src/com/hughes/android/dictionary/engine/DictionaryBuilderMain.java

   1 // Copyright 2011 Google Inc. All Rights Reserved.
   2 //
   3 // Licensed under the Apache License, Version 2.0 (the "License");
   4 // you may not use this file except in compliance with the License.
   5 // You may obtain a copy of the License at
   6 //
   7 //     http://www.apache.org/licenses/LICENSE-2.0
   8 //
   9 // Unless required by applicable law or agreed to in writing, software
  10 // distributed under the License is distributed on an "AS IS" BASIS,
  11 // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  12 // See the License for the specific language governing permissions and
  13 // limitations under the License.
  14
  15 package com.hughes.android.dictionary.engine;
  16
  17 import java.io.File;
  18 import java.io.PrintWriter;
  19 import java.io.RandomAccessFile;
  20 import java.util.ArrayList;
  21 import java.util.Collections;
  22 import java.util.List;
  23
  24 import junit.framework.TestCase;
  25
  26 public class DictionaryBuilderMain extends TestCase {
  27
  28   static final String INPUTS = "../DictionaryData/inputs/";
  29   static final String STOPLISTS = "../DictionaryData/inputs/stoplists/";
  30   static final String OUTPUTS = "../DictionaryData/outputs/";
  31
  32   static class Lang {
  33     final String nameRegex;
  34     final String isoCode;
  35     final String wikiSplit;
  36     final String stoplistFile;
  37     public Lang(String nameRegex, String code, final String wikiSplit, final String stoplistFile) {
  38       this.nameRegex = nameRegex;
  39       this.isoCode = code;
  40       this.wikiSplit = wikiSplit;
  41       this.stoplistFile = stoplistFile;
  42     }
  43   }
  44
  45
  46   public static void main(final String[] args) throws Exception {
  47
  48     Lang[] langs1 = new Lang[] {
  49         new Lang("^English$", "EN", null, "en.txt"),
  50     };
  51     Lang[] langs2 = new Lang[] {
  52 //        new Lang("^.*Italian.*$", "IT", "italian.data", "it.txt"),
  53 //        new Lang("^.*French.*$", "FR", "french.data", "empty.txt"),
  54 //        new Lang("^.*Spanish.*$", "ES", "spanish.data", "es.txt"),
  55 //        new Lang("^.*Greek.*$", "EL", "greek.data", "el.txt"),
  56 //        new Lang("^.*Japanese.*$", "JA", "japanese.data", "empty.txt"),
  57 //        new Lang("^.*Chinese.*$|^.*Mandarin.*$", "ZH", "mandarin.data", "empty.txt"),
  58         new Lang("^.*Afrikaans.*$", "AF", "afrikaans.data", "empty.txt"),
  59         new Lang("^.*Arabic.*$", "AR", "".data, "empty.txt"),
  60         new Lang("^.*Hebrew.*$", "HE"),
  61         new Lang("^.*Hindi.*$", "HI"),
  62         new Lang("^.*Icelandic.*$", "IS"),
  63         new Lang("^.*Irish.*$", "GA"),
  64         new Lang("^.*Korean.*$", "KO"),
  65         new Lang("^.*Maori.*$", "MI"),
  66         new Lang("^.*Norwegian.*$", "NO"),
  67         new Lang("^.*Persian.*$", "FA"),
  68         new Lang("^.*Portuguese.*$", "PT"),
  69         new Lang("^.*Romanian.*$", "RO"),
  70         new Lang("^.*Russian.*$", "RU"),
  71         new Lang("^.*Sanskrit.*$", "SA"),
  72         new Lang("^.*Serbian.*$", "SR"),
  73         new Lang("^.*Swedish.*$", "SV"),
  74         new Lang("^.*Tajik.*$", "TG"),
  75         new Lang("^.*Thai.*$", "TH"),
  76         new Lang("^.*Tibetan.*$", "BO"),
  77         new Lang("^.*Turkish.*$", "TR"),
  78         new Lang("^.*Ukranian.*$", "UK"),
  79         new Lang("^.*Vietnamese.*$", "VI"),
  80         new Lang("^.*Welsh.*$", "CY"),
  81         new Lang("^.*Zulu.*$", "ZU"),
  82         new Lang("^.*Croation.*$", "HR"),
  83         new Lang("^.*Czech.*$", "CS"),
  84         new Lang("^.*Dutch.*$", "NL"),
  85         new Lang("^.*Finnish.*$", "FI"),
  86         /*
  87         new Lang("^German$", "DE"),
  88         new Lang("^Armenian$", "HY"),
  89         new Lang("^English$", "EN"),
  90         new Lang("^Kurdish$", "KU"),
  91         new Lang("^Lithuanian$", "LT"),
  92         new Lang("^Malay$", "MS"),
  93         new Lang("^Mongolian$", "MN"),
  94         new Lang("^Somali$", "SO"),
  95         new Lang("^Sudanese$", "SU"),
  96         new Lang("^Yiddish$", "YI"),
  97         */
  98     };
  99
 100     for (final Lang lang1 : langs1) {
 101       for (final Lang lang2 : langs2) {
 102         if (lang1.nameRegex.equals(lang2.nameRegex)) {
 103           continue;
 104         }
 105
 106         int enIndex = -1;
 107         Lang nonEnglish = null;
 108         if (lang2.isoCode.equals("EN")) {
 109           enIndex = 2;
 110           nonEnglish = lang1;
 111         }
 112         if (lang1.isoCode.equals("EN")) {
 113           enIndex = 1;
 114           nonEnglish = lang2;
 115         }
 116         assert nonEnglish != null;
 117
 118         final String dictFile = String.format(OUTPUTS + "/%s-%s_enwiktionary.quickdic", lang1.isoCode, lang2.isoCode);
 119         System.out.println("building dictFile: " + dictFile);
 120         DictionaryBuilder.main(new String[] {
 121             String.format("--dictOut=%s", dictFile),
 122             String.format("--lang1=%s", lang1.isoCode),
 123             String.format("--lang2=%s", lang2.isoCode),
 124             String.format("--lang1Stoplist=%s", STOPLISTS + lang1.stoplistFile),
 125             String.format("--lang2Stoplist=%s", STOPLISTS + lang2.stoplistFile),
 126             String.format("--dictInfo=(EN)Wikitionary-based %s-%s dictionary", lang1.isoCode, lang2.isoCode),
 127
 128             "--input2=" + INPUTS + "enWikiSplit/" + nonEnglish.wikiSplit,
 129             "--input2Name=enwiktionary." + nonEnglish.wikiSplit,
 130             "--input2Format=enwiktionary",
 131             "--input2LangPattern=" + nonEnglish.nameRegex,
 132             "--input2LangCodePattern=" + nonEnglish.isoCode.toLowerCase(),
 133             "--input2EnIndex=" + enIndex,
 134
 135             "--input3=" + INPUTS + "enWikiSplit/english.data",
 136             "--input3Name=enwiktionary.english",
 137             "--input3Format=enwiktionary",
 138             "--input3LangPattern=" + nonEnglish.nameRegex,
 139             "--input3LangCodePattern=" + (enIndex == 1 ? lang2.isoCode : lang1.isoCode).toLowerCase(),
 140             "--input3EnIndex=" + enIndex,
 141
 142         });
 143
 144         // Print the entries for diffing.
 145         final RandomAccessFile raf = new RandomAccessFile(new File(dictFile), "r");
 146         final Dictionary dict = new Dictionary(raf);
 147         final PrintWriter textOut = new PrintWriter(new File(dictFile + ".text"));
 148         final List<PairEntry> sorted = new ArrayList<PairEntry>(dict.pairEntries);
 149         Collections.sort(sorted);
 150         for (final PairEntry pairEntry : sorted) {
 151           textOut.println(pairEntry.getRawText(false));
 152         }
 153         textOut.close();
 154         raf.close();
 155
 156       }  // langs2
 157     }  // langs1
 158
 159     DictionaryBuilder.main(new String[] {
 160         "--dictOut=" + OUTPUTS + "DE-EN_all_free.quickdic",
 161         "--lang1=DE",
 162         "--lang2=EN",
 163         "--dictInfo=@" + INPUTS + "de-en_all_free.info",
 164
 165         "--input1=" + INPUTS + "de-en_chemnitz.txt",
 166         "--input1Name=chemnitz",
 167         "--input1Charset=UTF8",
 168         "--input1Format=chemnitz",
 169     });
 170
 171     DictionaryBuilder.main(new String[] {
 172         "--dictOut=" + OUTPUTS + "de-en_all.quickdic",
 173         "--lang1=DE",
 174         "--lang2=EN",
 175         "--dictInfo=@" + INPUTS + "de-en_all.info",
 176
 177         "--input2=" + INPUTS + "de-en_chemnitz.txt",
 178         "--input2Name=dictcc",
 179         "--input2Charset=UTF8",
 180         "--input2Format=chemnitz",
 181
 182         "--input3=" + INPUTS + "/NONFREE/de-en_dictcc.txt",
 183         "--input3Name=dictcc",
 184         "--input3Charset=UTF8",
 185         "--input3Format=dictcc",
 186     });
 187
 188   }
 189
 190 }