| line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
|
1
|
|
|
|
|
|
|
package Test::Text::Sentence; |
|
2
|
|
|
|
|
|
|
|
|
3
|
|
|
|
|
|
|
#============================================================================== |
|
4
|
|
|
|
|
|
|
# |
|
5
|
|
|
|
|
|
|
# Start of POD |
|
6
|
|
|
|
|
|
|
# |
|
7
|
|
|
|
|
|
|
#============================================================================== |
|
8
|
|
|
|
|
|
|
|
|
9
|
|
|
|
|
|
|
=head1 NAME |
|
10
|
|
|
|
|
|
|
|
|
11
|
|
|
|
|
|
|
Test::Text::Sentence - module for splitting text into sentences |
|
12
|
|
|
|
|
|
|
|
|
13
|
|
|
|
|
|
|
=head1 SYNOPSIS |
|
14
|
|
|
|
|
|
|
|
|
15
|
|
|
|
|
|
|
use Test::Text::Sentence qw( split_sentences ); |
|
16
|
|
|
|
|
|
|
use locale; |
|
17
|
|
|
|
|
|
|
use POSIX qw( locale_h ); |
|
18
|
|
|
|
|
|
|
|
|
19
|
|
|
|
|
|
|
setlocale( LC_CTYPE, 'iso_8859_1' ); |
|
20
|
|
|
|
|
|
|
@sentences = split_sentences( $text ); |
|
21
|
|
|
|
|
|
|
|
|
22
|
|
|
|
|
|
|
=head1 DESCRIPTION |
|
23
|
|
|
|
|
|
|
|
|
24
|
|
|
|
|
|
|
The C module contains the function split_sentences, which |
|
25
|
|
|
|
|
|
|
splits text into its constituent sentences, based on a fairly approximate |
|
26
|
|
|
|
|
|
|
regex. If you set the locale before calling it, it will deal correctly with |
|
27
|
|
|
|
|
|
|
locale dependant capitalization to identify sentence boundaries. Certain well |
|
28
|
|
|
|
|
|
|
know exceptions, such as abbreviations, may cause incorrect segmentations. |
|
29
|
|
|
|
|
|
|
|
|
30
|
|
|
|
|
|
|
=head1 FUNCTIONS |
|
31
|
|
|
|
|
|
|
|
|
32
|
|
|
|
|
|
|
=head2 split_sentences( $text ) |
|
33
|
|
|
|
|
|
|
|
|
34
|
|
|
|
|
|
|
The split sentences function takes a scalar containing ascii text as an |
|
35
|
|
|
|
|
|
|
argument and returns an array of sentences that the text has been split into. |
|
36
|
|
|
|
|
|
|
|
|
37
|
|
|
|
|
|
|
@sentences = split_sentences( $text ); |
|
38
|
|
|
|
|
|
|
|
|
39
|
|
|
|
|
|
|
=head1 SEE ALSO |
|
40
|
|
|
|
|
|
|
|
|
41
|
|
|
|
|
|
|
locale |
|
42
|
|
|
|
|
|
|
POSIX |
|
43
|
|
|
|
|
|
|
|
|
44
|
|
|
|
|
|
|
=head1 REPOSITORY |
|
45
|
|
|
|
|
|
|
|
|
46
|
|
|
|
|
|
|
L |
|
47
|
|
|
|
|
|
|
|
|
48
|
|
|
|
|
|
|
=head1 AUTHOR |
|
49
|
|
|
|
|
|
|
|
|
50
|
|
|
|
|
|
|
Ave Wrigley Ewrigley@cre.canon.co.ukE |
|
51
|
|
|
|
|
|
|
|
|
52
|
|
|
|
|
|
|
=head1 COPYRIGHT |
|
53
|
|
|
|
|
|
|
|
|
54
|
|
|
|
|
|
|
Copyright (c) 1997 Canon Research Centre Europe (CRE). All rights reserved. |
|
55
|
|
|
|
|
|
|
|
|
56
|
|
|
|
|
|
|
This is free software; you can redistribute it and/or modify it under |
|
57
|
|
|
|
|
|
|
the same terms as the Perl 5 programming language system itself. |
|
58
|
|
|
|
|
|
|
|
|
59
|
|
|
|
|
|
|
=cut |
|
60
|
|
|
|
|
|
|
|
|
61
|
|
|
|
|
|
|
#============================================================================== |
|
62
|
|
|
|
|
|
|
# |
|
63
|
|
|
|
|
|
|
# End of POD |
|
64
|
|
|
|
|
|
|
# |
|
65
|
|
|
|
|
|
|
#============================================================================== |
|
66
|
|
|
|
|
|
|
|
|
67
|
|
|
|
|
|
|
#============================================================================== |
|
68
|
|
|
|
|
|
|
# |
|
69
|
|
|
|
|
|
|
# Pragmas |
|
70
|
|
|
|
|
|
|
# |
|
71
|
|
|
|
|
|
|
#============================================================================== |
|
72
|
|
|
|
|
|
|
|
|
73
|
|
|
|
|
|
|
require 5.006; |
|
74
|
4
|
|
|
4
|
|
94809
|
use strict; |
|
|
4
|
|
|
|
|
15
|
|
|
|
4
|
|
|
|
|
110
|
|
|
75
|
4
|
|
|
4
|
|
18
|
use warnings; |
|
|
4
|
|
|
|
|
7
|
|
|
|
4
|
|
|
|
|
142
|
|
|
76
|
|
|
|
|
|
|
|
|
77
|
|
|
|
|
|
|
#============================================================================== |
|
78
|
|
|
|
|
|
|
# |
|
79
|
|
|
|
|
|
|
# Modules |
|
80
|
|
|
|
|
|
|
# |
|
81
|
|
|
|
|
|
|
#============================================================================== |
|
82
|
|
|
|
|
|
|
|
|
83
|
|
|
|
|
|
|
require Exporter; |
|
84
|
|
|
|
|
|
|
|
|
85
|
|
|
|
|
|
|
#============================================================================== |
|
86
|
|
|
|
|
|
|
# |
|
87
|
|
|
|
|
|
|
# Public globals |
|
88
|
|
|
|
|
|
|
# |
|
89
|
|
|
|
|
|
|
#============================================================================== |
|
90
|
|
|
|
|
|
|
|
|
91
|
4
|
|
|
4
|
|
18
|
use vars qw( @ISA @EXPORT_OK @PUNCTUATION ); |
|
|
4
|
|
|
|
|
7
|
|
|
|
4
|
|
|
|
|
1140
|
|
|
92
|
|
|
|
|
|
|
|
|
93
|
|
|
|
|
|
|
@ISA = qw( Exporter ); |
|
94
|
|
|
|
|
|
|
@EXPORT_OK = qw( split_sentences ); |
|
95
|
|
|
|
|
|
|
@PUNCTUATION = ( '\.', '\!', '\?' ); |
|
96
|
|
|
|
|
|
|
|
|
97
|
|
|
|
|
|
|
#============================================================================== |
|
98
|
|
|
|
|
|
|
# |
|
99
|
|
|
|
|
|
|
# Public methods |
|
100
|
|
|
|
|
|
|
# |
|
101
|
|
|
|
|
|
|
#============================================================================== |
|
102
|
|
|
|
|
|
|
|
|
103
|
|
|
|
|
|
|
#------------------------------------------------------------------------------ |
|
104
|
|
|
|
|
|
|
# |
|
105
|
|
|
|
|
|
|
# split_sentences - takes text input an splits it into sentences, based on a |
|
106
|
|
|
|
|
|
|
# fairly approximate regex. Returns an array of the sentences. |
|
107
|
|
|
|
|
|
|
# |
|
108
|
|
|
|
|
|
|
#------------------------------------------------------------------------------ |
|
109
|
|
|
|
|
|
|
|
|
110
|
|
|
|
|
|
|
sub split_sentences |
|
111
|
|
|
|
|
|
|
{ |
|
112
|
16
|
|
|
16
|
1
|
3187
|
my $text = shift; |
|
113
|
|
|
|
|
|
|
|
|
114
|
16
|
50
|
|
|
|
53
|
return () unless $text; |
|
115
|
|
|
|
|
|
|
|
|
116
|
|
|
|
|
|
|
# capital letter is a character set; to account for local, this includes |
|
117
|
|
|
|
|
|
|
# all characters for which lc is different from that character |
|
118
|
|
|
|
|
|
|
|
|
119
|
|
|
|
|
|
|
my $capital_letter = |
|
120
|
|
|
|
|
|
|
'[' . |
|
121
|
|
|
|
|
|
|
join( '', |
|
122
|
3056
|
|
|
|
|
3684
|
grep { lc( $_ ) ne ( $_ ) } |
|
123
|
16
|
|
|
|
|
57
|
map { chr( $_ ) } ord( "A" ) .. ord( "\xff" ) |
|
|
3056
|
|
|
|
|
4005
|
|
|
124
|
|
|
|
|
|
|
) . |
|
125
|
|
|
|
|
|
|
']' |
|
126
|
|
|
|
|
|
|
; |
|
127
|
|
|
|
|
|
|
|
|
128
|
16
|
|
|
|
|
217
|
my $punctuation = '(?:' . join( '|', @PUNCTUATION ) . ')'; |
|
129
|
|
|
|
|
|
|
|
|
130
|
|
|
|
|
|
|
# this needs to be alternation, not character class, because of |
|
131
|
|
|
|
|
|
|
# multibyte characters |
|
132
|
|
|
|
|
|
|
|
|
133
|
16
|
|
|
|
|
33
|
my $opt_start_quote = q/['"]?/; # "' |
|
134
|
16
|
|
|
|
|
23
|
my $opt_close_quote = q/['"]?/; # "' |
|
135
|
|
|
|
|
|
|
|
|
136
|
|
|
|
|
|
|
# these are distinguished because (eventually!) I would like to do |
|
137
|
|
|
|
|
|
|
# locale stuff on quote characters |
|
138
|
|
|
|
|
|
|
|
|
139
|
16
|
|
|
|
|
26
|
my $opt_start_bracket = q/[[({]?/; # }{ |
|
140
|
16
|
|
|
|
|
19
|
my $opt_close_bracket = q/[\])}]?/; |
|
141
|
|
|
|
|
|
|
|
|
142
|
|
|
|
|
|
|
# return $text if there is no punctuation ... |
|
143
|
|
|
|
|
|
|
|
|
144
|
16
|
50
|
|
|
|
247
|
return $text unless $text =~ /$punctuation/; |
|
145
|
|
|
|
|
|
|
|
|
146
|
16
|
|
|
|
|
708
|
my @sentences = $text =~ / |
|
147
|
|
|
|
|
|
|
( |
|
148
|
|
|
|
|
|
|
# sentences start with ... |
|
149
|
|
|
|
|
|
|
$opt_start_quote # an optional start quote |
|
150
|
|
|
|
|
|
|
$opt_start_bracket # an optional start bracket |
|
151
|
|
|
|
|
|
|
$capital_letter # a capital letter ... |
|
152
|
|
|
|
|
|
|
.+? # at least some (non-greedy) anything ... |
|
153
|
|
|
|
|
|
|
$punctuation # ... followed by any one of !?. |
|
154
|
|
|
|
|
|
|
$opt_close_quote # an optional close quote |
|
155
|
|
|
|
|
|
|
$opt_close_bracket # and an optional close bracket |
|
156
|
|
|
|
|
|
|
) |
|
157
|
|
|
|
|
|
|
(?= # with lookahead that it is followed by ... |
|
158
|
|
|
|
|
|
|
(?: # either ... |
|
159
|
|
|
|
|
|
|
\s+ # some whitespace ... |
|
160
|
|
|
|
|
|
|
$opt_start_quote # an optional start quote |
|
161
|
|
|
|
|
|
|
$opt_start_bracket # an optional start bracket |
|
162
|
|
|
|
|
|
|
$capital_letter # an uppercase word character (for locale |
|
163
|
|
|
|
|
|
|
# sensitive matching) |
|
164
|
|
|
|
|
|
|
| # or ... |
|
165
|
|
|
|
|
|
|
\n\n # a couple (or more) of CRs |
|
166
|
|
|
|
|
|
|
| # or ... |
|
167
|
|
|
|
|
|
|
\s*$ # optional whitespace, followed by end of string |
|
168
|
|
|
|
|
|
|
) |
|
169
|
|
|
|
|
|
|
) |
|
170
|
|
|
|
|
|
|
/gxs |
|
171
|
|
|
|
|
|
|
; |
|
172
|
16
|
50
|
|
|
|
103
|
return @sentences if @sentences; |
|
173
|
0
|
|
|
|
|
|
return ( $text ); |
|
174
|
|
|
|
|
|
|
} |
|
175
|
|
|
|
|
|
|
|
|
176
|
|
|
|
|
|
|
#============================================================================== |
|
177
|
|
|
|
|
|
|
# |
|
178
|
|
|
|
|
|
|
# Return TRUE |
|
179
|
|
|
|
|
|
|
# |
|
180
|
|
|
|
|
|
|
#============================================================================== |
|
181
|
|
|
|
|
|
|
|
|
182
|
|
|
|
|
|
|
1; |