aristoteles.pl

   1 #!/usr/bin/env perl
   2 #
   3 # aristoteles - a regex-based, just-for-fun Italian-to-Portuguese translator
   4 # Copyright (C) 2009 Davide Mancusi <arekfu@gmail.com>
   5 #
   6 # This program is free software: you can redistribute it and/or modify
   7 # it under the terms of the GNU General Public License as published by
   8 # the Free Software Foundation, either version 3 of the License, or
   9 # (at your option) any later version.
  10 #
  11 # This program is distributed in the hope that it will be useful,
  12 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  13 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14 # GNU General Public License for more details.
  15 #
  16 # You should have received a copy of the GNU General Public License
  17 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
  18
  19 use strict;
  20 use warnings;
  21 use Tie::IxHash;
  22
  23 binmode(STDOUT, ":utf8");
  24 binmode(DATA, ":utf8");
  25
  26 # Check the number of arguments
  27 die "Usage: ", $0, " <filename>\n" unless $#ARGV == 0;
  28
  29 # Read the pattern dictionary
  30 tie my %patdic, 'Tie::IxHash';
  31
  32 while( <DATA> ) {
  33         chomp;
  34         my ($search, $subst) = split /\t+/;
  35         $patdic{$search} = $subst;
  36 }
  37
  38 # Open file to translate
  39 open( my $infile, "<utf8", $ARGV[0] ) or die "Can't open input file: $!\n";
  40
  41 # Translate!
  42 while( <$infile> ) {
  43         foreach my $search (keys %patdic) {
  44                 eval "s/$search/$patdic{$search}/g";
  45                 if( $search =~ /^\\b([[:lower:]].*)/ )
  46                 {
  47                         my $search2 = '\b'.ucfirst($1);
  48                         my $subst2 = ucfirst($patdic{$search});
  49                         eval "s/$search2/$subst2/g";
  50                 }
  51         }
  52         print;
  53 }
  54
  55 __END__
  56 \bdove\b        onde
  57 acqua\b agua
  58 \blatte\b       leite
  59 mezzo   meio
  60 mezza   meia
  61 \bche\b que
  62 \bchi\b quem
  63 \bè\b  é
  64 \be'\b  é
  65 \bho\b  tenho
  66 \bhai?\b        tem
  67 \babbiamo\b     tem
  68 \bavete\b       tem
  69 \bhanno\b       tem
  70 \bcos(ì|i')\b  assim
  71 \bperch(é|e')\b        porque
  72 \bbuon(o)?\b    bom
  73 \bbuona\b       boa
  74 \bdestra\b      direita
  75 \bsinistra\b    esquerda
  76 \btutto\b       todo
  77 \btutta\b       toda
  78 \btutti\b       todos
  79 \btutte\b       todas
  80 \bquesto\b      este
  81 \bquesta\b      esta
  82 \bquel(lo)?\b   esse
  83 \bquella?\b     essa
  84 \bil\b  o
  85 \blo\b  o
  86 \bl'([[:alpha:]]+)(a|i|u)\b     a $1$2
  87 \bl'([[:alpha:]]+)(o|e)\b       o $1$2
  88 \bla\b          a
  89 \bi\b   os
  90 \bgli\b os
  91 \ble\b          as
  92 \b(di|a|da|in|con|su|per|fra|tra)\Wme\b $1 mim
  93 \b(di|a|da|in|con|su|per|fra|tra)\Wte\b $1 ti
  94 \bal(lo)?\b     ao
  95 \balla\b        à
  96 \ba(i|gli)\b    aos
  97 \balle\b        às
  98 \ball'([[:alpha:]]+)(a|i)\b     à $1$2
  99 \ball'([[:alpha:]]+)(o|e)\b     ao $1$2
 100 \bdal(lo)?\b    do
 101 \bdalla\b       da
 102 \bda(i|gli)\b   dos
 103 \bdalle\b       das
 104 \bda\b          de
 105 \bdall'([[:alpha:]]+)(a|i)\b    da $1$2
 106 \bdall'([[:alpha:]]+)(o|e|u)\b  do $1$2
 107 \bsul(lo)?\b    sobre o
 108 \bsulla\b       sobre a
 109 \bsu(i|gli)\b   sobre os
 110 \bsulle\b       sobre as
 111 \bsu\b          sobre
 112 \bdell'([[:alpha:]]+)(a|i)\b    da $1$2
 113 \bdell'([[:alpha:]]+)(o|e|u)\b  do $1$2
 114 \bdel(lo)?\b    do
 115 \bdella\b       da
 116 \bde(i|gli)\b   dos
 117 \bdelle\b       das
 118 \bdi\b          de
 119 \bdell'([[:alpha:]]+)(a|i)\b    da $1$2
 120 \bdell'([[:alpha:]]+)(o|e|u)\b  do $1$2
 121 \bnel(lo)?\b    no
 122 \bnella\b       na
 123 \bne(i|gli)\b   nos
 124 \bnelle\b       nas
 125 \bnell'([[:alpha:]]+)(a|i)\b    à $1$2
 126 \bnell'([[:alpha:]]+)(o|e)\b    ao $1$2
 127 \bcon\b         com
 128 \bcome\b                como
 129 \bsotto\b       debaixo
 130 \bnon?\b        não
 131 \bun'   uma
 132 \buno?\b        um
 133 \buna\b         uma
 134 \bdue\b         dois
 135 \btre\b         três
 136 \bquattro\b     quatro
 137 \bcinque\b      cinco
 138 \bsei\b seis
 139 \bsette\b       sete
 140 otto\b  oito
 141 \bdieci\b       dez
 142 \bundici\b      onze
 143 \bdodici\b      doze
 144 \btredici\b     treze
 145 \bquattordici\b catorze
 146 \bquindici\b    quinze
 147 \bsedici\b      dezasseis
 148 \bdiciassette\b dezassete
 149 \bdiciotto\b    dezoito
 150 \bdiciannove\b  dezanove
 151 \bper\b         para
 152 \bin\b          em
 153 \bpi(ù|u')\b   mais
 154 \bma\b          mas
 155 ssione\b        ção
 156 ssioni\b        ções
 157 sione\b ção
 158 sioni\b ções
 159 zione\b ção
 160 zioni\b ções
 161 ine\b   inhas
 162 one\b   ão
 163 ini\b   inhos
 164 oni\b   ãos
 165 ina\b   inha
 166 ona\b   oa
 167 ino\b   inho
 168 one\b   oas
 169 ico\b   igo
 170 (b|c|d|f|g|l|m|n|p|q|t|v|z)\1   $1
 171 ([^sl[:^alpha:]])ce\b   $1z
 172 sc(ia|e|i|io|iu)        c$1
 173 gli?(a|e|i|o|u) lh$1
 174 gni?(a|e|i|o|u) nh$1
 175 cchi    lh
 176 \bpi(a|e|o|u)   ch$1
 177 \bs(t|p|d|f|c|b)        es$1
 178 (a|e|i|o)(r|l)e\b       $1$2
 179 zi(a|e|o|u)     ci$1
 180 ol(o|a)\b       ul$1
 181 at(o|a)\b       ad$1
 182 tà\b   tade
 183 ta'\b   tade
 184 (a|e)nza\b      $1nça
 185 (a|e)nze\b      $1nças
 186 che\b   cas
 187 chi\b   cos
 188 uo\w    o
 189 ie\w    e