| line | stmt | bran | cond | sub | pod | time | code | 
| 1 |  |  |  |  |  |  | package Text::Fingerprint; | 
| 2 |  |  |  |  |  |  | # ABSTRACT: perform simple text clustering by key collision | 
| 3 |  |  |  |  |  |  |  | 
| 4 |  |  |  |  |  |  |  | 
| 5 | 2 |  |  | 2 |  | 53580 | use strict; | 
|  | 2 |  |  |  |  | 5 |  | 
|  | 2 |  |  |  |  | 72 |  | 
| 6 | 2 |  |  | 2 |  | 10 | use utf8; | 
|  | 2 |  |  |  |  | 5 |  | 
|  | 2 |  |  |  |  | 18 |  | 
| 7 | 2 |  |  | 2 |  | 48 | use warnings qw(all); | 
|  | 2 |  |  |  |  | 3 |  | 
|  | 2 |  |  |  |  | 78 |  | 
| 8 |  |  |  |  |  |  |  | 
| 9 | 2 |  |  | 2 |  | 9 | use base q(Exporter); | 
|  | 2 |  |  |  |  | 3 |  | 
|  | 2 |  |  |  |  | 436 |  | 
| 10 |  |  |  |  |  |  |  | 
| 11 |  |  |  |  |  |  | our %EXPORT_TAGS    = (all => [qw(fingerprint fingerprint_ngram)]); | 
| 12 |  |  |  |  |  |  | our @EXPORT_OK      = (@{$EXPORT_TAGS{all}}); | 
| 13 |  |  |  |  |  |  | our @EXPORT         = qw(); | 
| 14 |  |  |  |  |  |  |  | 
| 15 | 2 |  |  | 2 |  | 1754 | use List::MoreUtils qw(uniq); | 
|  | 2 |  |  |  |  | 2688 |  | 
|  | 2 |  |  |  |  | 197 |  | 
| 16 | 2 |  |  | 2 |  | 2112 | use Text::Unidecode; | 
|  | 2 |  |  |  |  | 5151 |  | 
|  | 2 |  |  |  |  | 713 |  | 
| 17 |  |  |  |  |  |  |  | 
| 18 |  |  |  |  |  |  | our $VERSION = '0.006'; # VERSION | 
| 19 |  |  |  |  |  |  |  | 
| 20 |  |  |  |  |  |  |  | 
| 21 |  |  |  |  |  |  | my $NON_WORD = qr{ [\W_]+ }x; | 
| 22 |  |  |  |  |  |  |  | 
| 23 |  |  |  |  |  |  | sub fingerprint ($) { | 
| 24 | 1 |  |  | 1 | 1 | 9 | my ($string) = @_; | 
| 25 |  |  |  |  |  |  |  | 
| 26 | 1 |  |  |  |  | 6 | $string = lc unidecode $string; | 
| 27 | 1 |  |  |  |  | 3871 | $string =~ s{^ $NON_WORD | $NON_WORD $}{}gosx; | 
| 28 |  |  |  |  |  |  |  | 
| 29 | 1 |  |  |  |  | 1174 | return join q( ) => | 
| 30 |  |  |  |  |  |  | sort( | 
| 31 |  |  |  |  |  |  | uniq( | 
| 32 |  |  |  |  |  |  | split( | 
| 33 |  |  |  |  |  |  | m{ $NON_WORD }ox, | 
| 34 |  |  |  |  |  |  | $string | 
| 35 |  |  |  |  |  |  | ) | 
| 36 |  |  |  |  |  |  | ) | 
| 37 |  |  |  |  |  |  | ); | 
| 38 |  |  |  |  |  |  | } | 
| 39 |  |  |  |  |  |  |  | 
| 40 |  |  |  |  |  |  |  | 
| 41 |  |  |  |  |  |  | sub fingerprint_ngram ($;$) { | 
| 42 | 4 |  |  | 4 | 1 | 459 | my ($string, $n) = (@_, 2); | 
| 43 |  |  |  |  |  |  |  | 
| 44 | 4 |  |  |  |  | 13 | $string = lc unidecode $string; | 
| 45 | 4 |  |  |  |  | 358 | $string =~ s{ $NON_WORD }{}gosx; | 
| 46 |  |  |  |  |  |  |  | 
| 47 | 4 |  |  |  |  | 507 | return join '' => | 
| 48 |  |  |  |  |  |  | sort( | 
| 49 |  |  |  |  |  |  | uniq( | 
| 50 |  |  |  |  |  |  | $string =~ m{ | 
| 51 |  |  |  |  |  |  | (?= | 
| 52 |  |  |  |  |  |  | (.{$n}) | 
| 53 |  |  |  |  |  |  | ) | 
| 54 |  |  |  |  |  |  | }gx | 
| 55 |  |  |  |  |  |  | ) | 
| 56 |  |  |  |  |  |  | ); | 
| 57 |  |  |  |  |  |  | } | 
| 58 |  |  |  |  |  |  |  | 
| 59 |  |  |  |  |  |  |  | 
| 60 |  |  |  |  |  |  | 1; | 
| 61 |  |  |  |  |  |  |  | 
| 62 |  |  |  |  |  |  | __END__ |