1 |
# Documentation and Copyright exist after __END__ |
2 |
|
3 |
package Lingua::Spelling::Alternative; |
4 |
require 5.001; |
5 |
|
6 |
use strict; |
7 |
use vars qw($VERSION @ISA @EXPORT @EXPORT_OK); |
8 |
|
9 |
use Exporter; |
10 |
$VERSION = '0.01'; |
11 |
@ISA = ('Exporter'); |
12 |
|
13 |
#@EXPORT = qw(); |
14 |
@EXPORT_OK = qw( |
15 |
&alternatives |
16 |
); |
17 |
|
18 |
my $debug=0; |
19 |
|
20 |
# |
21 |
# make new instance of language, get args |
22 |
# |
23 |
sub new { |
24 |
my $class = shift; |
25 |
my $self = {}; |
26 |
bless($self, $class); |
27 |
$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; |
33 |
} |
34 |
|
35 |
|
36 |
# |
37 |
# load affix file in internal structures |
38 |
# |
39 |
|
40 |
sub load_affix { |
41 |
my $self = shift; |
42 |
my $filename = shift; |
43 |
|
44 |
my $suffixes=0; |
45 |
|
46 |
my ($regexp,$add,$sub); |
47 |
|
48 |
print STDERR "reading affix file $filename\n" if ($debug); |
49 |
|
50 |
open (A,$filename) || die "Can't open affix file $filename: $!"; |
51 |
while(<A>) { |
52 |
chomp; |
53 |
next if (/^#|^[\s\t\n\r]*$/); |
54 |
|
55 |
if (/^suffixes/i) { |
56 |
$suffixes++; |
57 |
next; |
58 |
} |
59 |
|
60 |
next if (! $suffixes); |
61 |
|
62 |
if (/^flag[\s\t]+\*{0,1}(.):/i) { |
63 |
undef $regexp; |
64 |
undef $add; |
65 |
undef $sub; |
66 |
next; |
67 |
} |
68 |
|
69 |
if (/^[\s\t]*([^>#]+)>[\s\t]+-([^\,\s\t]+),([^\s\t]+)/) { |
70 |
$regexp = $1; |
71 |
$add = $2; |
72 |
$sub = $3 if ($3 ne "-"); |
73 |
} elsif (/^[\s\t]*([^>#]+)>[\s\t]+([^\s\t\#]+)/) { |
74 |
$regexp = $1; |
75 |
$sub = $2; |
76 |
} |
77 |
|
78 |
sub nuke_s { |
79 |
my $tmp = $_[0]; |
80 |
return if (!$tmp); |
81 |
# $tmp=~s/^\s+//g; |
82 |
# $tmp=~s/\s+$//g; |
83 |
$tmp=~s/\s+//g; |
84 |
return $tmp; |
85 |
} |
86 |
|
87 |
push @{$self->{affix_regexp}},nuke_s($regexp); |
88 |
push @{$self->{affix_add}},nuke_s($add); |
89 |
push @{$self->{affix_sub}},nuke_s($sub); |
90 |
} |
91 |
return 1; |
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 |
119 |
# |
120 |
|
121 |
sub alternatives { |
122 |
my $self = shift; |
123 |
my @out; |
124 |
foreach my $word (@_) { |
125 |
push @out,$word; # save original word |
126 |
next if (length($word) < 3); # cludge: preskoci kratke |
127 |
for(my $i=0; $i<=$#{$self->{affix_regexp}}; $i++) { |
128 |
my $regexp = $self->{affix_regexp}[$i]; |
129 |
my $add = $self->{affix_add}[$i]; |
130 |
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); |
133 |
my $tmp_word = $word; |
134 |
if ($sub) { |
135 |
next if ($word !~ m/$sub$/i); |
136 |
if ($add) { |
137 |
$tmp_word =~ s/$sub$/$add/i; |
138 |
} else { |
139 |
$tmp_word =~ s/$sub$//i; |
140 |
} |
141 |
} else { |
142 |
$tmp_word = $word.$add; |
143 |
} |
144 |
print STDERR "\t ?:$tmp_word\n" if ($debug); |
145 |
if ($tmp_word =~ m/$regexp/ix) { |
146 |
# print "$word -> $tmp_word\t-$sub, +$add, regexp: $regexp\n"; |
147 |
push @out,lc($tmp_word); |
148 |
} |
149 |
} |
150 |
} |
151 |
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; |
174 |
__END__ |
175 |
|
176 |
=head1 NAME |
177 |
|
178 |
Alternative.pm - alternative spelling of a given word in a given language |
179 |
|
180 |
=head1 SYNOPSIS |
181 |
|
182 |
use Lingua::Spelling::Alternative; |
183 |
|
184 |
my $en = new Lingua::Spelling::Alternative; |
185 |
$en->load_affix('/usr/lib/ispell/default.aff') or die $!; |
186 |
print join(" ",$en->alternatives("cars")),"\n"; |
187 |
|
188 |
=head1 DESCRIPTION |
189 |
|
190 |
This module is designed to return all valid forms of a given word |
191 |
(for example when you want to see all possible forms of some word |
192 |
entered in search engine) |
193 |
|
194 |
=head1 PUBLIC METHODS |
195 |
|
196 |
=over 4 |
197 |
|
198 |
=item new |
199 |
|
200 |
The new() constructor (without parameters) create container for new language. |
201 |
Only parametar it supports is DEBUG which turns on (some) debugging output. |
202 |
|
203 |
=item load_affix |
204 |
|
205 |
Function load_affix() loads ispell's affix file for later usage. |
206 |
|
207 |
=item load_findaffix |
208 |
|
209 |
This function loads output of findaffix program from ispell package. |
210 |
This is better idea (if you are creating affix file for particular language |
211 |
yourself or you can get your hands on one) because affix file from ispel |
212 |
is limited to 26 entries (because each entry is denoted by single character). |
213 |
|
214 |
=item alternatives |
215 |
|
216 |
Function alternatives return array of all alternative spellings of particular |
217 |
word(s). It will also return spelling which are not correct if there is |
218 |
rule like that in affix file. |
219 |
|
220 |
=item minimal |
221 |
|
222 |
This function returns minimal of all alternatives of a given word(s). It's |
223 |
a poor man's version of normalize (because we don't know gramatic of |
224 |
particular language, just some spelling rules). |
225 |
|
226 |
=back |
227 |
|
228 |
=head1 PRIVATE METHODS |
229 |
|
230 |
Documented as being not documented. |
231 |
|
232 |
=head1 EXAMPLES |
233 |
|
234 |
Please see the test.pl program in distribution which exercises some |
235 |
aspects of Alternative.pm. |
236 |
|
237 |
=head1 BUGS |
238 |
|
239 |
There are no known bugs. If you find any, please report it in CPAN's |
240 |
request tracker at: http://rt.cpan.org/ |
241 |
|
242 |
=head1 CONTACT AND COPYRIGHT |
243 |
|
244 |
Copyright 2002-2003 Dobrica Pavlinusic (dpavlin@rot13.org). All |
245 |
rights reserved. This program is free software; you can redistribute |
246 |
it and/or modify it under the same terms as Perl itself. |
247 |
|
248 |
=cut |