1 // Copyright 2011 Google Inc. All Rights Reserved.
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
7 // http://www.apache.org/licenses/LICENSE-2.0
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.
15 package com.hughes.android.dictionary.engine;
17 import com.hughes.android.dictionary.parser.wiktionary.EnTranslationToTranslationParser;
18 import com.hughes.android.dictionary.parser.wiktionary.WholeSectionToHtmlParser;
19 import com.hughes.android.dictionary.parser.wiktionary.WiktionaryLangs;
21 import junit.framework.TestCase;
24 import java.util.ArrayList;
25 import java.util.Arrays;
26 import java.util.LinkedHashMap;
27 import java.util.LinkedHashSet;
28 import java.util.List;
32 public class DictionaryBuilderMain extends TestCase {
34 static final String INPUTS = "data/inputs/";
35 static final String STOPLISTS = "data/inputs/stoplists/";
36 static final String OUTPUTS = "data/outputs/";
38 // Build the non EN ones.
39 static final String[][] nonEnPairs = new String[][] {
57 {"AR", "TR" }, // Turkish
62 {"DE", "CA" }, // Catalan
63 {"DE", "CS" }, // Czech
64 {"DE", "EO" }, // Esperanto
68 {"DE", "HU" }, // Hungarian
71 {"DE", "LA" }, // Latin
72 {"DE", "NL" }, // Dutch
73 {"DE", "PL" }, // Polish
75 {"DE", "SV" }, // Swedish
76 {"DE", "TR" }, // Turkish
78 {"DE", "TA" }, // Tamil
80 {"ES", "RU" }, // Spanish-Russian
82 {"FR", "BG" }, // Bulgarian
83 {"FR", "CS" }, // Czech
89 {"FR", "NL" }, // Dutch
91 {"FR", "TR" }, // Turkish
96 {"IT", "EL" }, // Greek
101 {"IT", "LA" }, // Latin
102 {"IT", "LV" }, // Latvian
107 {"IT", "TR" }, // Turkish
126 {"PL", "FR" }, // Polish
127 {"PL", "RU" }, // Polish
128 {"PL", "HU" }, // Polish
129 {"PL", "ES" }, // Polish
131 {"TR", "EL" }, // Turkish, Greek
133 {"FA", "HY" }, // Persian, Armenian, by request.
134 {"FA", "SV" }, // Persian, Swedish, by request.
135 {"NL", "PL" }, // Dutch, Polish, by request.
141 static final Map<String,String> isoToDedication = new LinkedHashMap<String, String>();
143 isoToDedication.put("AF", "Afrikaans dictionary dedicated to Heiko and Mariëtte Horn.");
144 isoToDedication.put("HR", "Croatian dictionary dedicated to Ines Viskic and Miro Kresonja.");
145 isoToDedication.put("NL", "Dutch dictionary dedicated to Mike LeBeau.");
146 // German handled in file.
147 isoToDedication.put("EL", "Greek dictionary dedicated to Noah Egge.");
148 isoToDedication.put("IT", "Italian dictionary dedicated to Carolina Tropini, my favorite stardust in the whole universe! Ti amo!");
149 isoToDedication.put("KO", "Korean dictionary dedicated to Ande Elwood--fall fashion und Fernsehturms!");
150 isoToDedication.put("PT", "Portuguese dictionary dedicated to Carlos Melo, one Tough Mudder.");
151 isoToDedication.put("RO", "Romanian dictionary dedicated to Radu Teodorescu.");
152 isoToDedication.put("RU", "Russian dictionary dedicated to Maxim Aronin--best friend always!.");
153 isoToDedication.put("SR", "Serbian dictionary dedicated to Filip Crnogorac--thanks for the honey.");
154 isoToDedication.put("ES", "Spanish dictionary made especially for Carolina Tropini! <3 XoXoXXXXX!");
155 isoToDedication.put("SV", "Swedish dictionary dedicated to Kajsa Palmblad--björn kramar!");
157 private static String getDedication(String iso) {
158 return isoToDedication.containsKey(iso) ? "\n\n" + isoToDedication.get(iso) : "";
161 static final Map<String,String> isoToStoplist = new LinkedHashMap<String, String>();
163 isoToStoplist.put("DE", "de.txt");
164 isoToStoplist.put("EN", "en.txt");
165 isoToStoplist.put("ES", "es.txt");
166 isoToStoplist.put("IT", "it.txt");
167 isoToStoplist.put("FR", "fr.txt");
169 private static String getStoplist(String iso) {
170 return isoToStoplist.containsKey(iso) ? isoToStoplist.get(iso) : "empty.txt";
173 static String getOtherLang(final String[] pair, final String first) {
174 assert Arrays.asList(pair).contains(first);
175 assert pair.length == 2;
176 return pair[0].equals(first) ? pair[1] : pair[0];
179 static List<String> getMainArgs(final String[] pair) {
180 final List<String> result = new ArrayList<String>();
184 if (pair.length == 1) {
185 final String lang1 = pair[0];
186 final String dictFile = String.format("%s/%s.quickdic", OUTPUTS, lang1);
187 result.add(String.format("--dictOut=%s", dictFile));
188 result.add(String.format("--lang1=%s", lang1));
189 result.add(String.format("--lang1Stoplist=%s", STOPLISTS + getStoplist(lang1)));
190 result.add(String.format("--dictInfo=Wikitionary-based %s dictionary.", lang1));
193 final String wikiSplitFile = String.format("%s/wikiSplit/%s/%s.data", INPUTS, lang1.toLowerCase(), lang1);
194 if (new File(wikiSplitFile).canRead()) {
195 result.add(String.format("--input%d=%s", i, wikiSplitFile));
196 result.add(String.format("--input%dName=%s.wiktionary.org", i, lang1.toLowerCase()));
197 result.add(String.format("--input%dFormat=%s", i, WholeSectionToHtmlParser.NAME));
198 result.add(String.format("--input%dTitleIndex=%d", i, 1));
199 result.add(String.format("--input%dWiktionaryLang=%s", i, lang1));
200 result.add(String.format("--input%dSkipLang=%s", i, lang1));
201 result.add(String.format("--input%dWebUrlTemplate=http://%s.wiktionary.org/wiki/%%s", i, lang1.toLowerCase()));
202 //result.add(String.format("--input%dPageLimit=100", i));
205 System.err.println("Can't read file: " + wikiSplitFile);
208 if (lang1.equals("EN")) {
209 result.add(String.format("--input%d=%s/wikiSplit/en/%s.data", i, INPUTS, lang1));
210 result.add(String.format("--input%dName=ENWiktionary.%s", i, lang1)) ;
211 result.add(String.format("--input%dFormat=enwiktionary", i));
212 result.add(String.format("--input%dWiktionaryType=EnEnglish", i));
213 result.add(String.format("--input%dLangPattern=%s", i, "English"));
214 result.add(String.format("--input%dLangCodePattern=%s", i, lang1.toLowerCase()));
215 result.add(String.format("--input%dEnIndex=%d", i, 1));
216 //result.add(String.format("--input%dPageLimit=100", i));
221 } // Single-lang dictionaries.
223 final String lang1 = pair[0];
224 final String lang2 = pair[1];
226 final String dictFile = String.format("%s/%s-%s.quickdic",
227 OUTPUTS, lang1, lang2);
229 result.add(String.format("--dictOut=%s", dictFile));
230 result.add(String.format("--lang1Stoplist=%s", STOPLISTS + getStoplist(lang1)));
231 result.add(String.format("--lang2Stoplist=%s", STOPLISTS + getStoplist(lang2)));
233 // For a few langs, put the defs of the other language in DE/IT/FR using WholeSection.
234 for (final String wikitionaryLang : Arrays.asList("EN", "DE", "IT", "FR")) {
235 if (!Arrays.asList(pair).contains(wikitionaryLang)) {
238 final String foreignIso = getOtherLang(pair, wikitionaryLang);
239 final String wikiSplitFile = String.format("%s/wikiSplit/%s/%s.data", INPUTS, wikitionaryLang.toLowerCase(), foreignIso);
240 if (!new File(wikiSplitFile).canRead()) {
241 System.err.println("Can't read file: " + wikiSplitFile);
244 result.add(String.format("--input%d=%s", i, wikiSplitFile));
245 result.add(String.format("--input%dName=%s.wiktionary.org", i, wikitionaryLang.toLowerCase()));
246 result.add(String.format("--input%dFormat=%s", i, WholeSectionToHtmlParser.NAME));
247 result.add(String.format("--input%dTitleIndex=%d", i, Arrays.asList(pair).indexOf(foreignIso) + 1));
248 result.add(String.format("--input%dWiktionaryLang=%s", i, wikitionaryLang));
249 result.add(String.format("--input%dSkipLang=%s", i, foreignIso));
250 result.add(String.format("--input%dWebUrlTemplate=http://%s.wiktionary.org/wiki/%%s", i, wikitionaryLang.toLowerCase()));
254 // Deal with the pairs where one is English.
255 if (Arrays.asList(pair).contains("EN")) {
256 final String foreignIso = getOtherLang(pair, "EN");
258 String foreignRegex = WiktionaryLangs.isoCodeToEnWikiName.get(foreignIso);
259 if (foreignIso.equals("ZH")) {
260 // HACK: The missing "e" prevents a full match, causing "Cantonese" to be appended to the entries.
261 foreignRegex = "Chinese|Mandarin|Cantones";
264 result.add(String.format("--lang1=%s", lang1));
265 result.add(String.format("--lang2=%s", lang2));
266 result.add(String.format("--dictInfo=(EN)Wikitionary-based EN-%s dictionary.%s", foreignIso, getDedication(foreignIso)));
268 result.add(String.format("--input%d=%s/wikiSplit/en/%s.data", i, INPUTS, foreignIso));
269 result.add(String.format("--input%dName=ENWiktionary.%s", i, foreignIso)) ;
270 result.add(String.format("--input%dFormat=enwiktionary", i));
271 result.add(String.format("--input%dWiktionaryType=EnForeign", i));
272 result.add(String.format("--input%dLangPattern=%s", i, foreignRegex));
273 result.add(String.format("--input%dLangCodePattern=%s", i, foreignIso.toLowerCase()));
274 result.add(String.format("--input%dEnIndex=%d", i, Arrays.asList(pair).indexOf("EN") + 1));
277 result.add(String.format("--input%d=%swikiSplit/en/EN.data", i, INPUTS));
278 result.add(String.format("--input%dName=enwiktionary.english", i));
279 result.add(String.format("--input%dFormat=enwiktionary", i));
280 result.add(String.format("--input%dWiktionaryType=EnToTranslation", i));
281 result.add(String.format("--input%dLangPattern=%s", i, foreignRegex));
282 result.add(String.format("--input%dLangCodePattern=%s", i, foreignIso.toLowerCase()));
283 result.add(String.format("--input%dEnIndex=%d", i, Arrays.asList(pair).indexOf("EN") + 1));
286 if (foreignIso.equals("DE")) {
287 result.add(String.format("--input%d=%sde-en_chemnitz.txt", i, INPUTS));
288 result.add(String.format("--input%dName=chemnitz", i));
289 result.add(String.format("--input%dCharset=UTF8", i));
290 result.add(String.format("--input%dFormat=chemnitz", i));
295 // Pairs without English.
296 result.add(String.format("--lang1=%s", lang1));
297 result.add(String.format("--lang2=%s", lang2));
298 result.add(String.format("--dictInfo=Wikitionary-based %s-%s dictionary.", lang1, lang2));
300 result.add(String.format("--input%d=%swikiSplit/en/EN.data", i, INPUTS));
301 result.add(String.format("--input%dName=BETA!enwiktionary.%s-%s", i, lang1, lang2));
302 result.add(String.format("--input%dFormat=%s", i, EnTranslationToTranslationParser.NAME));
303 result.add(String.format("--input%dLangPattern1=%s", i, lang1));
304 result.add(String.format("--input%dLangPattern2=%s", i, lang2));
311 public static void main(final String[] args) throws Exception {
313 final List<String[]> allPairs = new ArrayList<String[]>();
315 allPairs.addAll(Arrays.asList(nonEnPairs));
316 // Add all the EN-XX pairs.
317 for (final String isoCode : WiktionaryLangs.isoCodeToEnWikiName.keySet()) {
318 if (!isoCode.equals("EN")) {
319 allPairs.add(new String[] {"EN", isoCode});
324 final Set<List<String>> done = new LinkedHashSet<List<String>>();
325 for (final String[] pair : allPairs) {
327 final List<String> pairList = Arrays.asList(pair);
328 if (done.contains(pairList)) {
333 if (!pairList.contains("EN") && !pairList.contains("EL")) {
337 DictionaryBuilder.main(getMainArgs(pair).toArray(new String[0]));