line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
package Template::Multilingual::Parser; |
2
|
|
|
|
|
|
|
|
3
|
3
|
|
|
3
|
|
29
|
use strict; |
|
3
|
|
|
|
|
8
|
|
|
3
|
|
|
|
|
164
|
|
4
|
3
|
|
|
3
|
|
17
|
use base qw(Template::Parser); |
|
3
|
|
|
|
|
8
|
|
|
3
|
|
|
|
|
3887
|
|
5
|
|
|
|
|
|
|
|
6
|
|
|
|
|
|
|
our $VERSION = '1.00'; |
7
|
|
|
|
|
|
|
|
8
|
|
|
|
|
|
|
sub new |
9
|
|
|
|
|
|
|
{ |
10
|
4
|
|
|
4
|
1
|
10
|
my ($class, $options) = @_; |
11
|
4
|
|
|
|
|
42
|
my $self = $class->SUPER::new($options); |
12
|
4
|
|
|
|
|
784
|
$self->{_sections} = []; |
13
|
4
|
|
50
|
|
|
73
|
$self->{_langvar} = $options->{LANGUAGE_VAR} || 'language'; |
14
|
|
|
|
|
|
|
|
15
|
4
|
|
|
|
|
12
|
my $style = $self->{ STYLE }->[-1]; |
16
|
4
|
|
|
|
|
20
|
@$self{ qw(_start _end) } = @$style{ qw( START_TAG END_TAG ) }; |
17
|
4
|
|
|
|
|
13
|
for (qw( _start _end )) { |
18
|
8
|
|
|
|
|
59
|
$self->{$_} =~ s/\\([^\\])/$1/g; |
19
|
|
|
|
|
|
|
} |
20
|
|
|
|
|
|
|
|
21
|
4
|
|
|
|
|
20
|
return $self; |
22
|
|
|
|
|
|
|
} |
23
|
|
|
|
|
|
|
|
24
|
|
|
|
|
|
|
sub parse |
25
|
|
|
|
|
|
|
{ |
26
|
30
|
|
|
30
|
1
|
6444
|
my ($self, $text) = @_; |
27
|
|
|
|
|
|
|
|
28
|
|
|
|
|
|
|
# isolate multilingual sections |
29
|
30
|
|
|
|
|
234
|
$self->_tokenize($text); |
30
|
|
|
|
|
|
|
|
31
|
|
|
|
|
|
|
# replace multilingual sections with TT directives |
32
|
30
|
|
|
|
|
415
|
my ($S, $E, $LANGVAR) = map $self->{$_}, qw(_start _end _langvar); |
33
|
|
|
|
|
|
|
|
34
|
|
|
|
|
|
|
# if language is a variant (en_US), create a template variable holding the fallback value (en) |
35
|
30
|
|
|
|
|
245
|
$text = "$S IF (tm_matches = $LANGVAR.match('^(\\w+)[-_].*\$')); tm_fb = tm_matches.0; END $E"; |
36
|
|
|
|
|
|
|
|
37
|
30
|
|
|
|
|
68
|
for my $section (@{$self->{_sections}}) { |
|
30
|
|
|
|
|
89
|
|
38
|
34
|
100
|
|
|
|
177
|
if ($section->{nolang}) { |
|
|
50
|
|
|
|
|
|
39
|
8
|
|
|
|
|
21
|
$text .= $section->{nolang}; |
40
|
|
|
|
|
|
|
} |
41
|
|
|
|
|
|
|
elsif (my $t = $section->{lang}) { |
42
|
26
|
|
|
|
|
98
|
my @languages = keys %$t; |
43
|
|
|
|
|
|
|
|
44
|
|
|
|
|
|
|
# first loop through languages: look for exact match |
45
|
26
|
|
|
|
|
86
|
$text .= "$S tm_f = 0; SWITCH $LANGVAR $E"; |
46
|
26
|
|
|
|
|
111
|
for my $lang (@languages) { |
47
|
44
|
|
|
|
|
175
|
$text .= "$S CASE '$lang' $E" . $t->{$lang}; |
48
|
|
|
|
|
|
|
} |
49
|
|
|
|
|
|
|
# add a default case to trigger fallback |
50
|
26
|
|
|
|
|
69
|
$text .= "$S CASE; tm_f=1; END; $E"; |
51
|
|
|
|
|
|
|
|
52
|
|
|
|
|
|
|
# second loop: fallback to primary language (en_US matches en) |
53
|
26
|
|
|
|
|
65
|
$text .= "$S IF tm_fb AND tm_f; tm_f=0; SWITCH tm_fb; $E"; |
54
|
26
|
|
|
|
|
134
|
for my $lang (@languages) { |
55
|
44
|
|
|
|
|
458
|
$text .= "$S CASE '$lang' $E" . $t->{$lang}; |
56
|
|
|
|
|
|
|
} |
57
|
|
|
|
|
|
|
# add a default case to trigger last resort fallback |
58
|
|
|
|
|
|
|
# LANG is fr_XX or fr but template has neither |
59
|
|
|
|
|
|
|
# we try to fallback to fr_YY is present |
60
|
26
|
|
|
|
|
39
|
my %seen; |
61
|
26
|
100
|
66
|
|
|
184
|
my @fallbacks = map { /^(\w+)[-_].*$/ && !$seen{$_}++ ? [ $1 => $_] : () } sort @languages; |
|
44
|
|
|
|
|
446
|
|
62
|
26
|
100
|
|
|
|
80
|
if (@fallbacks) { |
63
|
|
|
|
|
|
|
# third loop: fallback to first available variant |
64
|
12
|
|
|
|
|
39
|
$text .= "$S CASE; tm_f=1; END; END; IF tm_f; SWITCH tm_fb || $LANGVAR; $E"; |
65
|
12
|
|
|
|
|
25
|
for my $ref (@fallbacks) { |
66
|
20
|
|
|
|
|
39
|
my ($lang, $variant) = @$ref; |
67
|
20
|
|
|
|
|
154
|
$text .= "$S CASE '$lang' $E" . $t->{$variant}; |
68
|
|
|
|
|
|
|
} |
69
|
|
|
|
|
|
|
} |
70
|
26
|
|
|
|
|
129
|
$text .= "$S END; END $E"; |
71
|
|
|
|
|
|
|
} |
72
|
|
|
|
|
|
|
} |
73
|
30
|
|
|
|
|
210
|
return $self->SUPER::parse ($text); |
74
|
|
|
|
|
|
|
} |
75
|
|
|
|
|
|
|
|
76
|
|
|
|
|
|
|
sub _tokenize |
77
|
|
|
|
|
|
|
{ |
78
|
30
|
|
|
30
|
|
69
|
my ($self, $text) = @_; |
79
|
|
|
|
|
|
|
|
80
|
|
|
|
|
|
|
# extract all sections from the text |
81
|
30
|
|
|
|
|
91
|
$self->{_sections} = []; |
82
|
30
|
|
|
|
|
379
|
my @tokens = split m!(.*?)!s, $text; |
83
|
30
|
|
|
|
|
127
|
my $i = 0; |
84
|
30
|
|
|
|
|
81
|
for my $t (@tokens) { |
85
|
56
|
100
|
|
|
|
144
|
if ($i) { # ... multilingual section |
86
|
26
|
|
|
|
|
139
|
my %section; |
87
|
26
|
|
|
|
|
228
|
while ($t =~ m!<([^<>]+)>(.*?)\1>!gs) { |
88
|
44
|
|
|
|
|
740
|
$section{$1} = $2; |
89
|
|
|
|
|
|
|
} |
90
|
26
|
50
|
|
|
|
277
|
push @{$self->{_sections}}, { lang => \%section } |
|
26
|
|
|
|
|
138
|
|
91
|
|
|
|
|
|
|
if %section; |
92
|
|
|
|
|
|
|
} |
93
|
|
|
|
|
|
|
else { # bare text |
94
|
30
|
100
|
|
|
|
101
|
push @{$self->{_sections}}, { nolang => $t } if $t; |
|
8
|
|
|
|
|
57
|
|
95
|
|
|
|
|
|
|
} |
96
|
56
|
|
|
|
|
297
|
$i = 1 - $i; |
97
|
|
|
|
|
|
|
} |
98
|
|
|
|
|
|
|
} |
99
|
30
|
|
|
30
|
1
|
541863
|
sub sections { $_[0]->{_sections} } |
100
|
|
|
|
|
|
|
|
101
|
|
|
|
|
|
|
=head1 NAME |
102
|
|
|
|
|
|
|
|
103
|
|
|
|
|
|
|
Template::Multilingual::Parser - Multilingual template parser |
104
|
|
|
|
|
|
|
|
105
|
|
|
|
|
|
|
=head1 SYNOPSIS |
106
|
|
|
|
|
|
|
|
107
|
|
|
|
|
|
|
use Template; |
108
|
|
|
|
|
|
|
use Template::Multilingual::Parser; |
109
|
|
|
|
|
|
|
|
110
|
|
|
|
|
|
|
my $parser = Template::Multilingual::Parser->new(); |
111
|
|
|
|
|
|
|
my $template = Template->new(PARSER => $parser); |
112
|
|
|
|
|
|
|
$template->process('example.ttml', { language => 'en'}); |
113
|
|
|
|
|
|
|
|
114
|
|
|
|
|
|
|
=head1 DESCRIPTION |
115
|
|
|
|
|
|
|
|
116
|
|
|
|
|
|
|
This subclass of Template Toolkit's C parses multilingual |
117
|
|
|
|
|
|
|
templates: templates that contain text in several languages. |
118
|
|
|
|
|
|
|
|
119
|
|
|
|
|
|
|
|
120
|
|
|
|
|
|
|
Hello! |
121
|
|
|
|
|
|
|
Bonjour ! |
122
|
|
|
|
|
|
|
|
123
|
|
|
|
|
|
|
|
124
|
|
|
|
|
|
|
Use this module directly if you have subclassed C, otherwise you |
125
|
|
|
|
|
|
|
may find it easier to use C. |
126
|
|
|
|
|
|
|
|
127
|
|
|
|
|
|
|
Language codes can be any string that matches C<\w+>, but we suggest |
128
|
|
|
|
|
|
|
sticking to ISO-639 which provides 2-letter codes for common languages |
129
|
|
|
|
|
|
|
and 3-letter codes for many others. |
130
|
|
|
|
|
|
|
|
131
|
|
|
|
|
|
|
=head1 METHODS |
132
|
|
|
|
|
|
|
|
133
|
|
|
|
|
|
|
=head2 new(\%params) |
134
|
|
|
|
|
|
|
|
135
|
|
|
|
|
|
|
The new() constructor creates and returns a reference to a new |
136
|
|
|
|
|
|
|
parser object. A reference to a hash may be supplied as a |
137
|
|
|
|
|
|
|
parameter to provide configuration values. |
138
|
|
|
|
|
|
|
|
139
|
|
|
|
|
|
|
Parser objects are typically provided as the C option |
140
|
|
|
|
|
|
|
to the C constructor. |
141
|
|
|
|
|
|
|
|
142
|
|
|
|
|
|
|
Configuration values are all valid C superclass |
143
|
|
|
|
|
|
|
options, and one specific to this class: |
144
|
|
|
|
|
|
|
|
145
|
|
|
|
|
|
|
=over |
146
|
|
|
|
|
|
|
|
147
|
|
|
|
|
|
|
=item LANGUAGE_VAR |
148
|
|
|
|
|
|
|
|
149
|
|
|
|
|
|
|
The LANGUAGE_VAR option can be used to set the name of the template |
150
|
|
|
|
|
|
|
variable which contains the current language. Defaults to |
151
|
|
|
|
|
|
|
I. |
152
|
|
|
|
|
|
|
|
153
|
|
|
|
|
|
|
my $parser = Template::Multilingual::Parser->new({ |
154
|
|
|
|
|
|
|
LANGUAGE_VAR => 'global.language', |
155
|
|
|
|
|
|
|
}); |
156
|
|
|
|
|
|
|
|
157
|
|
|
|
|
|
|
You will need to set this variable with the current language value |
158
|
|
|
|
|
|
|
at request time, usually in your C subclass' C |
159
|
|
|
|
|
|
|
method. |
160
|
|
|
|
|
|
|
|
161
|
|
|
|
|
|
|
=back |
162
|
|
|
|
|
|
|
|
163
|
|
|
|
|
|
|
=head2 parse($text) |
164
|
|
|
|
|
|
|
|
165
|
|
|
|
|
|
|
parse() is called by the Template Toolkit. It parses multilingual |
166
|
|
|
|
|
|
|
sections from the input text and translates them to Template Toolkit |
167
|
|
|
|
|
|
|
directives. The result is then passed to the C superclass. |
168
|
|
|
|
|
|
|
|
169
|
|
|
|
|
|
|
=head2 sections |
170
|
|
|
|
|
|
|
|
171
|
|
|
|
|
|
|
Returns a reference to an array of tokenized sections. Each section is a |
172
|
|
|
|
|
|
|
reference to hash with either a C key or a C key. |
173
|
|
|
|
|
|
|
|
174
|
|
|
|
|
|
|
A C key denotes text outside of any multilingual sections. The value |
175
|
|
|
|
|
|
|
is the text itself. |
176
|
|
|
|
|
|
|
|
177
|
|
|
|
|
|
|
A C key denotes text inside a multilingual section. The value is a |
178
|
|
|
|
|
|
|
reference to a hash, whose keys are language codes and values the corresponding |
179
|
|
|
|
|
|
|
text. For example, the following multilingual template: |
180
|
|
|
|
|
|
|
|
181
|
|
|
|
|
|
|
foo bonjourHello bar |
182
|
|
|
|
|
|
|
|
183
|
|
|
|
|
|
|
will parse to the following sections: |
184
|
|
|
|
|
|
|
|
185
|
|
|
|
|
|
|
[ { nolang => 'foo ' }, |
186
|
|
|
|
|
|
|
{ lang => { fr => 'bonjour', en => 'hello' } }, |
187
|
|
|
|
|
|
|
{ nolang => ' bar' }, |
188
|
|
|
|
|
|
|
] |
189
|
|
|
|
|
|
|
|
190
|
|
|
|
|
|
|
=head1 LANGUAGE SUBTAG HANDLING |
191
|
|
|
|
|
|
|
|
192
|
|
|
|
|
|
|
This module supports language subtags to express variants, e.g. "en_US" or "en-US". |
193
|
|
|
|
|
|
|
Here are the rules used for language matching: |
194
|
|
|
|
|
|
|
|
195
|
|
|
|
|
|
|
=over |
196
|
|
|
|
|
|
|
|
197
|
|
|
|
|
|
|
=item * |
198
|
|
|
|
|
|
|
|
199
|
|
|
|
|
|
|
Exact match: the current language is found in the template |
200
|
|
|
|
|
|
|
|
201
|
|
|
|
|
|
|
language template output |
202
|
|
|
|
|
|
|
fr foobar foo |
203
|
|
|
|
|
|
|
fr_CA foobar bar |
204
|
|
|
|
|
|
|
|
205
|
|
|
|
|
|
|
=item * |
206
|
|
|
|
|
|
|
|
207
|
|
|
|
|
|
|
Fallback to the primary language |
208
|
|
|
|
|
|
|
|
209
|
|
|
|
|
|
|
language template output |
210
|
|
|
|
|
|
|
fr_CA foobar foo |
211
|
|
|
|
|
|
|
|
212
|
|
|
|
|
|
|
=item * |
213
|
|
|
|
|
|
|
|
214
|
|
|
|
|
|
|
Fallback to first (in alphabetical order) other variant of the primary language |
215
|
|
|
|
|
|
|
|
216
|
|
|
|
|
|
|
language template output |
217
|
|
|
|
|
|
|
fr foobar bar |
218
|
|
|
|
|
|
|
fr_CA foobar bar |
219
|
|
|
|
|
|
|
|
220
|
|
|
|
|
|
|
=back |
221
|
|
|
|
|
|
|
|
222
|
|
|
|
|
|
|
=head1 AUTHOR |
223
|
|
|
|
|
|
|
|
224
|
|
|
|
|
|
|
Eric Cholet, C<< >> |
225
|
|
|
|
|
|
|
|
226
|
|
|
|
|
|
|
=head1 BUGS |
227
|
|
|
|
|
|
|
|
228
|
|
|
|
|
|
|
Multilingual text sections cannot be used inside TT directives. |
229
|
|
|
|
|
|
|
The following is illegal and will trigger a TT syntax error: |
230
|
|
|
|
|
|
|
|
231
|
|
|
|
|
|
|
[% title = "BonjourHello" %] |
232
|
|
|
|
|
|
|
|
233
|
|
|
|
|
|
|
Use this instead: |
234
|
|
|
|
|
|
|
|
235
|
|
|
|
|
|
|
[% title = BLOCK %]BonjourHello[% END %] |
236
|
|
|
|
|
|
|
|
237
|
|
|
|
|
|
|
|
238
|
|
|
|
|
|
|
The TAG_STYLE, START_TAG and END_TAG directives are supported, but the |
239
|
|
|
|
|
|
|
TAGS directive is not. |
240
|
|
|
|
|
|
|
|
241
|
|
|
|
|
|
|
Please report any bugs or feature requests to |
242
|
|
|
|
|
|
|
C, or through the web interface at |
243
|
|
|
|
|
|
|
L. |
244
|
|
|
|
|
|
|
I will be notified, and then you'll automatically be notified of progress on |
245
|
|
|
|
|
|
|
your bug as I make changes. |
246
|
|
|
|
|
|
|
|
247
|
|
|
|
|
|
|
=head1 SEE ALSO |
248
|
|
|
|
|
|
|
|
249
|
|
|
|
|
|
|
L |
250
|
|
|
|
|
|
|
|
251
|
|
|
|
|
|
|
ISO 639-2 Codes for the Representation of Names of Languages: |
252
|
|
|
|
|
|
|
http://www.loc.gov/standards/iso639-2/langcodes.html |
253
|
|
|
|
|
|
|
|
254
|
|
|
|
|
|
|
=head1 COPYRIGHT & LICENSE |
255
|
|
|
|
|
|
|
|
256
|
|
|
|
|
|
|
Copyright 2009 Eric Cholet, All Rights Reserved. |
257
|
|
|
|
|
|
|
|
258
|
|
|
|
|
|
|
This program is free software; you can redistribute it and/or modify it |
259
|
|
|
|
|
|
|
under the same terms as Perl itself. |
260
|
|
|
|
|
|
|
|
261
|
|
|
|
|
|
|
=cut |
262
|
|
|
|
|
|
|
|
263
|
|
|
|
|
|
|
1; # End of Template::Multilingual::Parser |