line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
package WWW::Babelfish; |
2
|
|
|
|
|
|
|
|
3
|
|
|
|
|
|
|
require 5.008; |
4
|
|
|
|
|
|
|
|
5
|
1
|
|
|
1
|
|
582
|
use strict; |
|
1
|
|
|
|
|
1
|
|
|
1
|
|
|
|
|
32
|
|
6
|
1
|
|
|
1
|
|
3
|
use vars qw($VERSION @ISA @EXPORT @EXPORT_OK); |
|
1
|
|
|
|
|
2
|
|
|
1
|
|
|
|
|
100
|
|
7
|
|
|
|
|
|
|
|
8
|
|
|
|
|
|
|
require Exporter; |
9
|
|
|
|
|
|
|
require AutoLoader; |
10
|
|
|
|
|
|
|
|
11
|
|
|
|
|
|
|
@ISA = qw(Exporter AutoLoader); |
12
|
|
|
|
|
|
|
# Items to export into callers namespace by default. Note: do not export |
13
|
|
|
|
|
|
|
# names by default without a very good reason. Use EXPORT_OK instead. |
14
|
|
|
|
|
|
|
# Do not simply export all your public functions/methods/constants. |
15
|
|
|
|
|
|
|
@EXPORT = qw(); |
16
|
|
|
|
|
|
|
|
17
|
|
|
|
|
|
|
$VERSION = '0.16'; |
18
|
|
|
|
|
|
|
|
19
|
|
|
|
|
|
|
# Preloaded methods go here. |
20
|
|
|
|
|
|
|
|
21
|
1
|
|
|
1
|
|
719
|
use HTTP::Request::Common qw(POST); |
|
1
|
|
|
|
|
55179
|
|
|
1
|
|
|
|
|
98
|
|
22
|
1
|
|
|
1
|
|
1705
|
use LWP::UserAgent; |
|
1
|
|
|
|
|
37803
|
|
|
1
|
|
|
|
|
33
|
|
23
|
1
|
|
|
1
|
|
742
|
use HTML::TokeParser; |
|
1
|
|
|
|
|
14152
|
|
|
1
|
|
|
|
|
32
|
|
24
|
1
|
|
|
1
|
|
889
|
use IO::String; |
|
1
|
|
|
|
|
4650
|
|
|
1
|
|
|
|
|
29
|
|
25
|
1
|
|
|
1
|
|
3056
|
use Encode; |
|
1
|
|
|
|
|
12827
|
|
|
1
|
|
|
|
|
2463
|
|
26
|
|
|
|
|
|
|
|
27
|
|
|
|
|
|
|
my $MAXCHUNK = 1000; # Maximum number of characters |
28
|
|
|
|
|
|
|
# Bablefish will translate at one time |
29
|
|
|
|
|
|
|
|
30
|
|
|
|
|
|
|
my $MAXRETRIES = 50; # Maximum number of retries for a chunk of text |
31
|
|
|
|
|
|
|
$| = 1; |
32
|
|
|
|
|
|
|
|
33
|
|
|
|
|
|
|
my $Services = { |
34
|
|
|
|
|
|
|
Babelfish => { |
35
|
|
|
|
|
|
|
agent => $0 . ":" . __PACKAGE__ . "/" . $VERSION, |
36
|
|
|
|
|
|
|
|
37
|
|
|
|
|
|
|
languagesrequest => sub { |
38
|
|
|
|
|
|
|
my $req = new HTTP::Request(GET => 'http://babelfish.altavista.com/babelfish/tr?il=en'); |
39
|
|
|
|
|
|
|
return $req; |
40
|
|
|
|
|
|
|
}, |
41
|
|
|
|
|
|
|
|
42
|
|
|
|
|
|
|
translaterequest => sub { |
43
|
|
|
|
|
|
|
my($langpair, $text) = @_; |
44
|
|
|
|
|
|
|
my $req = POST ( 'http://babelfish.altavista.com/babelfish/tr?il=en', |
45
|
|
|
|
|
|
|
[ 'doit' => 'done', 'urltext' => encode("utf8",$text), 'lp' => $langpair, 'Submit' => 'Translate', 'enc' => 'utf8' ], qw(Accept-Charset utf-8) ); |
46
|
|
|
|
|
|
|
return $req; |
47
|
|
|
|
|
|
|
}, |
48
|
|
|
|
|
|
|
|
49
|
|
|
|
|
|
|
# Extract the text from the html we get back from babelfish and return |
50
|
|
|
|
|
|
|
# it (keying on the fact that it's the first thing after a tag, |
51
|
|
|
|
|
|
|
# possibly removing a textarea tag after it). |
52
|
|
|
|
|
|
|
|
53
|
|
|
|
|
|
|
# extract_text => sub { |
54
|
|
|
|
|
|
|
# my($html) = @_; |
55
|
|
|
|
|
|
|
# my $p = HTML::TokeParser->new(\$html); |
56
|
|
|
|
|
|
|
# my $tag; |
57
|
|
|
|
|
|
|
# while ($tag = $p->get_tag('input')) { |
58
|
|
|
|
|
|
|
# $_ = @{$tag}[1]->{value} if @{$tag}[1]->{name} eq 'q'; |
59
|
|
|
|
|
|
|
# return decode("utf8",$_); |
60
|
|
|
|
|
|
|
# } |
61
|
|
|
|
|
|
|
|
62
|
|
|
|
|
|
|
extract_text => sub { |
63
|
|
|
|
|
|
|
my($html) = @_; |
64
|
|
|
|
|
|
|
my $p = HTML::TokeParser->new(\$html); |
65
|
|
|
|
|
|
|
while ( my $_tag = $p->get_tag('div') ) { |
66
|
|
|
|
|
|
|
my($tag,$attr,$attrseq) = @$_tag; |
67
|
|
|
|
|
|
|
next unless @$attrseq == 1 |
68
|
|
|
|
|
|
|
&& $attrseq->[-1] eq 'style' |
69
|
|
|
|
|
|
|
&& $attr->{style} eq 'padding:10px;'; |
70
|
|
|
|
|
|
|
my($token) = $p->get_token or return; |
71
|
|
|
|
|
|
|
my ( $type, $text, $is_data ) = @$token; |
72
|
|
|
|
|
|
|
next if $type ne 'T'; |
73
|
|
|
|
|
|
|
return decode( utf8 => $text ); |
74
|
|
|
|
|
|
|
} |
75
|
|
|
|
|
|
|
|
76
|
|
|
|
|
|
|
|
77
|
|
|
|
|
|
|
} |
78
|
|
|
|
|
|
|
}, |
79
|
|
|
|
|
|
|
|
80
|
|
|
|
|
|
|
Google => { |
81
|
|
|
|
|
|
|
agent => 'Mozilla/5.0', # Google is finicky |
82
|
|
|
|
|
|
|
|
83
|
|
|
|
|
|
|
languagesrequest => sub { |
84
|
|
|
|
|
|
|
my $req = new HTTP::Request(GET => 'http://www.google.com/language_tools?hl=en'); |
85
|
|
|
|
|
|
|
return $req; |
86
|
|
|
|
|
|
|
}, |
87
|
|
|
|
|
|
|
|
88
|
|
|
|
|
|
|
translaterequest => sub { |
89
|
|
|
|
|
|
|
my($langpair, $text) = @_; |
90
|
|
|
|
|
|
|
my $req = POST ( 'http://translate.google.com/translate_t', |
91
|
|
|
|
|
|
|
[ 'text' => encode("utf8",$text), 'langpair' => $langpair, hl => 'en', ie => "UTF8", oe => "UTF8",]); |
92
|
|
|
|
|
|
|
return $req; |
93
|
|
|
|
|
|
|
}, |
94
|
|
|
|
|
|
|
|
95
|
|
|
|
|
|
|
extract_text => sub { |
96
|
|
|
|
|
|
|
my($html) = @_; |
97
|
|
|
|
|
|
|
my $p = HTML::TokeParser->new(\$html); |
98
|
|
|
|
|
|
|
my $tag; |
99
|
|
|
|
|
|
|
while ($tag = $p->get_tag('div')) { |
100
|
|
|
|
|
|
|
if (@{$tag}[1]->{id} eq 'result_box') { |
101
|
|
|
|
|
|
|
$_ = $p->get_text; |
102
|
|
|
|
|
|
|
return decode("utf8",$_); |
103
|
|
|
|
|
|
|
} |
104
|
|
|
|
|
|
|
} |
105
|
|
|
|
|
|
|
} |
106
|
|
|
|
|
|
|
}, |
107
|
|
|
|
|
|
|
|
108
|
|
|
|
|
|
|
Yahoo => { |
109
|
|
|
|
|
|
|
agent => $0 . ":" . __PACKAGE__ . "/" . $VERSION, |
110
|
|
|
|
|
|
|
|
111
|
|
|
|
|
|
|
languagesrequest => sub { |
112
|
|
|
|
|
|
|
my $req = new HTTP::Request(GET => 'http://babelfish.yahoo.com/translate_txt'); |
113
|
|
|
|
|
|
|
return $req; |
114
|
|
|
|
|
|
|
}, |
115
|
|
|
|
|
|
|
|
116
|
|
|
|
|
|
|
translaterequest => sub { |
117
|
|
|
|
|
|
|
my($langpair, $text) = @_; |
118
|
|
|
|
|
|
|
my $req = POST ( 'http://babelfish.yahoo.com/translate_txt', |
119
|
|
|
|
|
|
|
[ 'ei' => 'UTF-8', 'doit' => 'done', 'tt' => 'urltext', 'trtext' => encode("utf8",$text), 'lp' => $langpair, 'btnTrTxt' => 'Translate', 'intl' => '1' ]); |
120
|
|
|
|
|
|
|
return $req; |
121
|
|
|
|
|
|
|
}, |
122
|
|
|
|
|
|
|
|
123
|
|
|
|
|
|
|
# Extract the text from the html we get back from Yahoo |
124
|
|
|
|
|
|
|
extract_text => sub { |
125
|
|
|
|
|
|
|
my($html) = @_; |
126
|
|
|
|
|
|
|
my $p = HTML::TokeParser->new(\$html); |
127
|
|
|
|
|
|
|
my $tag; |
128
|
|
|
|
|
|
|
while ($tag = $p->get_tag('div')) { |
129
|
|
|
|
|
|
|
next if (@{$tag}[1]->{id} ne 'result'); |
130
|
|
|
|
|
|
|
$_ = $p->get_text('/div'); |
131
|
|
|
|
|
|
|
return decode("utf8",$_); |
132
|
|
|
|
|
|
|
} |
133
|
|
|
|
|
|
|
} |
134
|
|
|
|
|
|
|
}, |
135
|
|
|
|
|
|
|
|
136
|
|
|
|
|
|
|
}; |
137
|
|
|
|
|
|
|
|
138
|
|
|
|
|
|
|
|
139
|
|
|
|
|
|
|
sub new { |
140
|
0
|
|
|
0
|
1
|
|
my ($this, @args) = @_; |
141
|
0
|
|
0
|
|
|
|
my $class = ref($this) || $this; |
142
|
0
|
|
|
|
|
|
my $self = {}; |
143
|
0
|
|
|
|
|
|
bless $self, $class; |
144
|
0
|
0
|
|
|
|
|
return undef unless( $self->initialize(@args) ); |
145
|
0
|
|
|
|
|
|
return $self; |
146
|
|
|
|
|
|
|
} |
147
|
|
|
|
|
|
|
|
148
|
|
|
|
|
|
|
sub initialize { |
149
|
0
|
|
|
0
|
0
|
|
my($self, %params) = @_; |
150
|
|
|
|
|
|
|
|
151
|
0
|
|
0
|
|
|
|
$self->{service} = $params{service} || 'Babelfish'; |
152
|
0
|
0
|
|
|
|
|
die "No such service: " . $self->{service} unless defined $Services->{ $self->{service} }; |
153
|
|
|
|
|
|
|
|
154
|
|
|
|
|
|
|
# Caller can set user agent; we default to "script:WWW::Babelfish/0.01" |
155
|
0
|
|
0
|
|
|
|
$self->{agent} = $params{agent} || $Services->{agent}; |
156
|
|
|
|
|
|
|
|
157
|
0
|
0
|
|
|
|
|
$self->{proxy} = $params{proxy} if defined $params{proxy}; |
158
|
|
|
|
|
|
|
|
159
|
|
|
|
|
|
|
# Get the page |
160
|
0
|
|
|
|
|
|
my $ua = new LWP::UserAgent; |
161
|
0
|
0
|
|
|
|
|
$ua->proxy('http','http://' . $self->{proxy}) if defined $self->{proxy}; |
162
|
0
|
|
|
|
|
|
$ua->agent($self->{agent}); |
163
|
0
|
|
|
|
|
|
$self->{ua} = $ua; |
164
|
|
|
|
|
|
|
|
165
|
0
|
|
|
|
|
|
my $req = &{ $Services->{ $self->{service} }->{languagesrequest} }; |
|
0
|
|
|
|
|
|
|
166
|
0
|
|
|
|
|
|
my $res = $ua->request($req); |
167
|
0
|
0
|
|
|
|
|
unless($res->is_success){ |
168
|
0
|
|
|
|
|
|
warn(__PACKAGE__ . ":" . $res->status_line); |
169
|
0
|
|
|
|
|
|
return 0; |
170
|
|
|
|
|
|
|
} |
171
|
0
|
|
|
|
|
|
my $page = $res->content; |
172
|
|
|
|
|
|
|
|
173
|
|
|
|
|
|
|
# Extract the language names and the mapping of languages to options to |
174
|
|
|
|
|
|
|
# be passed back, and store them on our object in "Langs" hash of hashes |
175
|
|
|
|
|
|
|
# Incredibly, this works for both Babelfish and Google; it should really |
176
|
|
|
|
|
|
|
# be a method in $Services |
177
|
0
|
|
|
|
|
|
my $p = HTML::TokeParser->new(\$page); |
178
|
0
|
|
|
|
|
|
my $a2b; |
179
|
0
|
0
|
|
|
|
|
if ( $p->get_tag("select") ) { |
180
|
0
|
|
|
|
|
|
while ( $_ = $p->get_tag("option") ) { |
181
|
0
|
|
|
|
|
|
$a2b = $p->get_trimmed_text; |
182
|
0
|
0
|
|
|
|
|
next if $a2b =~ /Select from and to languages/; # This for babelfish |
183
|
0
|
0
|
|
|
|
|
$a2b =~ /(\S+)\sto\s(\S+)/ or next; |
184
|
0
|
|
|
|
|
|
$self->{Langs}{$1}{$2} = $_->[1]{value}; |
185
|
0
|
|
0
|
|
|
|
$self->{Langs}{$2} ||= {}; |
186
|
|
|
|
|
|
|
} |
187
|
|
|
|
|
|
|
} |
188
|
|
|
|
|
|
|
|
189
|
0
|
|
|
|
|
|
return 1; |
190
|
|
|
|
|
|
|
} |
191
|
|
|
|
|
|
|
|
192
|
|
|
|
|
|
|
sub services { |
193
|
0
|
|
|
0
|
1
|
|
my $self = shift; |
194
|
0
|
0
|
|
|
|
|
if($self){ |
195
|
0
|
|
|
|
|
|
return keys %{$self->Services}; |
|
0
|
|
|
|
|
|
|
196
|
|
|
|
|
|
|
} |
197
|
|
|
|
|
|
|
else{ |
198
|
0
|
|
|
|
|
|
return keys %{$Services}; |
|
0
|
|
|
|
|
|
|
199
|
|
|
|
|
|
|
} |
200
|
|
|
|
|
|
|
} |
201
|
|
|
|
|
|
|
|
202
|
|
|
|
|
|
|
sub languages { |
203
|
0
|
|
|
0
|
1
|
|
my $self = shift; |
204
|
0
|
|
|
|
|
|
return sort keys %{$self->{Langs}}; |
|
0
|
|
|
|
|
|
|
205
|
|
|
|
|
|
|
} |
206
|
|
|
|
|
|
|
|
207
|
|
|
|
|
|
|
sub languagepairs { |
208
|
0
|
|
|
0
|
1
|
|
my $self = shift; |
209
|
0
|
|
|
|
|
|
return $self->{Langs}; |
210
|
|
|
|
|
|
|
} |
211
|
|
|
|
|
|
|
|
212
|
|
|
|
|
|
|
sub translate { |
213
|
0
|
|
|
0
|
1
|
|
my ($self, %params) = @_; |
214
|
|
|
|
|
|
|
|
215
|
|
|
|
|
|
|
# Paragraph separator is "\n\n" by default |
216
|
0
|
|
0
|
|
|
|
local $/ = $params{delimiter} || "\n\n"; |
217
|
0
|
|
|
|
|
|
local $_; |
218
|
|
|
|
|
|
|
|
219
|
0
|
0
|
|
|
|
|
$params{delimiter} = "\n\n" if ( ! defined( $params{delimiter} ) ); |
220
|
|
|
|
|
|
|
|
221
|
0
|
|
|
|
|
|
undef $self->{error}; |
222
|
0
|
0
|
|
|
|
|
unless ( exists($self->{Langs}->{$params{source}}) ) { |
223
|
0
|
|
|
|
|
|
$self->{error} = qq(Language "$params{source}" is not available); |
224
|
0
|
|
|
|
|
|
warn(__PACKAGE__ . ": " . $self->{error} . "\n"); |
225
|
0
|
|
|
|
|
|
return undef; |
226
|
|
|
|
|
|
|
} |
227
|
|
|
|
|
|
|
|
228
|
|
|
|
|
|
|
# This "feature" is actually useful as a pass-thru filter. |
229
|
|
|
|
|
|
|
# Babelfish doesn't do same-to-same anyway (though it would be |
230
|
|
|
|
|
|
|
# pretty interesting if it did) |
231
|
0
|
0
|
|
|
|
|
return $params{text} if $params{source} eq $params{destination}; |
232
|
|
|
|
|
|
|
|
233
|
0
|
0
|
|
|
|
|
unless ( exists($self->{Langs}->{$params{source}}{$params{destination}}) ) { |
234
|
0
|
|
|
|
|
|
$self->{error} = |
235
|
|
|
|
|
|
|
qq(Cannot translate from "$params{source}" to "$params{destination}"); |
236
|
0
|
|
|
|
|
|
warn(__PACKAGE__ . ": " . $self->{error} . "\n"); |
237
|
0
|
|
|
|
|
|
return undef; |
238
|
|
|
|
|
|
|
} |
239
|
|
|
|
|
|
|
|
240
|
0
|
|
|
|
|
|
my $langopt = $self->{Langs}{$params{source}}{$params{destination}}; |
241
|
|
|
|
|
|
|
|
242
|
0
|
|
|
|
|
|
my $th; # "Text Handle" |
243
|
0
|
0
|
|
|
|
|
if ( ref $params{text} ) { # We've been passed a filehandle |
244
|
0
|
|
|
|
|
|
$th = $params{text}; |
245
|
|
|
|
|
|
|
} else { # We've been passed a string |
246
|
0
|
|
|
|
|
|
$th = new IO::String($params{text}); |
247
|
|
|
|
|
|
|
} |
248
|
|
|
|
|
|
|
|
249
|
0
|
|
|
|
|
|
my $Text = ""; |
250
|
0
|
|
|
|
|
|
my $WANT_STRING_RETURNED = 0; |
251
|
0
|
0
|
|
|
|
|
unless ( defined $params{ofh} ) { |
252
|
0
|
|
|
|
|
|
$params{ofh} = new IO::String($Text); |
253
|
0
|
|
|
|
|
|
$WANT_STRING_RETURNED = 1; |
254
|
|
|
|
|
|
|
} |
255
|
|
|
|
|
|
|
|
256
|
|
|
|
|
|
|
# Variables we use in the next mega-block |
257
|
0
|
|
|
|
|
|
my $para; # paragraph |
258
|
0
|
|
|
|
|
|
my $num_paras = 0; # number of paragraphs |
259
|
0
|
|
|
|
|
|
my $transpara; # translated paragraph |
260
|
0
|
|
|
|
|
|
my $para_start_ws = ""; # initial whitespace in paragraph |
261
|
0
|
|
|
|
|
|
my $chunk; # paragraph piece to feed to babelfish |
262
|
|
|
|
|
|
|
my $req; # LWP request object |
263
|
0
|
|
|
|
|
|
my $ua; # LWP user agent |
264
|
0
|
|
|
|
|
|
my $res; # LWP result |
265
|
0
|
|
|
|
|
|
my $text; # translated chunk |
266
|
0
|
|
|
|
|
|
my $i; # a counter |
267
|
0
|
|
|
|
|
|
while ($para = <$th>) { |
268
|
0
|
|
|
|
|
|
$num_paras++; |
269
|
0
|
|
|
|
|
|
$transpara = ""; |
270
|
|
|
|
|
|
|
|
271
|
|
|
|
|
|
|
# Extract any leading whitespace from the start of the paragraph |
272
|
|
|
|
|
|
|
# Babelfish will eat it anyway. |
273
|
0
|
0
|
|
|
|
|
if ($para =~ s/(^\s+)(\S)/$2/) { |
274
|
0
|
|
0
|
|
|
|
$para_start_ws = $1 || ""; |
275
|
|
|
|
|
|
|
} |
276
|
0
|
|
|
|
|
|
$para =~ s/$params{delimiter}//; # Remove the para delimiter |
277
|
|
|
|
|
|
|
|
278
|
|
|
|
|
|
|
CHUNK: |
279
|
0
|
|
|
|
|
|
foreach $chunk ( $self->_chunk_text($MAXCHUNK, $para) ) { |
280
|
0
|
|
|
|
|
|
$req = &{ $Services->{ $self->{service} }->{translaterequest} }($langopt, $chunk); |
|
0
|
|
|
|
|
|
|
281
|
0
|
|
|
|
|
|
$ua = $self->{ua}; |
282
|
|
|
|
|
|
|
|
283
|
|
|
|
|
|
|
RETRY: |
284
|
0
|
|
|
|
|
|
for ($i = 0; $i <= $MAXRETRIES; $i++) { |
285
|
0
|
|
|
|
|
|
$res = $ua->request($req); |
286
|
|
|
|
|
|
|
|
287
|
0
|
0
|
|
|
|
|
if ( $res->is_success ) { |
288
|
|
|
|
|
|
|
|
289
|
|
|
|
|
|
|
#$text = $self->_extract_text($res->as_string); #REMOVE |
290
|
0
|
|
|
|
|
|
$text = &{ $Services->{ $self->{service} }->{extract_text} }($res->as_string); |
|
0
|
|
|
|
|
|
|
291
|
0
|
0
|
0
|
|
|
|
if ( ( ! defined( $text ) ) || |
292
|
|
|
|
|
|
|
( $text =~ /^\*\*time-out\*\*/ ) |
293
|
|
|
|
|
|
|
) # in-band signalling; yuck |
294
|
|
|
|
|
|
|
{ |
295
|
0
|
|
|
|
|
|
next RETRY; |
296
|
|
|
|
|
|
|
|
297
|
|
|
|
|
|
|
} ## end if |
298
|
|
|
|
|
|
|
|
299
|
0
|
|
|
|
|
|
$text =~ s/\n$//; # Babelfish likes to append newlines |
300
|
0
|
|
|
|
|
|
$transpara .= $text; |
301
|
|
|
|
|
|
|
|
302
|
0
|
|
|
|
|
|
next CHUNK; |
303
|
|
|
|
|
|
|
} |
304
|
|
|
|
|
|
|
} |
305
|
0
|
|
|
|
|
|
$self->{error} = "Request timed out more than $MAXRETRIES times"; |
306
|
0
|
|
|
|
|
|
return undef; |
307
|
|
|
|
|
|
|
} |
308
|
0
|
0
|
|
|
|
|
print { $params{ofh} } $/ if $num_paras > 1; |
|
0
|
|
|
|
|
|
|
309
|
0
|
|
|
|
|
|
print { $params{ofh} } $para_start_ws . $transpara; |
|
0
|
|
|
|
|
|
|
310
|
|
|
|
|
|
|
} |
311
|
|
|
|
|
|
|
|
312
|
0
|
0
|
|
|
|
|
if ( $WANT_STRING_RETURNED ) { |
313
|
0
|
|
|
|
|
|
return $Text; |
314
|
|
|
|
|
|
|
} else { |
315
|
0
|
|
|
|
|
|
return 1; |
316
|
|
|
|
|
|
|
} |
317
|
|
|
|
|
|
|
} |
318
|
|
|
|
|
|
|
|
319
|
|
|
|
|
|
|
sub error { |
320
|
0
|
|
|
0
|
1
|
|
my $self = shift; |
321
|
0
|
|
|
|
|
|
return $self->{error}; |
322
|
|
|
|
|
|
|
} |
323
|
|
|
|
|
|
|
|
324
|
|
|
|
|
|
|
# Given a maximum chunk size and some text, return |
325
|
|
|
|
|
|
|
# an array of pieces of the text chopped up in a |
326
|
|
|
|
|
|
|
# logical way and less than or equal to the chunk size |
327
|
|
|
|
|
|
|
sub _chunk_text { |
328
|
0
|
|
|
0
|
|
|
my($self, $max, $text) = @_; |
329
|
|
|
|
|
|
|
|
330
|
0
|
|
|
|
|
|
my @result; |
331
|
|
|
|
|
|
|
|
332
|
|
|
|
|
|
|
# The trivial case |
333
|
0
|
0
|
|
|
|
|
return($text) if length($text) <= $max; |
334
|
|
|
|
|
|
|
|
335
|
|
|
|
|
|
|
# Hmmm. There are a couple of ways we could do this. |
336
|
|
|
|
|
|
|
# I'm guessing that Babelfish doesn't look at any structure larger than |
337
|
|
|
|
|
|
|
# a sentence; in fact I'm often tempted to guess that it doesn't look |
338
|
|
|
|
|
|
|
# at anything larger than a word, but we'll give it the benefit of the doubt. |
339
|
|
|
|
|
|
|
# |
340
|
|
|
|
|
|
|
|
341
|
|
|
|
|
|
|
# FIXME there are no built-in regexps for matching sentence |
342
|
|
|
|
|
|
|
# breaks; I'm not sure if terminal punctuation will work for all |
343
|
|
|
|
|
|
|
# languages... |
344
|
|
|
|
|
|
|
|
345
|
0
|
|
|
|
|
|
my $total = length($text); |
346
|
0
|
|
|
|
|
|
my $offset = 0; |
347
|
0
|
|
|
|
|
|
my $lastoffset = 0; |
348
|
0
|
|
|
|
|
|
my $test; |
349
|
|
|
|
|
|
|
my $chunk; |
350
|
|
|
|
|
|
|
|
351
|
0
|
|
|
|
|
|
while ( ($total - $lastoffset) > $max) { |
352
|
0
|
|
|
|
|
|
$test = $lastoffset + $max; |
353
|
|
|
|
|
|
|
|
354
|
|
|
|
|
|
|
# Split by terminal punctuation... |
355
|
0
|
|
|
|
|
|
@_ = sort {$b <=> $a} ( rindex($text, '.', $test), |
|
0
|
|
|
|
|
|
|
356
|
|
|
|
|
|
|
rindex($text, '!', $test), |
357
|
|
|
|
|
|
|
rindex($text, '?', $test), |
358
|
|
|
|
|
|
|
); |
359
|
0
|
|
|
|
|
|
$offset = shift(@_) + 1; |
360
|
|
|
|
|
|
|
|
361
|
|
|
|
|
|
|
# or by clause... |
362
|
0
|
0
|
0
|
|
|
|
if ( $offset == -1 or $offset <= $lastoffset ) { |
363
|
0
|
|
|
|
|
|
@_ = sort {$b <=> $a} ( rindex($text, ',', $test), |
|
0
|
|
|
|
|
|
|
364
|
|
|
|
|
|
|
rindex($text, ';', $test), |
365
|
|
|
|
|
|
|
rindex($text, ':', $test), |
366
|
|
|
|
|
|
|
); |
367
|
0
|
|
|
|
|
|
$offset = shift(@_) + 1; |
368
|
|
|
|
|
|
|
|
369
|
|
|
|
|
|
|
|
370
|
|
|
|
|
|
|
# or by word |
371
|
0
|
0
|
0
|
|
|
|
if ( $offset == -1 or $offset <= $lastoffset) { |
372
|
0
|
|
|
|
|
|
$offset = rindex($text, " ", $test); |
373
|
|
|
|
|
|
|
} |
374
|
|
|
|
|
|
|
|
375
|
|
|
|
|
|
|
# or give up |
376
|
0
|
0
|
|
|
|
|
return undef if $offset == -1; |
377
|
|
|
|
|
|
|
} |
378
|
|
|
|
|
|
|
|
379
|
0
|
|
|
|
|
|
$chunk = substr($text, $lastoffset, $offset - $lastoffset); |
380
|
|
|
|
|
|
|
|
381
|
0
|
|
|
|
|
|
push( @result, $chunk); |
382
|
0
|
|
|
|
|
|
$lastoffset = $offset; |
383
|
|
|
|
|
|
|
} |
384
|
|
|
|
|
|
|
|
385
|
0
|
|
|
|
|
|
push( @result, substr($text, $lastoffset) ); |
386
|
0
|
|
|
|
|
|
return @result; |
387
|
|
|
|
|
|
|
} |
388
|
|
|
|
|
|
|
|
389
|
|
|
|
|
|
|
# This code is now obsoleted by the new result page format, but I'm |
390
|
|
|
|
|
|
|
# leaving it here commented out in case we end up needing the |
391
|
|
|
|
|
|
|
# whitespace hack again. |
392
|
|
|
|
|
|
|
# |
393
|
|
|
|
|
|
|
# my ($tag,$token); |
394
|
|
|
|
|
|
|
# my $text=""; |
395
|
|
|
|
|
|
|
# if ($tag = $p->get_tag('br')) { |
396
|
|
|
|
|
|
|
# while ($token = $p->get_token) { |
397
|
|
|
|
|
|
|
# next if shift(@{$token}) ne "T"; |
398
|
|
|
|
|
|
|
# $text = shift(@{$token}); |
399
|
|
|
|
|
|
|
|
400
|
|
|
|
|
|
|
# #$text =~ s/[\r\n]//g; |
401
|
|
|
|
|
|
|
# # This patch for whitespace handling from Olivier Scherler |
402
|
|
|
|
|
|
|
# $text =~ s/[\r\n]/ /g; |
403
|
|
|
|
|
|
|
# $text =~ s/^\s*//; |
404
|
|
|
|
|
|
|
# $text =~ s/\s+/ /g; |
405
|
|
|
|
|
|
|
# $text =~ s/\s+$//; |
406
|
|
|
|
|
|
|
|
407
|
|
|
|
|
|
|
# last if defined($text) and $text ne ""; |
408
|
|
|
|
|
|
|
# } |
409
|
|
|
|
|
|
|
# } |
410
|
|
|
|
|
|
|
# return $text; |
411
|
|
|
|
|
|
|
|
412
|
|
|
|
|
|
|
|
413
|
|
|
|
|
|
|
#} |
414
|
|
|
|
|
|
|
|
415
|
|
|
|
|
|
|
|
416
|
|
|
|
|
|
|
# Autoload methods go after =cut, and are processed by the autosplit program. |
417
|
|
|
|
|
|
|
|
418
|
|
|
|
|
|
|
1; |
419
|
|
|
|
|
|
|
__END__ |