4 |
use strict; |
use strict; |
5 |
use warnings; |
use warnings; |
6 |
|
|
|
require Exporter; |
|
|
|
|
|
our @ISA = qw(Exporter); |
|
|
|
|
|
our %EXPORT_TAGS = ( 'all' => [ qw( |
|
|
) ] ); |
|
|
|
|
|
our @EXPORT_OK = ( @{ $EXPORT_TAGS{'all'} } ); |
|
|
|
|
|
our @EXPORT = qw( |
|
|
); |
|
|
|
|
7 |
our $VERSION = '0.00'; |
our $VERSION = '0.00'; |
8 |
|
|
|
use Carp; |
|
|
|
|
9 |
=head1 NAME |
=head1 NAME |
10 |
|
|
11 |
Search::Estraier - pure perl module to use Hyper Estraier search engine |
Search::Estraier - pure perl module to use Hyper Estraier search engine |
27 |
|
|
28 |
=cut |
=cut |
29 |
|
|
30 |
|
=head2 _s |
31 |
|
|
32 |
|
Remove multiple whitespaces from string, as well as whitespaces at beginning or end |
33 |
|
|
34 |
|
my $text = $self->_s(" this is a text "); |
35 |
|
$text = 'this is a text'; |
36 |
|
|
37 |
|
=cut |
38 |
|
|
39 |
|
sub _s { |
40 |
|
my $text = $_[1] || return; |
41 |
|
$text =~ s/\s\s+/ /gs; |
42 |
|
$text =~ s/^\s+//; |
43 |
|
$text =~ s/\s+$//; |
44 |
|
return $text; |
45 |
|
} |
46 |
|
|
47 |
package Search::Estraier::Document; |
package Search::Estraier::Document; |
48 |
|
|
49 |
use Carp qw/croak confess/; |
use Carp qw/croak confess/; |
50 |
|
|
51 |
|
use Search::Estraier; |
52 |
|
our @ISA = qw/Search::Estraier/; |
53 |
|
|
54 |
=head1 Search::Estraier::Document |
=head1 Search::Estraier::Document |
55 |
|
|
56 |
Document for HyperEstraier |
This class implements Document which is collection of attributes |
57 |
|
(key=value), vectors (also key value) display text and hidden text. |
58 |
|
|
59 |
=head2 new |
=head2 new |
60 |
|
|
61 |
|
Create new document, empty or from draft. |
62 |
|
|
63 |
my $doc = new Search::HyperEstraier::Document; |
my $doc = new Search::HyperEstraier::Document; |
64 |
|
my $doc2 = new Search::HyperEstraier::Document( $draft ); |
65 |
|
|
66 |
=cut |
=cut |
67 |
|
|
68 |
sub new { |
sub new { |
69 |
my $class = shift; |
my $class = shift; |
70 |
my $self = {@_}; |
my $self = {}; |
71 |
bless($self, $class); |
bless($self, $class); |
72 |
|
|
73 |
$self->{id} = -1; |
$self->{id} = -1; |
74 |
|
|
75 |
|
my $draft = shift; |
76 |
|
|
77 |
|
if ($draft) { |
78 |
|
my $in_text = 0; |
79 |
|
foreach my $line (split(/\n/, $draft)) { |
80 |
|
|
81 |
|
if ($in_text) { |
82 |
|
if ($line =~ /^\t/) { |
83 |
|
push @{ $self->{htexts} }, substr($line, 1); |
84 |
|
} else { |
85 |
|
push @{ $self->{dtexts} }, $line; |
86 |
|
} |
87 |
|
next; |
88 |
|
} |
89 |
|
|
90 |
|
if ($line =~ m/^%VECTOR\t(.+)$/) { |
91 |
|
my @fields = split(/\t/, $1); |
92 |
|
for my $i ( 0 .. ($#fields - 1) ) { |
93 |
|
$self->{kwords}->{ $fields[ $i ] } = $fields[ $i + 1 ]; |
94 |
|
$i++; |
95 |
|
} |
96 |
|
next; |
97 |
|
} elsif ($line =~ m/^%/) { |
98 |
|
# What is this? comment? |
99 |
|
#warn "$line\n"; |
100 |
|
next; |
101 |
|
} elsif ($line =~ m/^$/) { |
102 |
|
$in_text = 1; |
103 |
|
next; |
104 |
|
} elsif ($line =~ m/^(.+)=(.+)$/) { |
105 |
|
$self->{attrs}->{ $1 } = $2; |
106 |
|
next; |
107 |
|
} |
108 |
|
|
109 |
|
warn "draft ignored: $line\n"; |
110 |
|
} |
111 |
|
} |
112 |
|
|
113 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
114 |
} |
} |
115 |
|
|
132 |
|
|
133 |
while (my ($name, $value) = each %{ $attrs }) { |
while (my ($name, $value) = each %{ $attrs }) { |
134 |
if (! defined($value)) { |
if (! defined($value)) { |
135 |
delete( $self->{attrs}->{_s($name)} ); |
delete( $self->{attrs}->{ $self->_s($name) } ); |
136 |
} else { |
} else { |
137 |
$self->{attrs}->{_s($name)} = _s($value); |
$self->{attrs}->{ $self->_s($name) } = $self->_s($value); |
138 |
} |
} |
139 |
} |
} |
140 |
|
|
155 |
my $text = shift; |
my $text = shift; |
156 |
return unless defined($text); |
return unless defined($text); |
157 |
|
|
158 |
push @{ $self->{dtexts} }, _s($text); |
push @{ $self->{dtexts} }, $self->_s($text); |
159 |
} |
} |
160 |
|
|
161 |
|
|
172 |
my $text = shift; |
my $text = shift; |
173 |
return unless defined($text); |
return unless defined($text); |
174 |
|
|
175 |
push @{ $self->{htexts} }, _s($text); |
push @{ $self->{htexts} }, $self->_s($text); |
176 |
} |
} |
177 |
|
|
178 |
=head2 id |
=head2 id |
284 |
|
|
285 |
$doc->delete; |
$doc->delete; |
286 |
|
|
287 |
|
This function is addition to original Ruby API, and since it was included in C wrappers it's here as a |
288 |
|
convinience. Document objects which go out of scope will be destroyed |
289 |
|
automatically. |
290 |
|
|
291 |
=cut |
=cut |
292 |
|
|
293 |
sub delete { |
sub delete { |
294 |
my $self = shift; |
my $self = shift; |
295 |
|
|
296 |
foreach my $data (qw/attrs dtexts stexts/) { |
foreach my $data (qw/attrs dtexts stexts kwords/) { |
297 |
delete($self->{$data}); |
delete($self->{$data}); |
298 |
} |
} |
299 |
|
|
303 |
} |
} |
304 |
|
|
305 |
|
|
|
=head2 _s |
|
306 |
|
|
307 |
Remove multiple whitespaces from string, as well as whitespaces at beginning or end |
package Search::Estraier::Condition; |
308 |
|
|
309 |
my $text = _s(" this is a text "); |
use Carp qw/confess croak/; |
310 |
$text = 'this is a text'; |
|
311 |
|
use Search::Estraier; |
312 |
|
our @ISA = qw/Search::Estraier/; |
313 |
|
|
314 |
|
=head1 Search::Estraier::Condition |
315 |
|
|
316 |
|
=head2 new |
317 |
|
|
318 |
|
my $cond = new Search::HyperEstraier::Condition; |
319 |
|
|
320 |
=cut |
=cut |
321 |
|
|
322 |
sub _s { |
sub new { |
323 |
my $text = shift || return; |
my $class = shift; |
324 |
$text =~ s/\s\s+/ /gs; |
my $self = {}; |
325 |
$text =~ s/^\s+//; |
bless($self, $class); |
326 |
$text =~ s/\s+$//; |
|
327 |
return $text; |
$self->{max} = -1; |
328 |
|
$self->{options} = 0; |
329 |
|
|
330 |
|
$self ? return $self : return undef; |
331 |
|
} |
332 |
|
|
333 |
|
=head2 set_phrase |
334 |
|
|
335 |
|
$cond->set_phrase('search phrase'); |
336 |
|
|
337 |
|
=cut |
338 |
|
|
339 |
|
sub set_phrase { |
340 |
|
my $self = shift; |
341 |
|
$self->{phrase} = $self->_s( shift ); |
342 |
} |
} |
343 |
|
|
344 |
|
=head2 add_attr |
345 |
|
|
346 |
|
$cond->add_attr('@URI STRINC /~dpavlin/'); |
347 |
|
|
348 |
|
=cut |
349 |
|
|
350 |
|
sub add_attr { |
351 |
|
my $self = shift; |
352 |
|
my $attr = shift || return; |
353 |
|
push @{ $self->{attrs} }, $self->_s( $attr ); |
354 |
|
} |
355 |
|
|
356 |
|
=head2 set_order |
357 |
|
|
358 |
|
$cond->set_order('@mdate NUMD'); |
359 |
|
|
360 |
|
=cut |
361 |
|
|
362 |
|
sub set_order { |
363 |
|
my $self = shift; |
364 |
|
$self->{order} = shift; |
365 |
|
} |
366 |
|
|
367 |
|
=head2 set_max |
368 |
|
|
369 |
|
$cond->set_max(42); |
370 |
|
|
371 |
|
=cut |
372 |
|
|
373 |
|
sub set_max { |
374 |
|
my $self = shift; |
375 |
|
my $max = shift; |
376 |
|
croak "set_max needs number" unless ($max =~ m/^\d+$/); |
377 |
|
$self->{max} = $max; |
378 |
|
} |
379 |
|
|
380 |
|
=head2 set_options |
381 |
|
|
382 |
|
$cond->set_options( SURE => 1 ); |
383 |
|
|
384 |
|
=cut |
385 |
|
|
386 |
|
my $options = { |
387 |
|
# check N-gram keys skipping by three |
388 |
|
SURE => 1 << 0, |
389 |
|
# check N-gram keys skipping by two |
390 |
|
USUAL => 1 << 1, |
391 |
|
# without TF-IDF tuning |
392 |
|
FAST => 1 << 2, |
393 |
|
# with the simplified phrase |
394 |
|
AGITO => 1 << 3, |
395 |
|
# check every N-gram key |
396 |
|
NOIDF => 1 << 4, |
397 |
|
# check N-gram keys skipping by one |
398 |
|
SIMPLE => 1 << 10, |
399 |
|
}; |
400 |
|
|
401 |
|
sub set_options { |
402 |
|
my $self = shift; |
403 |
|
my $option = shift; |
404 |
|
confess "unknown option" unless ($options->{$option}); |
405 |
|
$self->{options} ||= $options->{$option}; |
406 |
|
} |
407 |
|
|
408 |
|
=head2 phrase |
409 |
|
|
410 |
|
Return search phrase. |
411 |
|
|
412 |
|
print $cond->phrase; |
413 |
|
|
414 |
|
=cut |
415 |
|
|
416 |
|
sub phrase { |
417 |
|
my $self = shift; |
418 |
|
return $self->{phrase}; |
419 |
|
} |
420 |
|
|
421 |
|
=head2 order |
422 |
|
|
423 |
|
Return search result order. |
424 |
|
|
425 |
|
print $cond->order; |
426 |
|
|
427 |
|
=cut |
428 |
|
|
429 |
|
sub order { |
430 |
|
my $self = shift; |
431 |
|
return $self->{order}; |
432 |
|
} |
433 |
|
|
434 |
|
=head2 attrs |
435 |
|
|
436 |
|
Return search result attrs. |
437 |
|
|
438 |
|
my @cond_attrs = $cond->attrs; |
439 |
|
|
440 |
|
=cut |
441 |
|
|
442 |
|
sub attrs { |
443 |
|
my $self = shift; |
444 |
|
#croak "attrs return array, not scalar" if (! wantarray); |
445 |
|
return @{ $self->{attrs} }; |
446 |
|
} |
447 |
|
|
448 |
|
=head2 max |
449 |
|
|
450 |
|
Return maximum number of results. |
451 |
|
|
452 |
|
print $cond->max; |
453 |
|
|
454 |
|
C<-1> is returned for unitialized value, C<0> is unlimited. |
455 |
|
|
456 |
|
=cut |
457 |
|
|
458 |
|
sub max { |
459 |
|
my $self = shift; |
460 |
|
return $self->{max}; |
461 |
|
} |
462 |
|
|
463 |
|
=head2 options |
464 |
|
|
465 |
|
Return options for this condition. |
466 |
|
|
467 |
|
print $cond->options; |
468 |
|
|
469 |
|
Options are returned in numerical form. |
470 |
|
|
471 |
|
=cut |
472 |
|
|
473 |
|
sub options { |
474 |
|
my $self = shift; |
475 |
|
return $self->{options}; |
476 |
|
} |
477 |
|
|
478 |
|
|
479 |
package Search::Estraier::Master; |
package Search::Estraier::Master; |
488 |
|
|
489 |
{ |
{ |
490 |
package RequestAgent; |
package RequestAgent; |
491 |
@ISA = qw(LWP::UserAgent); |
our @ISA = qw(LWP::UserAgent); |
492 |
|
|
493 |
sub new { |
sub new { |
494 |
my $self = LWP::UserAgent::new(@_); |
my $self = LWP::UserAgent::new(@_); |
549 |
|
|
550 |
=head1 COPYRIGHT AND LICENSE |
=head1 COPYRIGHT AND LICENSE |
551 |
|
|
552 |
Copyright (C) 2005 by Dobrica Pavlinusic |
Copyright (C) 2005-2006 by Dobrica Pavlinusic |
553 |
|
|
554 |
This library is free software; you can redistribute it and/or modify |
This library is free software; you can redistribute it and/or modify |
555 |
it under the GPL v2 or later. |
it under the GPL v2 or later. |