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 $text; |
45 |
|
} |
46 |
|
|
47 |
package Search::Estraier::Document; |
package Search::Estraier::Document; |
48 |
|
|
49 |
use Carp qw/croak confess/; |
use Carp qw/croak confess/; |
50 |
|
|
51 |
|
use Search::Estraier; |
52 |
|
our @ISA = qw/Search::Estraier/; |
53 |
|
|
54 |
=head1 Search::Estraier::Document |
=head1 Search::Estraier::Document |
55 |
|
|
56 |
This class implements Document which is collection of attributes |
This class implements Document which is collection of attributes |
57 |
(key=value), vectors (also key value) display text and hidden text. |
(key=value), vectors (also key value) display text and hidden text. |
58 |
|
|
|
Document for HyperEstraier |
|
|
|
|
59 |
=head2 new |
=head2 new |
60 |
|
|
61 |
Create new document, empty or from draft. |
Create new document, empty or from draft. |
132 |
|
|
133 |
while (my ($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 { |
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 Search::Estraier; |
310 |
$text = 'this is a text'; |
our @ISA = qw/Search::Estraier/; |
311 |
|
|
312 |
|
my $options = { |
313 |
|
# check N-gram keys skipping by three |
314 |
|
SURE => 1 << 0, |
315 |
|
# check N-gram keys skipping by two |
316 |
|
USUAL => 1 << 1, |
317 |
|
# without TF-IDF tuning |
318 |
|
FAST => 1 << 2, |
319 |
|
# with the simplified phrase |
320 |
|
AGITO => 1 << 3, |
321 |
|
# check every N-gram key |
322 |
|
NOIDF => 1 << 4, |
323 |
|
# check N-gram keys skipping by one |
324 |
|
SIMPLE => 1 << 10, |
325 |
|
}; |
326 |
|
|
327 |
|
=head1 Search::Estraier::Condition |
328 |
|
|
329 |
|
=head2 new |
330 |
|
|
331 |
|
my $cond = new Search::HyperEstraier::Condition; |
332 |
|
|
333 |
=cut |
=cut |
334 |
|
|
335 |
sub _s { |
sub new { |
336 |
my $text = shift || return; |
my $class = shift; |
337 |
$text =~ s/\s\s+/ /gs; |
my $self = {}; |
338 |
$text =~ s/^\s+//; |
bless($self, $class); |
|
$text =~ s/\s+$//; |
|
|
return $text; |
|
|
} |
|
339 |
|
|
340 |
|
$self ? return $self : return undef; |
341 |
|
} |
342 |
|
|
343 |
|
|
344 |
package Search::Estraier::Master; |
package Search::Estraier::Master; |
353 |
|
|
354 |
{ |
{ |
355 |
package RequestAgent; |
package RequestAgent; |
356 |
@ISA = qw(LWP::UserAgent); |
our @ISA = qw(LWP::UserAgent); |
357 |
|
|
358 |
sub new { |
sub new { |
359 |
my $self = LWP::UserAgent::new(@_); |
my $self = LWP::UserAgent::new(@_); |
414 |
|
|
415 |
=head1 COPYRIGHT AND LICENSE |
=head1 COPYRIGHT AND LICENSE |
416 |
|
|
417 |
Copyright (C) 2005 by Dobrica Pavlinusic |
Copyright (C) 2005-2006 by Dobrica Pavlinusic |
418 |
|
|
419 |
This library is free software; you can redistribute it and/or modify |
This library is free software; you can redistribute it and/or modify |
420 |
it under the GPL v2 or later. |
it under the GPL v2 or later. |