Clusterize.pm

filter: Statementstmt 25.0% 12 / 48 Branchbran 0.0% 0 / 10 Conditioncond 0.0% 0 / 2 MC/DCmcdc 0.0% 0 / 1 Subroutinesub 30.7% 4 / 13 Podpod 40.0% 4 / 10 totaltotal 23.8% 20 / 84 CC 10 SCAR 21.5
CC
10
Cov
20%
add_pair
20.0
remove_pair
12.0
list
12.0
CRAP 8.5
↑ Summary Clusterize/Pattern.pm »
1 package Clusterize;
2 3 ▶use warnings;
Subroutine BEGIN: called CC=1 SCAR=0.0
3 3 ▶use strict;
Subroutine BEGIN: called CC=1 SCAR=0.0
4 3 ▶use Clusterize::Pattern;
Subroutine BEGIN: called CC=1 SCAR=0.0
5 3 our $VERSION = '0.02';
6  
7 0 ▶sub pair {
Subroutine pair: not called CC=2 SCAR=17.9
Pod undocumented
8 0 my ($self, $key, $val) = @_;
9 0 ▶ $self->{pairs}{$key} = $val if defined $val;
Branch: 1 branch
BranchTrueFalseTotal
if defined $val 0 0 0
10 0 $self->{pairs}{$key};
11 }
12  
13 0 ▶sub delete_pair {
Subroutine delete_pair: not called CC=1 SCAR=6.9
Pod undocumented
14 0 my ($self, $key) = @_;
15 0 delete $self->{pairs}{$key};
16 }
17  
18 0 ▶sub cluster_list { keys %{shift->{clusters}} }
Subroutine cluster_list: not called CC=1 SCAR=6.9
Pod undocumented
19 0 ▶sub remove_cluster_pair {
Subroutine remove_cluster_pair: not called CC=2 SCAR=17.9
Pod undocumented
20 0 my ($self, $digest, $key) = @_;
21 0 delete $self->{clusters}{$digest}{$key};
22 delete $self->{clusters}{$digest}
23 0 ▶ unless %{$self->{clusters}{$digest}};
Branch: 1 branch
BranchTrueFalseTotal
unless %{$self->{"clusters"}{$digest};} 0 0 0
24 }
25  
26 0 ▶sub add_cluster_pair {
Subroutine add_cluster_pair: not called CC=1 SCAR=6.9
Pod undocumented
27 0 my ($self, $digest, $pair) = @_;
28 0 $self->{clusters}{$digest}{$pair->{key}} = $pair->{val};
29 }
30  
31 0 ▶sub cluster_pairs {
Subroutine cluster_pairs: not called CC=1 SCAR=6.9
Pod undocumented
32 0 my ($self, $digest) = @_;
33 0 return $self->{clusters}{$digest};
34 }
35  
36 1 ▶sub new { return bless {}, shift }
Subroutine new: called CC=1 SCAR=0.0
Pod documented
37 0 ▶sub add_pair {
Subroutine add_pair: not called CC=4 SCAR=30.0
Pod documented
38 0 my ($self, $key, $digest) = @_;
39 0 ▶ return if $self->pair($key);
Branch: 1 branch
BranchTrueFalseTotal
if $self->pair($key) 0 0 0
40 0 ▶ $digest = Clusterize::Pattern->text2digest($digest)
Branch: 1 branch
BranchTrueFalseTotal
if ref $digest eq "ARRAY" 0 0 0
41 if ref $digest eq 'ARRAY';
42 0 $self->pair($key, $digest);
43 0 for (keys %{$digest}) {
44 0 $self->add_cluster_pair($_, {key => $key, val => $digest->{$_}});
45 }
46 }
47  
48 0 ▶sub remove_pair {
Subroutine remove_pair: not called CC=3 SCAR=24.8
Pod documented
49 0 my ($self, $key) = @_;
50 0 ▶ my $cluster_pair = $self->pair($key) || return;
Condition0%0/2
$self->pair($key) || (return)
l!l
00
MC/DC0%0/1
$self->pair($key) || (return)
$self->pair($key)
Truth table0%0/2
A
A= $self->pair($key)
Aresult
11
00
51 0 for (keys %{$cluster_pair}) {$self->remove_cluster_pair($_, $key)}
52 0 $self->delete_pair($key);
53 }
54  
55 0 ▶sub list {
Subroutine list: not called CC=3 SCAR=24.8
Pod documented
56 0 my ($self, $opt) = @_;
57 0 my (%md5, @clusters);
58 0 for (map { Clusterize::Pattern->new($self->cluster_pairs($_)) }
59 $self->cluster_list($opt)) {
60 0 ▶ next if $md5{$_->digest};
Branch: 1 branch
BranchTrueFalseTotal
if $md5{$_->digest} 0 0 0
61 0 $md5{$_->digest} = 1;
62 0 push @clusters, $_
63 }
64 0 @clusters;
65 }
66  
67 3 1;
68  
69 =head1 NAME
70  
71 Clusterize - clustering text documents.
72  
73 =head1 VERSION
74  
75 Version 0.02
76  
77 =head1 SYNOPSIS
78  
79 use Clusterize;
80  
81 my %pairs = (
82 key1 => [ string1, string2, ...stringN ],
83 key2 => [ string5, string6, ...stringM ],
84 ...
85 keyN => [ ... ],
86 );
87  
88 my $clusterize = Clusterize->new();
89 while (my @pair = each %files) { $clusterize->add_pair(@pair) }
90  
91 foreach my $c ( $clusterize->list ) {
92 printf "# /%s/ (digest=%s) (accuracy=%.3f) (size=%d)",
93 $c->pattern, $digest, $c->accuracy, $c->size;
94 my $pairs = $c->pairs;
95 for ( keys %{$pairs} ) { print $_." ".$pairs->{$_} }
96 }
97  
98  
99 =head1 DESCRIPTION
100  
101 B<Clusterize> module implements specific algorithm for clustering text documents.
102  
103 =head1 PUBLIC METHODS
104  
105 =head2 new
106  
107 This is the constructor. No parameter is required.
108  
109 =head2 add_pair
110  
111 This method is used to add new document into cluster set:
112  
113 $clusterize->add_pair($key, [$string1, $string2, ...]);
114  
115 $key - is uniq name of the document (e.g. filename),
116 [$string1, $string2, ...] - text of the document.
117  
118 =head2 remove_pair
119  
120 This method is used to remove document from cluster set:
121  
122 $clusterize->remove_pair($key);
123  
124 $key - is name of the document (e.g. filename).
125  
126 =head2 list
127  
128 This method is used to get list of built clusters:
129  
130 my @clusters = $clusterize->list();
131  
132 Returns list of B<Clusterize::Pattern> objects with the following attributes:
133  
134 $c->pattern - regexp that matches all strings in the given cluster;
135  
136 $c->accuracy - this value reflects how similar strings in the cluster (value from 0 to 1);
137  
138 $c->size - how many documents in the cluster;
139  
140 $c->digest - MD5 digest of the cluster to identify duplicate clusters;
141  
142 $c->pairs - list of { key => $key1, val => $val1 } hash pairs, where:
143 key - is name of document, val - is string from 'key' document;
144  
145 =head1 AUTHOR
146  
147 Slava Moiseev, <slava.moiseev@yahoo.com>
148  
↑ Summary Clusterize/Pattern.pm »