7 |
use vars qw($VERSION @ISA @EXPORT @EXPORT_OK); |
use vars qw($VERSION @ISA @EXPORT @EXPORT_OK); |
8 |
|
|
9 |
use Exporter; |
use Exporter; |
10 |
$VERSION = '0.01'; |
$VERSION = '0.02'; |
11 |
@ISA = ('Exporter'); |
@ISA = ('Exporter'); |
12 |
|
|
13 |
#@EXPORT = qw(); |
#@EXPORT = qw(); |
14 |
@EXPORT_OK = qw( |
@EXPORT_OK = qw( |
15 |
&alternatives |
&alternatives |
|
&load_affix |
|
16 |
); |
); |
17 |
|
|
18 |
#my @affix_regexp; |
my $debug=0; |
|
#my @affix_add; |
|
|
#my @affix_sub; |
|
|
|
|
|
my $debug=1; |
|
|
|
|
|
# stub |
|
19 |
|
|
20 |
|
# |
21 |
|
# make new instance of language, get args |
22 |
|
# |
23 |
sub new { |
sub new { |
24 |
my $class = shift; |
my $class = shift; |
25 |
my $self = {}; |
my $self = {}; |
26 |
bless($self, $class); |
bless($self, $class); |
27 |
$self->{ARGS} = {@_}; |
$self->{ARGS} = {@_}; |
28 |
|
$debug = $self->{ARGS}->{DEBUG}; |
29 |
|
@{$self->{affix_regexp}} = (); |
30 |
|
@{$self->{affix_add}} = (); |
31 |
|
@{$self->{affix_sub}} = (); |
32 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
33 |
} |
} |
34 |
|
|
78 |
sub nuke_s { |
sub nuke_s { |
79 |
my $tmp = $_[0]; |
my $tmp = $_[0]; |
80 |
return if (!$tmp); |
return if (!$tmp); |
81 |
$tmp=~s/^ *//g; |
# $tmp=~s/^\s+//g; |
82 |
$tmp=~s/ *$//g; |
# $tmp=~s/\s+$//g; |
83 |
$tmp=~s/ *//g; |
$tmp=~s/\s+//g; |
84 |
return $tmp; |
return $tmp; |
85 |
} |
} |
86 |
|
|
92 |
} |
} |
93 |
|
|
94 |
# |
# |
95 |
|
# function for reading raw findaffix output |
96 |
|
# |
97 |
|
|
98 |
|
sub load_findaffix { |
99 |
|
my $self = shift; |
100 |
|
my $filename = shift; |
101 |
|
|
102 |
|
print STDERR "reading findaffix output $filename\n" if ($debug); |
103 |
|
|
104 |
|
open (A,$filename) || die "Can't open findaffix output $filename: $!"; |
105 |
|
while(<A>) { |
106 |
|
chomp; |
107 |
|
my @line=split(m;/;,$_,4); |
108 |
|
if ($#line > 2) { |
109 |
|
push @{$self->{affix_regexp}},'.'; |
110 |
|
push @{$self->{affix_sub}},$line[0]; |
111 |
|
push @{$self->{affix_add}},$line[1]; |
112 |
|
} |
113 |
|
} |
114 |
|
return 1; |
115 |
|
} |
116 |
|
|
117 |
|
# |
118 |
# function which returns original word and all alternatives |
# function which returns original word and all alternatives |
119 |
# |
# |
120 |
|
|
128 |
my $regexp = $self->{affix_regexp}[$i]; |
my $regexp = $self->{affix_regexp}[$i]; |
129 |
my $add = $self->{affix_add}[$i]; |
my $add = $self->{affix_add}[$i]; |
130 |
my $sub = $self->{affix_sub}[$i]; |
my $sub = $self->{affix_sub}[$i]; |
131 |
|
print STDERR "r:'$regexp'\t-'",$sub||'',"'\t+'",$add||'',"'\n" if ($debug); |
132 |
next if length($word) < length($sub); |
next if length($word) < length($sub); |
133 |
my $tmp_word = $word; |
my $tmp_word = $word; |
134 |
if ($sub) { |
if ($sub) { |
141 |
} else { |
} else { |
142 |
$tmp_word = $word.$add; |
$tmp_word = $word.$add; |
143 |
} |
} |
144 |
|
print STDERR "\t ?:$tmp_word\n" if ($debug); |
145 |
if ($tmp_word =~ m/$regexp/ix) { |
if ($tmp_word =~ m/$regexp/ix) { |
146 |
# print "$word -> $tmp_word\t-$sub, +$add, regexp: $regexp\n"; |
# print "$word -> $tmp_word\t-$sub, +$add, regexp: $regexp\n"; |
147 |
push @out,lc($tmp_word); |
push @out,lc($tmp_word); |
151 |
return @out; |
return @out; |
152 |
} |
} |
153 |
|
|
154 |
|
# |
155 |
|
# function which return minimal word of all alternatives |
156 |
|
# |
157 |
|
|
158 |
|
sub minimal { |
159 |
|
my $self = shift; |
160 |
|
my @out; |
161 |
|
foreach my $word (@_) { |
162 |
|
my @alt = $self->alternatives($word); |
163 |
|
my $minimal = shift @alt; |
164 |
|
foreach (@alt) { |
165 |
|
$minimal=$_ if (length($_) < length($minimal)); |
166 |
|
} |
167 |
|
push @out,$minimal; |
168 |
|
} |
169 |
|
return @out; |
170 |
|
} |
171 |
|
|
172 |
############################################################################### |
############################################################################### |
173 |
1; |
1; |
174 |
__END__ |
__END__ |
175 |
|
|
176 |
=head1 NAME |
=head1 NAME |
177 |
|
|
178 |
Alternative.pm - see all alternatives of a given word in a given language |
Alternative.pm - alternative spelling of a given word in a given language |
179 |
|
|
180 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
181 |
|
|
182 |
use Lingua::Spelling:Alternative; |
use Lingua::Spelling::Alternative; |
183 |
|
|
184 |
my $en = new Alternative; |
my $en = new Lingua::Spelling::Alternative; |
185 |
$en->load_affix('/usr/lib/ispell/english.aff') or die $!; |
$en->load_affix('/usr/lib/ispell/default.aff') or die $!; |
186 |
print $en->alternatives("cars"); |
print join(" ",$en->alternatives("cars")),"\n"; |
187 |
|
|
188 |
=head1 DESCRIPTION |
=head1 DESCRIPTION |
189 |
|
|
190 |
This module is designed to return all valid forms of a given word |
This module is designed to return all forms of a given word |
191 |
(for example when you want to see all possible forms of some word |
(for example when you want to see all possible forms of some word |
192 |
entered in search engine) |
entered in search engine) which can be generated using affix file (from |
193 |
|
ispell) or using findaffix output file (also part of ispell package) |
194 |
|
|
195 |
=head1 PUBLIC METHODS |
=head1 PUBLIC METHODS |
196 |
|
|
198 |
|
|
199 |
=item new |
=item new |
200 |
|
|
201 |
The new() constructor (without parameters) create container for new |
The new() constructor (without parameters) create container for new language. |
202 |
language. |
Only parameter it supports is DEBUG which turns on (some) debugging output. |
203 |
|
|
204 |
=item load_affix |
=item load_affix |
205 |
|
|
206 |
Function load_affix loads ispell's affix file. |
Function load_affix() loads ispell's affix file for later usage. |
207 |
|
|
208 |
|
=item load_findaffix |
209 |
|
|
210 |
|
This function loads output of findaffix program from ispell package. |
211 |
|
This is better idea (if you are creating affix file for particular language |
212 |
|
yourself or you can get your hands on one) because affix file from ispell |
213 |
|
is limited to 26 entries (because each entry is denoted by single character). |
214 |
|
|
215 |
=item alternatives |
=item alternatives |
216 |
|
|
217 |
Function alternatives |
Function alternatives return array of all alternative spellings of particular |
218 |
|
word(s). It will also return spelling which are not correct if there is |
219 |
|
rule like that in affix file. |
220 |
|
|
221 |
|
=item minimal |
222 |
|
|
223 |
|
This function returns minimal of all alternatives of a given word(s). It's |
224 |
|
a poor man's version of normalize (because we don't know grammatic of |
225 |
|
particular language, just some spelling rules). |
226 |
|
|
227 |
|
=back |
228 |
|
|
229 |
=head1 PRIVATE METHODS |
=head1 PRIVATE METHODS |
230 |
|
|
237 |
|
|
238 |
=head1 BUGS |
=head1 BUGS |
239 |
|
|
240 |
There are no known bugs. |
There are no known bugs. If you find any, please report it in CPAN's |
241 |
|
request tracker at: http://rt.cpan.org/ |
242 |
|
|
243 |
=head1 CONTACT AND COPYRIGHT |
=head1 CONTACT AND COPYRIGHT |
244 |
|
|
245 |
Copyright 2002 Dobrica Pavlinusic (dpavlin@rot13.org). All |
Copyright 2002-2003 Dobrica Pavlinusic (dpavlin@rot13.org). All |
246 |
rights reserved. This program is free software; you can redistribute |
rights reserved. This program is free software; you can redistribute |
247 |
it and/or modify it under the same terms as Perl itself. |
it and/or modify it under the same terms as Perl itself. |
248 |
|
|