line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
package Algorithm::LSH::Hash::Hamming; |
2
|
2
|
|
|
2
|
|
958
|
use strict; |
|
2
|
|
|
|
|
4
|
|
|
2
|
|
|
|
|
79
|
|
3
|
2
|
|
|
2
|
|
11
|
use warnings; |
|
2
|
|
|
|
|
4
|
|
|
2
|
|
|
|
|
64
|
|
4
|
2
|
|
|
2
|
|
10
|
use List::Util qw(max); |
|
2
|
|
|
|
|
3
|
|
|
2
|
|
|
|
|
247
|
|
5
|
2
|
|
|
2
|
|
14
|
use Carp; |
|
2
|
|
|
|
|
2
|
|
|
2
|
|
|
|
|
183
|
|
6
|
2
|
|
|
2
|
|
12
|
use base qw(Algorithm::LSH::Hash); |
|
2
|
|
|
|
|
4
|
|
|
2
|
|
|
|
|
1630
|
|
7
|
|
|
|
|
|
|
|
8
|
|
|
|
|
|
|
__PACKAGE__->mk_accessors($_) for qw( _indexes ); |
9
|
|
|
|
|
|
|
|
10
|
|
|
|
|
|
|
our $SCALE = 100; |
11
|
|
|
|
|
|
|
|
12
|
|
|
|
|
|
|
sub do_hashing { |
13
|
0
|
|
|
0
|
1
|
|
my $self = shift; |
14
|
0
|
|
|
|
|
|
my $vector = shift; |
15
|
0
|
0
|
|
|
|
|
if ( ref $vector ne 'ARRAY' ) { |
16
|
0
|
0
|
|
|
|
|
carp("args should be an array_ref") and return; |
17
|
|
|
|
|
|
|
} |
18
|
0
|
|
|
0
|
|
|
my $d = $self->d || sub { $self->d( int @$vector ) } |
19
|
0
|
|
0
|
|
|
|
->(); |
20
|
0
|
0
|
|
|
|
|
if ( $d != scalar @$vector ) { |
21
|
0
|
0
|
|
|
|
|
carp("invalid dimention number") and return; |
22
|
|
|
|
|
|
|
} |
23
|
0
|
|
|
|
|
|
my $L = $self->L; |
24
|
0
|
|
|
|
|
|
my $unary_code = $self->_unarize($vector); |
25
|
0
|
|
|
|
|
|
my @hashes; |
26
|
0
|
|
|
|
|
|
for my $i ( 0 .. $L - 1 ) { |
27
|
0
|
|
|
|
|
|
my @array; |
28
|
0
|
|
|
|
|
|
for my $j ( @{ $self->_pickuped_indexes->[$i] } ) { |
|
0
|
|
|
|
|
|
|
29
|
0
|
|
|
|
|
|
push( @array, $unary_code->[$j] ); |
30
|
|
|
|
|
|
|
} |
31
|
0
|
|
|
|
|
|
push( @hashes, \@array ); |
32
|
|
|
|
|
|
|
} |
33
|
0
|
|
|
|
|
|
return \@hashes; |
34
|
|
|
|
|
|
|
} |
35
|
|
|
|
|
|
|
|
36
|
|
|
|
|
|
|
sub _pickuped_indexes { |
37
|
0
|
|
|
0
|
|
|
my $self = shift; |
38
|
|
|
|
|
|
|
$self->_indexes or sub { |
39
|
0
|
|
|
0
|
|
|
my @indexes; |
40
|
0
|
|
|
|
|
|
my $L = $self->L; |
41
|
0
|
|
|
|
|
|
for my $i ( 0 .. $L - 1 ) { |
42
|
0
|
|
|
|
|
|
my %seen; |
43
|
0
|
|
|
|
|
|
while (1) { |
44
|
0
|
|
|
|
|
|
my $rand = int( rand( $self->d * $SCALE ) ); |
45
|
0
|
0
|
|
|
|
|
if ( !$seen{$rand} ) { |
46
|
0
|
|
|
|
|
|
$seen{$rand} = 1; |
47
|
0
|
0
|
|
|
|
|
last if keys %seen == $self->k * $SCALE; |
48
|
|
|
|
|
|
|
} |
49
|
|
|
|
|
|
|
} |
50
|
0
|
|
|
|
|
|
push( @indexes, [ sort { $a <=> $b } keys %seen ] ); |
|
0
|
|
|
|
|
|
|
51
|
|
|
|
|
|
|
} |
52
|
0
|
|
|
|
|
|
$self->_indexes( \@indexes ); |
53
|
|
|
|
|
|
|
} |
54
|
0
|
0
|
|
|
|
|
->(); |
55
|
|
|
|
|
|
|
} |
56
|
|
|
|
|
|
|
|
57
|
|
|
|
|
|
|
sub _unarize { |
58
|
0
|
|
|
0
|
|
|
my $self = shift; |
59
|
0
|
|
|
|
|
|
my $vector = shift; |
60
|
0
|
|
|
|
|
|
my $max = max(@$vector); |
61
|
0
|
|
|
|
|
|
my $n = $SCALE / $max; |
62
|
0
|
|
|
|
|
|
my @unary; |
63
|
0
|
|
|
|
|
|
for (@$vector) { |
64
|
0
|
|
|
|
|
|
my $i = int( $_ * $n ); |
65
|
0
|
|
|
|
|
|
my $j = $SCALE - $i; |
66
|
0
|
|
|
|
|
|
for ( 1 .. $i ) { |
67
|
0
|
|
|
|
|
|
push( @unary, 1 ); |
68
|
|
|
|
|
|
|
} |
69
|
0
|
|
|
|
|
|
for ( 1 .. $j ) { |
70
|
0
|
|
|
|
|
|
push( @unary, 0 ); |
71
|
|
|
|
|
|
|
} |
72
|
|
|
|
|
|
|
} |
73
|
0
|
|
|
|
|
|
return \@unary; |
74
|
|
|
|
|
|
|
} |
75
|
|
|
|
|
|
|
|
76
|
|
|
|
|
|
|
1; |
77
|
|
|
|
|
|
|
__END__ |