc( document_id ) or die "can't get document"; |
1185 |
|
1186 |
Return true on success or false on failture. |
1187 |
|
1188 |
=cut |
1189 |
|
1190 |
sub get_doc { |
1191 |
my $self = shift; |
1192 |
my $id = shift || return; |
1193 |
return $self->_fetch_doc( id => $id ); |
1194 |
} |
1195 |
|
1196 |
|
1197 |
=head2 get_doc_by_uri |
1198 |
|
1199 |
Retreive document |
1200 |
|
1201 |
my $doc = $node->get_doc_by_uri( 'file:///document/uri/42' ) or die "can't get document"; |
1202 |
|
1203 |
Return true on success or false on failture. |
1204 |
|
1205 |
=cut |
1206 |
|
1207 |
sub get_doc_by_uri { |
1208 |
my $self = shift; |
1209 |
my $uri = shift || return; |
1210 |
return $self->_fetch_doc( uri => $uri ); |
1211 |
} |
1212 |
|
1213 |
|
1214 |
=head2 get_doc_attr |
1215 |
|
1216 |
Retrieve the value of an atribute from object |
1217 |
|
1218 |
my $val = $node->get_doc_attr( document_id, 'attribute_name' ) or |
1219 |
die "can't get document attribute"; |
1220 |
|
1221 |
=cut |
1222 |
|
1223 |
sub get_doc_attr { |
1224 |
my $self = shift; |
1225 |
my ($id,$name) = @_; |
1226 |
return unless ($id && $name); |
1227 |
return $self->_fetch_doc( id => $id, attr => $name ); |
1228 |
} |
1229 |
|
1230 |
|
1231 |
=head2 get_doc_attr_by_uri |
1232 |
|
1233 |
Retrieve the value of an atribute from object |
1234 |
|
1235 |
my $val = $node->get_doc_attr_by_uri( document_id, 'attribute_name' ) or |
1236 |
die "can't get document attribute"; |
1237 |
|
1238 |
=cut |
1239 |
|
1240 |
sub get_doc_attr_by_uri { |
1241 |
my $self = shift; |
1242 |
my ($uri,$name) = @_; |
1243 |
return unless ($uri && $name); |
1244 |
return $self->_fetch_doc( uri => $uri, attr => $name ); |
1245 |
} |
1246 |
|
1247 |
|
1248 |
=head2 etch_doc |
1249 |
|
1250 |
Exctract document keywords |
1251 |
|
1252 |
my $keywords = $node->etch_doc( document_id ) or die "can't etch document"; |
1253 |
|
1254 |
=cut |
1255 |
|
1256 |
sub etch_doc { |
1257 |
my $self = shift; |
1258 |
my $id = shift || return; |
1259 |
return $self->_fetch_doc( id => $id, etch => 1 ); |
1260 |
} |
1261 |
|
1262 |
=head2 etch_doc_by_uri |
1263 |
|
1264 |
Retreive document |
1265 |
|
1266 |
my $keywords = $node->etch_doc_by_uri( 'file:///document/uri/42' ) or die "can't etch document"; |
1267 |
|
1268 |
Return true on success or false on failture. |
1269 |
|
1270 |
=cut |
1271 |
|
1272 |
sub etch_doc_by_uri { |
1273 |
my $self = shift; |
1274 |
my $uri = shift || return; |
1275 |
return $self->_fetch_doc( uri => $uri, etch => 1 ); |
1276 |
} |
1277 |
|
1278 |
|
1279 |
=head2 uri_to_id |
1280 |
|
1281 |
Get ID of document specified by URI |
1282 |
|
1283 |
my $id = $node->uri_to_id( 'file:///document/uri/42' ); |
1284 |
|
1285 |
This method won't croak, even if using C<croak_on_error>. |
1286 |
|
1287 |
=cut |
1288 |
|
1289 |
sub uri_to_id { |
1290 |
my $self = shift; |
1291 |
my $uri = shift || return; |
1292 |
return $self->_fetch_doc( uri => $uri, path => '/uri_to_id', chomp_resbody => 1, croak_on_error => 0 ); |
1293 |
} |
1294 |
|
1295 |
|
1296 |
=head2 _fetch_doc |
1297 |
|
1298 |
Private function used for implementing of C<get_doc>, C<get_doc_by_uri>, |
1299 |
C<etch_doc>, C<etch_doc_by_uri>. |
1300 |
|
1301 |
# this will decode received draft into Search::Estraier::Document object |
1302 |
my $doc = $node->_fetch_doc( id => 42 ); |
1303 |
my $doc = $node->_fetch_doc( uri => 'file:///document/uri/42' ); |
1304 |
|
1305 |
# to extract keywords, add etch |
1306 |
my $doc = $node->_fetch_doc( id => 42, etch => 1 ); |
1307 |
my $doc = $node->_fetch_doc( uri => 'file:///document/uri/42', etch => 1 ); |
1308 |
|
1309 |
# to get document attrubute add attr |
1310 |
my $doc = $node->_fetch_doc( id => 42, attr => '@mdate' ); |
1311 |
my $doc = $node->_fetch_doc( uri => 'file:///document/uri/42', attr => '@mdate' ); |
1312 |
|
1313 |
# more general form which allows implementation of |
1314 |
# uri_to_id |
1315 |
my $id = $node->_fetch_doc( |
1316 |
uri => 'file:///document/uri/42', |
1317 |
path => '/uri_to_id', |
1318 |
chomp_resbody => 1 |
1319 |
); |
1320 |
|
1321 |
=cut |
1322 |
|
1323 |
sub _fetch_doc { |
1324 |
my $self = shift; |
1325 |
my $a = {@_}; |
1326 |
return unless ( ($a->{id} || $a->{uri}) && $self->{url} ); |
1327 |
|
1328 |
my ($arg, $resbody); |
1329 |
|
1330 |
my $path = $a->{path} || '/get_doc'; |
1331 |
$path = '/etch_doc' if ($a->{etch}); |
1332 |
|
1333 |
if ($a->{id}) { |
1334 |
croak "id must be numberm not '$a->{id}'" unless ($a->{id} =~ m/^\d+$/); |
1335 |
$arg = 'id=' . $a->{id}; |
1336 |
} elsif ($a->{uri}) { |
1337 |
$arg = 'uri=' . uri_escape($a->{uri}); |
1338 |
} else { |
1339 |
confess "unhandled argument. Need id or uri."; |
1340 |
} |
1341 |
|
1342 |
if ($a->{attr}) { |
1343 |
$path = '/get_doc_attr'; |
1344 |
$arg .= '&attr=' . uri_escape($a->{attr}); |
1345 |
$a->{chomp_resbody} = 1; |
1346 |
} |
1347 |
|
1348 |
my $rv = $self->shuttle_url( $self->{url} . $path, |
1349 |
'application/x-www-form-urlencoded', |
1350 |
$arg, |
1351 |
\$resbody, |
1352 |
$a->{croak_on_error}, |
1353 |
); |
1354 |
|
1355 |
return if ($rv != 200); |
1356 |
|
1357 |
if ($a->{etch}) { |
1358 |
$self->{kwords} = {}; |
1359 |
return +{} unless ($resbody); |
1360 |
foreach my $l (split(/\n/, $resbody)) { |
1361 |
my ($k,$v) = split(/\t/, $l, 2); |
1362 |
$self->{kwords}->{$k} = $v if ($v); |
1363 |
} |
1364 |
return $self->{kwords}; |
1365 |
} elsif ($a->{chomp_resbody}) { |
1366 |
return unless (defined($resbody)); |
1367 |
chomp($resbody); |
1368 |
return $resbody; |
1369 |
} else { |
1370 |
return new Search::Estraier::Document($resbody); |
1371 |
} |
1372 |
} |
1373 |
|
1374 |
|
1375 |
=head2 name |
1376 |
|
1377 |
my $node_name = $node->name; |
1378 |
|
1379 |
=cut |
1380 |
|
1381 |
sub name { |
1382 |
my $self = shift; |
1383 |
$self->_set_info unless ($self->{inform}->{name}); |
1384 |
return $self->{inform}->{name}; |
1385 |
} |
1386 |
|
1387 |
|
1388 |
=head2 label |
1389 |
|
1390 |
my $node_label = $node->label; |
1391 |
|
1392 |
=cut |
1393 |
|
1394 |
sub label { |
1395 |
my $self = shift; |
1396 |
$self->_set_info unless ($self->{inform}->{label}); |
1397 |
return $self->{inform}->{label}; |
1398 |
} |
1399 |
|
1400 |
|
1401 |
=head2 doc_num |
1402 |
|
1403 |
my $documents_in_node = $node->doc_num; |
1404 |
|
1405 |
=cut |
1406 |
|
1407 |
sub doc_num { |
1408 |
my $self = shift; |
1409 |
$self->_set_info if ($self->{inform}->{dnum} < 0); |
1410 |
return $self->{inform}->{dnum}; |
1411 |
} |
1412 |
|
1413 |
|
1414 |
=head2 word_num |
1415 |
|
1416 |
my $words_in_node = $node->word_num; |
1417 |
|
1418 |
=cut |
1419 |
|
1420 |
sub word_num { |
1421 |
my $self = shift; |
1422 |
$self->_set_info if ($self->{inform}->{wnum} < 0); |
1423 |
return $self->{inform}->{wnum}; |
1424 |
} |
1425 |
|
1426 |
|
1427 |
=head2 size |
1428 |
|
1429 |
my $node_size = $node->size; |
1430 |
|
1431 |
=cut |
1432 |
|
1433 |
sub size { |
1434 |
my $self = shift; |
1435 |
$self->_set_info if ($self->{inform}->{size} < 0); |
1436 |
return $self->{inform}->{size}; |
1437 |
} |
1438 |
|
1439 |
|
1440 |
=head2 search |
1441 |
|
1442 |
Search documents which match condition |
1443 |
|
1444 |
my $nres = $node->search( $cond, $depth ); |
1445 |
|
1446 |
C<$cond> is C<Search::Estraier::Condition> object, while <$depth> specifies |
1447 |
depth for meta search. |
1448 |
|
1449 |
Function results C<Search::Estraier::NodeResult> object. |
1450 |
|
1451 |
=cut |
1452 |
|
1453 |
sub search { |
1454 |
my $self = shift; |
1455 |
my ($cond, $depth) = @_; |
1456 |
return unless ($cond && defined($depth) && $self->{url}); |
1457 |
croak "cond mush be Search::Estraier::Condition, not '$cond->isa'" unless ($cond->isa('Search::Estraier::Condition')); |
1458 |
croak "depth needs number, not '$depth'" unless ($depth =~ m/^\d+$/); |
1459 |
|
1460 |
my $resbody; |
1461 |
|
1462 |
my $rv = $self->shuttle_url( $self->{url} . '/search', |
1463 |
'application/x-www-form-urlencoded', |
1464 |
$self->cond_to_query( $cond, $depth ), |
1465 |
\$resbody, |
1466 |
); |
1467 |
return if ($rv != 200); |
1468 |
|
1469 |
my @records = split /--------\[.*?\]--------(?::END)?\r?\n/, $resbody; |
1470 |
my $hintsText = splice @records, 0, 2; # starts with empty record |
1471 |
my $hints = { $hintsText =~ m/^(.*?)\t(.*?)$/gsm }; |
1472 |
|
1473 |
# process records |
1474 |
my $docs = []; |
1475 |
foreach my $record (@records) |
1476 |
{ |
1477 |
# split into keys and snippets |
1478 |
my ($keys, $snippet) = $record =~ m/^(.*?)\n\n(.*?)$/s; |
1479 |
|
1480 |
# create document hash |
1481 |
my $doc = { $keys =~ m/^(.*?)=(.*?)$/gsm }; |
1482 |
$doc->{'@keywords'} = $doc->{keywords}; |
1483 |
($doc->{keywords}) = $keys =~ m/^%VECTOR\t(.*?)$/gm; |
1484 |
$doc->{snippet} = $snippet; |
1485 |
|
1486 |
push @$docs, new Search::Estraier::ResultDocument( |
1487 |
attrs => $doc, |
1488 |
uri => $doc->{'@uri'}, |
1489 |
snippet => $snippet, |
1490 |
keywords => $doc->{'keywords'}, |
1491 |
); |
1492 |
} |
1493 |
|
1494 |
return new Search::Estraier::NodeResult( docs => $docs, hints => $hints ); |
1495 |
} |
1496 |
|
1497 |
|
1498 |
=head2 cond_to_query |
1499 |
|
1500 |
Return URI encoded string generated from Search::Estraier::Condition |
1501 |
|
1502 |
my $args = $node->cond_to_query( $cond, $depth ); |
1503 |
|
1504 |
=cut |
1505 |
|
1506 |
sub cond_to_query { |
1507 |
my $self = shift; |
1508 |
|
1509 |
my $cond = shift || return; |
1510 |
croak "condition must be Search::Estraier::Condition, not '$cond->isa'" unless ($cond->isa('Search::Estraier::Condition')); |
1511 |
my $depth = shift; |
1512 |
|
1513 |
my @args; |
1514 |
|
1515 |
if (my $phrase = $cond->phrase) { |
1516 |
push @args, 'phrase=' . uri_escape($phrase); |
1517 |
} |
1518 |
|
1519 |
if (my @attrs = $cond->attrs) { |
1520 |
for my $i ( 0 .. $#attrs ) { |
1521 |
push @args,'attr' . ($i+1) . '=' . uri_escape( $attrs[$i] ) if ($attrs[$i]); |
1522 |
} |
1523 |
} |
1524 |
|
1525 |
if (my $order = $cond->order) { |
1526 |
push @args, 'order=' . uri_escape($order); |
1527 |
} |
1528 |
|
1529 |
if (my $max = $cond->max) { |
1530 |
push @args, 'max=' . $max; |
1531 |
} else { |
1532 |
push @args, 'max=' . (1 << 30); |
1533 |
} |
1534 |
|
1535 |
if (my $options = $cond->options) { |
1536 |
push @args, 'options=' . $options; |
1537 |
} |
1538 |
|
1539 |
push @args, 'depth=' . $depth if ($depth); |
1540 |
push @args, 'wwidth=' . $self->{wwidth}; |
1541 |
push @args, 'hwidth=' . $self->{hwidth}; |
1542 |
push @args, 'awidth=' . $self->{awidth}; |
1543 |
push @args, 'skip=' . $cond->{skip} if ($cond->{skip}); |
1544 |
|
1545 |
return join('&', @args); |
1546 |
} |
1547 |
|
1548 |
|
1549 |
=head2 shuttle_url |
1550 |
|
1551 |
This is method which uses C<LWP::UserAgent> to communicate with Hyper Estraier node |
1552 |
master. |
1553 |
|
1554 |
my $rv = shuttle_url( $url, $content_type, $req_body, \$resbody ); |
1555 |
|
1556 |
C<$resheads> and C<$resbody> booleans controll if response headers and/or response |
1557 |
body will be saved within object. |
1558 |
|
1559 |
=cut |
1560 |
|
1561 |
use LWP::UserAgent; |
1562 |
|
1563 |
sub shuttle_url { |
1564 |
my $self = shift; |
1565 |
|
1566 |
my ($url, $content_type, $reqbody, $resbody, $croak_on_error) = @_; |
1567 |
|
1568 |
$croak_on_error = $self->{croak_on_error} unless defined($croak_on_error); |
1569 |
|
1570 |
$self->{status} = -1; |
1571 |
|
1572 |
warn "## $url\n" if ($self->{debug}); |
1573 |
|
1574 |
$url = new URI($url); |
1575 |
if ( |
1576 |
!$url || !$url->scheme || !$url->scheme eq 'http' || |
1577 |
!$url->host || !$url->port || $url->port < 1 |
1578 |
) { |
1579 |
carp "can't parse $url\n"; |
1580 |
return -1; |
1581 |
} |
1582 |
|
1583 |
my $ua = LWP::UserAgent->new; |
1584 |
$ua->agent( "Search-Estraier/$Search::Estraier::VERSION" ); |
1585 |
|
1586 |
my $req; |
1587 |
if ($reqbody) { |
1588 |
$req = HTTP::Request->new(POST => $url); |
1589 |
} else { |
1590 |
$req = HTTP::Request->new(GET => $url); |
1591 |
} |
1592 |
|
1593 |
$req->headers->header( 'Host' => $url->host . ":" . $url->port ); |
1594 |
$req->headers->header( 'Connection', 'close' ); |
1595 |
$req->headers->header( 'Authorization', 'Basic ' . $self->{auth} ) if ($self->{auth}); |
1596 |
$req->content_type( $content_type ); |
1597 |
|
1598 |
warn $req->headers->as_string,"\n" if ($self->{debug}); |
1599 |
|
1600 |
if ($reqbody) { |
1601 |
warn "$reqbody\n" if ($self->{debug}); |
1602 |
$req->content( $reqbody ); |
1603 |
} |
1604 |
|
1605 |
my $res = $ua->request($req) || croak "can't make request to $url: $!"; |
1606 |
|
1607 |
warn "## response status: ",$res->status_line,"\n" if ($self->{debug}); |
1608 |
|
1609 |
($self->{status}, $self->{status_message}) = split(/\s+/, $res->status_line, 2); |
1610 |
|
1611 |
if (! $res->is_success) { |
1612 |
if ($croak_on_error) { |
1613 |
croak("can't get $url: ",$res->status_line); |
1614 |
} else { |
1615 |
return -1; |
1616 |
} |
1617 |
} |
1618 |
|
1619 |
$$resbody .= $res->content; |
1620 |
|
1621 |
warn "## response body:\n$$resbody\n" if ($resbody && $self->{debug}); |
1622 |
|
1623 |
return $self->{status}; |
1624 |
} |
1625 |
|
1626 |
|
1627 |
=head2 set_snippet_width |
1628 |
|
1629 |
Set width of snippets in results |
1630 |
|
1631 |
$node->set_snippet_width( $wwidth, $hwidth, $awidth ); |
1632 |
|
1633 |
C<$wwidth> specifies whole width of snippet. It's C<480> by default. If it's C<0> snippet |
1634 |
is not sent with results. If it is negative, whole document text is sent instead of snippet. |
1635 |
|
1636 |
C<$hwidth> specified width of strings from beginning of string. Default |
1637 |
value is C<96>. Negative or zero value keep previous value. |
1638 |
|
1639 |
C<$awidth> specifies width of strings around each highlighted word. It's C<96> by default. |
1640 |
If negative of zero value is provided previous value is kept unchanged. |
1641 |
|
1642 |
=cut |
1643 |
|
1644 |
sub set_snippet_width { |
1645 |
my $self = shift; |
1646 |
|
1647 |
my ($wwidth, $hwidth, $awidth) = @_; |
1648 |
$self->{wwidth} = $wwidth; |
1649 |
$self->{hwidth} = $hwidth if ($hwidth >= 0); |
1650 |
$self->{awidth} = $awidth if ($awidth >= 0); |
1651 |
} |
1652 |
|
1653 |
|
1654 |
=head2 set_user |
1655 |
|
1656 |
Manage users of node |
1657 |
|
1658 |
$node->set_user( 'name', $mode ); |
1659 |
|
1660 |
C<$mode> can be one of: |
1661 |
|
1662 |
=over 4 |
1663 |
|
1664 |
=item 0 |
1665 |
|
1666 |
delete account |
1667 |
|
1668 |
=item 1 |
1669 |
|
1670 |
set administrative right for user |
1671 |
|
1672 |
=item 2 |
1673 |
|
1674 |
set user account as guest |
1675 |
|
1676 |
=back |
1677 |
|
1678 |
Return true on success, otherwise false. |
1679 |
|
1680 |
=cut |
1681 |
|
1682 |
sub set_user { |
1683 |
my $self = shift; |
1684 |
my ($name, $mode) = @_; |
1685 |
|
1686 |
return unless ($self->{url}); |
1687 |
croak "mode must be number, not '$mode'" unless ($mode =~ m/^\d+$/); |
1688 |
|
1689 |
$self->shuttle_url( $self->{url} . '/_set_user', |
1690 |
'application/x-www-form-urlencoded', |
1691 |
'name=' . uri_escape($name) . '&mode=' . $mode, |
1692 |
undef |
1693 |
) == 200; |
1694 |
} |
1695 |
|
1696 |
|
1697 |
=head2 set_link |
1698 |
|
1699 |
Manage node links |
1700 |
|
1701 |
$node->set_link('http://localhost:1978/node/another', 'another node label', $credit); |
1702 |
|
1703 |
If C<$credit> is negative, link is removed. |
1704 |
|
1705 |
=cut |
1706 |
|
1707 |
sub set_link { |
1708 |
my $self = shift; |
1709 |
my ($url, $label, $credit) = @_; |
1710 |
|
1711 |
return unless ($self->{url}); |
1712 |
croak "mode credit be number, not '$credit'" unless ($credit =~ m/^\d+$/); |
1713 |
|
1714 |
my $reqbody = 'url=' . uri_escape($url) . '&label=' . uri_escape($label); |
1715 |
$reqbody .= '&credit=' . $credit if ($credit > 0); |
1716 |
|
1717 |
if ($self->shuttle_url( $self->{url} . '/_set_link', |
1718 |
'application/x-www-form-urlencoded', |
1719 |
$reqbody, |
1720 |
undef |
1721 |
) == 200) { |
1722 |
# refresh node info after adding link |
1723 |
$self->_clear_info; |
1724 |
return 1; |
1725 |
} |
1726 |
return undef; |
1727 |
} |
1728 |
|
1729 |
=head2 admins |
1730 |
|
1731 |
my @admins = @{ $node->admins }; |
1732 |
|
1733 |
Return array of users with admin rights on node |
1734 |
|
1735 |
=cut |
1736 |
|
1737 |
sub admins { |
1738 |
my $self = shift; |
1739 |
$self->_set_info unless ($self->{inform}->{name}); |
1740 |
return $self->{inform}->{admins}; |
1741 |
} |
1742 |
|
1743 |
=head2 guests |
1744 |
|
1745 |
my @guests = @{ $node->guests }; |
1746 |
|
1747 |
Return array of users with guest rights on node |
1748 |
|
1749 |
=cut |
1750 |
|
1751 |
sub guests { |
1752 |
my $self = shift; |
1753 |
$self->_set_info unless ($self->{inform}->{name}); |
1754 |
return $self->{inform}->{guests}; |
1755 |
} |
1756 |
|
1757 |
=head2 links |
1758 |
|
1759 |
my $links = @{ $node->links }; |
1760 |
|
1761 |
Return array of links for this node |
1762 |
|
1763 |
=cut |
1764 |
|
1765 |
sub links { |
1766 |
my $self = shift; |
1767 |
$self->_set_info unless ($self->{inform}->{name}); |
1768 |
return $self->{inform}->{links}; |
1769 |
} |
1770 |
|
1771 |
=head2 cacheusage |
1772 |
|
1773 |
Return cache usage for a node |
1774 |
|
1775 |
my $cache = $node->cacheusage; |
1776 |
|
1777 |
=cut |
1778 |
|
1779 |
sub cacheusage { |
1780 |
my $self = shift; |
1781 |
|
1782 |
return unless ($self->{url}); |
1783 |
|
1784 |
my $resbody; |
1785 |
my $rv = $self->shuttle_url( $self->{url} . '/cacheusage', |
1786 |
'text/plain', |
1787 |
undef, |
1788 |
\$resbody, |
1789 |
); |
1790 |
|
1791 |
return if ($rv != 200 || !$resbody); |
1792 |
|
1793 |
return $resbody; |
1794 |
} |
1795 |
|
1796 |
=head2 master |
1797 |
|
1798 |
Set actions on Hyper Estraier node master (C<estmaster> process) |
1799 |
|
1800 |
$node->master( |
1801 |
action => 'sync' |
1802 |
); |
1803 |
|
1804 |
All available actions are documented in |
1805 |
L<http://hyperestraier.sourceforge.net/nguide-en.html#protocol> |
1806 |
|
1807 |
=cut |
1808 |
|
1809 |
my $estmaster_rest = { |
1810 |
shutdown => { |
1811 |
status => 202, |
1812 |
}, |
1813 |
sync => { |
1814 |
status => 202, |
1815 |
}, |
1816 |
backup => { |
1817 |
status => 202, |
1818 |
}, |
1819 |
userlist => { |
1820 |
status => 200, |
1821 |
returns => [ qw/name passwd flags fname misc/ ], |
1822 |
}, |
1823 |
useradd => { |
1824 |
required => [ qw/name passwd flags/ ], |
1825 |
optional => [ qw/fname misc/ ], |
1826 |
status => 200, |
1827 |
}, |
1828 |
userdel => { |
1829 |
required => [ qw/name/ ], |
1830 |
status => 200, |
1831 |
}, |
1832 |
nodelist => { |
1833 |
status => 200, |
1834 |
returns => [ qw/name label doc_num word_num size/ ], |
1835 |
}, |
1836 |
nodeadd => { |
1837 |
required => [ qw/name/ ], |
1838 |
optional => [ qw/label/ ], |
1839 |
status => 200, |
1840 |
}, |
1841 |
nodedel => { |
1842 |
required => [ qw/name/ ], |
1843 |
status => 200, |
1844 |
}, |
1845 |
nodeclr => { |
1846 |
required => [ qw/name/ ], |
1847 |
status => 200, |
1848 |
}, |
1849 |
nodertt => { |
1850 |
status => 200, |
1851 |
}, |
1852 |
}; |
1853 |
|
1854 |
sub master { |
1855 |
my $self = shift; |
1856 |
|
1857 |
my $args = {@_}; |
1858 |
|
1859 |
# have action? |
1860 |
my $action = $args->{action} || croak "need action, available: ", |
1861 |
join(", ",keys %{ $estmaster_rest }); |
1862 |
|
1863 |
# check if action is valid |
1864 |
my $rest = $estmaster_rest->{$action}; |
1865 |
croak "action '$action' is not supported, available actions: ", |
1866 |
join(", ",keys %{ $estmaster_rest }) unless ($rest); |
1867 |
|
1868 |
croak "BUG: action '$action' needs return status" unless ($rest->{status}); |
1869 |
|
1870 |
my @args; |
1871 |
|
1872 |
if ($rest->{required} || $rest->{optional}) { |
1873 |
|
1874 |
map { |
1875 |
croak "need parametar '$_' for action '$action'" unless ($args->{$_}); |
1876 |
push @args, $_ . '=' . uri_escape( $args->{$_} ); |
1877 |
} ( @{ $rest->{required} } ); |
1878 |
|
1879 |
map { |
1880 |
push @args, $_ . '=' . uri_escape( $args->{$_} ) if ($args->{$_}); |
1881 |
} ( @{ $rest->{optional} } ); |
1882 |
|
1883 |
} |
1884 |
|
1885 |
my $uri = new URI( $self->{url} ); |
1886 |
|
1887 |
my $resbody; |
1888 |
|
1889 |
my $status = $self->shuttle_url( |
1890 |
'http://' . $uri->host_port . '/master?action=' . $action , |
1891 |
'application/x-www-form-urlencoded', |
1892 |
join('&', @args), |
1893 |
\$resbody, |
1894 |
1, |
1895 |
) or confess "shuttle_url failed"; |
1896 |
|
1897 |
if ($status == $rest->{status}) { |
1898 |
|
1899 |
# refresh node info after sync |
1900 |
$self->_clear_info if ($action eq 'sync' || $action =~ m/^node(?:add|del|clr)$/); |
1901 |
|
1902 |
if ($rest->{returns} && wantarray) { |
1903 |
|
1904 |
my @results; |
1905 |
my $fields = $#{$rest->{returns}}; |
1906 |
|
1907 |
foreach my $line ( split(/[\r\n]/,$resbody) ) { |
1908 |
my @e = split(/\t/, $line, $fields + 1); |
1909 |
my $row; |
1910 |
foreach my $i ( 0 .. $fields) { |
1911 |
$row->{ $rest->{returns}->[$i] } = $e[ $i ]; |
1912 |
} |
1913 |
push @results, $row; |
1914 |
} |
1915 |
|
1916 |
return @results; |
1917 |
|
1918 |
} elsif ($resbody) { |
1919 |
chomp $resbody; |
1920 |
return $resbody; |
1921 |
} else { |
1922 |
return 0E0; |
1923 |
} |
1924 |
} |
1925 |
|
1926 |
carp "expected status $rest->{status}, but got $status"; |
1927 |
return undef; |
1928 |
} |
1929 |
|
1930 |
=head1 PRIVATE METHODS |
1931 |
|
1932 |
You could call those directly, but you don't have to. I hope. |
1933 |
|
1934 |
=head2 _set_info |
1935 |
|
1936 |
Set information for node |
1937 |
|
1938 |
$node->_set_info; |
1939 |
|
1940 |
=cut |
1941 |
|
1942 |
sub _set_info { |
1943 |
my $self = shift; |
1944 |
|
1945 |
$self->{status} = -1; |
1946 |
return unless ($self->{url}); |
1947 |
|
1948 |
my $resbody; |
1949 |
my $rv = $self->shuttle_url( $self->{url} . '/inform', |
1950 |
'text/plain', |
1951 |
undef, |
1952 |
\$resbody, |
1953 |
); |
1954 |
|
1955 |
return if ($rv != 200 || !$resbody); |
1956 |
|
1957 |
my @lines = split(/[\r\n]/,$resbody); |
1958 |
|
1959 |
$self->_clear_info; |
1960 |
|
1961 |
( $self->{inform}->{name}, $self->{inform}->{label}, $self->{inform}->{dnum}, |
1962 |
$self->{inform}->{wnum}, $self->{inform}->{size} ) = split(/\t/, shift @lines, 5); |
1963 |
|
1964 |
return $resbody unless (@lines); |
1965 |
|
1966 |
shift @lines; |
1967 |
|
1968 |
while(my $admin = shift @lines) { |
1969 |
push @{$self->{inform}->{admins}}, $admin; |
1970 |
} |
1971 |
|
1972 |
while(my $guest = shift @lines) { |
1973 |
push @{$self->{inform}->{guests}}, $guest; |
1974 |
} |
1975 |
|
1976 |
while(my $link = shift @lines) { |
1977 |
push @{$self->{inform}->{links}}, $link; |
1978 |
} |
1979 |
|
1980 |
return $resbody; |
1981 |
|
1982 |
} |
1983 |
|
1984 |
=head2 _clear_info |
1985 |
|
1986 |
Clear information for node |
1987 |
|
1988 |
$node->_clear_info; |
1989 |
|
1990 |
On next call to C<name>, C<label>, C<doc_num>, C<word_num> or C<size> node |
1991 |
info will be fetch again from Hyper Estraier. |
1992 |
|
1993 |
=cut |
1994 |
sub _clear_info { |
1995 |
my $self = shift; |
1996 |
$self->{inform} = { |
1997 |
dnum => -1, |
1998 |
wnum => -1, |
1999 |
size => -1.0, |
2000 |
}; |
2001 |
} |
2002 |
|
2003 |
### |
2004 |
|
2005 |
=head1 EXPORT |
2006 |
|
2007 |
Nothing. |
2008 |
|
2009 |
=head1 SEE ALSO |
2010 |
|
2011 |
L<http://hyperestraier.sourceforge.net/> |
2012 |
|
2013 |
Hyper Estraier Ruby interface on which this module is based. |
2014 |
|
2015 |
Hyper Estraier now also has pure-perl binding included in distribution. It's |
2016 |
a faster way to access databases directly if you are not running |
2017 |
C<estmaster> P2P server. |
2018 |
|
2019 |
=head1 AUTHOR |
2020 |
|
2021 |
Dobrica Pavlinusic, E<lt>dpavlin@rot13.orgE<gt> |
2022 |
|
2023 |
Robert Klep E<lt>robert@klep.nameE<gt> contributed refactored search code |
2024 |
|
2025 |
=head1 COPYRIGHT AND LICENSE |
2026 |
|
2027 |
Copyright (C) 2005-2006 by Dobrica Pavlinusic |
2028 |
|
2029 |
This library is free software; you can redistribute it and/or modify |
2030 |
it under the GPL v2 or later. |
2031 |
|
2032 |
=cut |
2033 |
|
2034 |
1; |