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/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 |
|
|
120 |
|
|
121 |
$doc->add_attr( name => 'value' ); |
$doc->add_attr( name => 'value' ); |
122 |
|
|
123 |
B<FIXME>: delete attribute using |
Delete attribute using |
124 |
|
|
125 |
$doc->add_attr( name => undef ); |
$doc->add_attr( name => undef ); |
126 |
|
|
131 |
my $attrs = {@_}; |
my $attrs = {@_}; |
132 |
|
|
133 |
while (my ($name, $value) = each %{ $attrs }) { |
while (my ($name, $value) = each %{ $attrs }) { |
134 |
$self->{attrs}->{_s($name)} = _s($value); |
if (! defined($value)) { |
135 |
|
delete( $self->{attrs}->{ $self->_s($name) } ); |
136 |
|
} else { |
137 |
|
$self->{attrs}->{ $self->_s($name) } = $self->_s($value); |
138 |
|
} |
139 |
} |
} |
140 |
|
|
141 |
return 1; |
return 1; |
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 |
190 |
|
|
191 |
=head2 attr_names |
=head2 attr_names |
192 |
|
|
193 |
Get a list of attribute names of a document object. |
Returns array with attribute names from document object. |
194 |
|
|
195 |
my @attrs = $doc->attr_names; |
my @attrs = $doc->attr_names; |
196 |
|
|
198 |
|
|
199 |
sub attr_names { |
sub attr_names { |
200 |
my $self = shift; |
my $self = shift; |
201 |
confess "attr_names return array, not scalar" if (! wantarray); |
croak "attr_names return array, not scalar" if (! wantarray); |
202 |
return sort keys %{ $self->{attrs} }; |
return sort keys %{ $self->{attrs} }; |
203 |
} |
} |
204 |
|
|
205 |
|
|
206 |
=head2 attr |
=head2 attr |
207 |
|
|
208 |
Get the value of an attribute. |
Returns value of an attribute. |
209 |
|
|
210 |
my $value = $doc->attr( 'attribute' ); |
my $value = $doc->attr( 'attribute' ); |
211 |
|
|
218 |
return $self->{'attrs'}->{ $name }; |
return $self->{'attrs'}->{ $name }; |
219 |
} |
} |
220 |
|
|
221 |
|
|
222 |
|
=head2 texts |
223 |
|
|
224 |
|
Returns array with text sentences. |
225 |
|
|
226 |
|
my @texts = $doc->texts; |
227 |
|
|
228 |
|
=cut |
229 |
|
|
230 |
|
sub texts { |
231 |
|
my $self = shift; |
232 |
|
confess "texts return array, not scalar" if (! wantarray); |
233 |
|
return @{ $self->{dtexts} }; |
234 |
|
} |
235 |
|
|
236 |
|
=head2 cat_texts |
237 |
|
|
238 |
|
Return whole text as single scalar. |
239 |
|
|
240 |
|
my $text = $doc->cat_texts; |
241 |
|
|
242 |
|
=cut |
243 |
|
|
244 |
|
sub cat_texts { |
245 |
|
my $self = shift; |
246 |
|
return join(' ',@{ $self->{dtexts} }); |
247 |
|
} |
248 |
|
|
249 |
=head2 dump_draft |
=head2 dump_draft |
250 |
|
|
251 |
|
Dump draft data from document object. |
252 |
|
|
253 |
print $doc->dump_draft; |
print $doc->dump_draft; |
254 |
|
|
255 |
=cut |
=cut |
256 |
|
|
257 |
sub dump_draft { |
sub dump_draft { |
258 |
|
my $self = shift; |
259 |
|
my $draft; |
260 |
|
|
261 |
|
foreach my $attr_name (sort keys %{ $self->{attrs} }) { |
262 |
|
$draft .= $attr_name . '=' . $self->{attrs}->{$attr_name} . "\n"; |
263 |
|
} |
264 |
|
|
265 |
|
if ($self->{kwords}) { |
266 |
|
$draft .= '%%VECTOR'; |
267 |
|
while (my ($key, $value) = each %{ $self->{kwords} }) { |
268 |
|
$draft .= "\t$key\t$value"; |
269 |
|
} |
270 |
|
$draft .= "\n"; |
271 |
|
} |
272 |
|
|
273 |
|
$draft .= "\n"; |
274 |
|
|
275 |
|
$draft .= join("\n", @{ $self->{dtexts} }) . "\n"; |
276 |
|
$draft .= "\t" . join("\n\t", @{ $self->{htexts} }) . "\n"; |
277 |
|
|
278 |
|
return $draft; |
279 |
} |
} |
280 |
|
|
281 |
=head2 delete |
=head2 delete |
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 |
|
|
300 |
|
$self->{id} = -1; |
301 |
|
|
302 |
return 1; |
return 1; |
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. |