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[][] {
58 {"AR", "TR" }, // Turkish
63 {"DE", "CA" }, // Catalan
64 {"DE", "CS" }, // Czech
65 {"DE", "EO" }, // Esperanto
69 {"DE", "HU" }, // Hungarian
72 {"DE", "LA" }, // Latin
73 {"DE", "NL" }, // Dutch
74 {"DE", "PL" }, // Polish
76 {"DE", "SV" }, // Swedish
77 {"DE", "TR" }, // Turkish
79 {"DE", "TA" }, // Tamil
81 {"ES", "RU" }, // Spanish-Russian
83 {"FR", "BG" }, // Bulgarian
84 {"FR", "CS" }, // Czech
90 {"FR", "NL" }, // Dutch
92 {"FR", "TR" }, // Turkish
97 {"IT", "EL" }, // Greek
102 {"IT", "LA" }, // Latin
103 {"IT", "LV" }, // Latvian
108 {"IT", "TR" }, // Turkish
127 {"PL", "FR" }, // Polish
128 {"PL", "RU" }, // Polish
129 {"PL", "HU" }, // Polish
130 {"PL", "ES" }, // Polish
132 {"TR", "EL" }, // Turkish, Greek
134 {"FA", "HY" }, // Persian, Armenian, by request.
135 {"FA", "SV" }, // Persian, Swedish, by request.
136 {"NL", "PL" }, // Dutch, Polish, by request.
144 static final Map<String,String> isoToDedication = new LinkedHashMap<String, String>();
146 isoToDedication.put("AF", "Afrikaans dictionary dedicated to Heiko and Mariëtte Horn.");
147 isoToDedication.put("HR", "Croatian dictionary dedicated to Ines Viskic and Miro Kresonja.");
148 isoToDedication.put("NL", "Dutch dictionary dedicated to Mike LeBeau.");
149 // German handled in file.
150 isoToDedication.put("EL", "Greek dictionary dedicated to Noah Egge.");
151 isoToDedication.put("IT", "Italian dictionary dedicated to Carolina Tropini, my favorite stardust in the whole universe! Ti amo!");
152 isoToDedication.put("KO", "Korean dictionary dedicated to Ande Elwood--fall fashion und Fernsehturms!");
153 isoToDedication.put("PT", "Portuguese dictionary dedicated to Carlos Melo, one Tough Mudder.");
154 isoToDedication.put("RO", "Romanian dictionary dedicated to Radu Teodorescu.");
155 isoToDedication.put("RU", "Russian dictionary dedicated to Maxim Aronin--best friend always!.");
156 isoToDedication.put("SR", "Serbian dictionary dedicated to Filip Crnogorac--thanks for the honey.");
157 isoToDedication.put("ES", "Spanish dictionary made especially for Carolina Tropini! <3 XoXoXXXXX!");
158 isoToDedication.put("SV", "Swedish dictionary dedicated to Kajsa Palmblad--björn kramar!");
160 private static String getDedication(String iso) {
161 return isoToDedication.containsKey(iso) ? "\n\n" + isoToDedication.get(iso) : "";
164 static final Map<String,String> isoToStoplist = new LinkedHashMap<String, String>();
166 isoToStoplist.put("DE", "de.txt");
167 isoToStoplist.put("EN", "en.txt");
168 isoToStoplist.put("ES", "es.txt");
169 isoToStoplist.put("IT", "it.txt");
170 isoToStoplist.put("FR", "fr.txt");
172 private static String getStoplist(String iso) {
173 return isoToStoplist.containsKey(iso) ? isoToStoplist.get(iso) : "empty.txt";
176 static String getOtherLang(final String[] pair, final String first) {
177 assert Arrays.asList(pair).contains(first);
178 assert pair.length == 2;
179 return pair[0].equals(first) ? pair[1] : pair[0];
182 static List<String> getMainArgs(final String[] pair) {
183 final List<String> result = new ArrayList<String>();
187 if (pair.length == 1) {
188 final String lang1 = pair[0];
189 final String dictFile = String.format("%s/%s.quickdic", OUTPUTS, lang1);
190 result.add(String.format("--dictOut=%s", dictFile));
191 result.add(String.format("--lang1=%s", lang1));
192 result.add(String.format("--lang1Stoplist=%s", STOPLISTS + getStoplist(lang1)));
193 result.add(String.format("--dictInfo=Wikitionary-based %s dictionary.", lang1));
196 final String wikiSplitFile = String.format("%s/wikiSplit/%s/%s.data", INPUTS, lang1.toLowerCase(), lang1);
197 if (new File(wikiSplitFile).canRead()) {
198 result.add(String.format("--input%d=%s", i, wikiSplitFile));
199 result.add(String.format("--input%dName=%s.wiktionary.org", i, lang1.toLowerCase()));
200 result.add(String.format("--input%dFormat=%s", i, WholeSectionToHtmlParser.NAME));
201 result.add(String.format("--input%dTitleIndex=%d", i, 1));
202 result.add(String.format("--input%dWiktionaryLang=%s", i, lang1));
203 result.add(String.format("--input%dSkipLang=%s", i, lang1));
204 result.add(String.format("--input%dWebUrlTemplate=http://%s.wiktionary.org/wiki/%%s", i, lang1.toLowerCase()));
205 //result.add(String.format("--input%dPageLimit=100", i));
208 System.err.println("Can't read file: " + wikiSplitFile);
211 if (lang1.equals("EN")) {
212 result.add(String.format("--input%d=%s/wikiSplit/en/%s.data", i, INPUTS, lang1));
213 result.add(String.format("--input%dName=ENWiktionary.%s", i, lang1)) ;
214 result.add(String.format("--input%dFormat=enwiktionary", i));
215 result.add(String.format("--input%dWiktionaryType=EnEnglish", i));
216 result.add(String.format("--input%dLangPattern=%s", i, "English"));
217 result.add(String.format("--input%dLangCodePattern=%s", i, lang1.toLowerCase()));
218 result.add(String.format("--input%dEnIndex=%d", i, 1));
219 //result.add(String.format("--input%dPageLimit=100", i));
224 } // Single-lang dictionaries.
226 final String lang1 = pair[0];
227 final String lang2 = pair[1];
229 final String dictFile = String.format("%s/%s-%s.quickdic",
230 OUTPUTS, lang1, lang2);
232 result.add(String.format("--dictOut=%s", dictFile));
233 result.add(String.format("--lang1Stoplist=%s", STOPLISTS + getStoplist(lang1)));
234 result.add(String.format("--lang2Stoplist=%s", STOPLISTS + getStoplist(lang2)));
236 // For a few langs, put the defs of the other language in DE/IT/FR using WholeSection.
237 for (final String wikitionaryLang : Arrays.asList("EN", "DE", "IT", "FR")) {
238 if (!Arrays.asList(pair).contains(wikitionaryLang)) {
241 final String foreignIso = getOtherLang(pair, wikitionaryLang);
242 final String wikiSplitFile = String.format("%s/wikiSplit/%s/%s.data", INPUTS, wikitionaryLang.toLowerCase(), foreignIso);
243 if (!new File(wikiSplitFile).canRead()) {
244 System.err.println("Can't read file: " + wikiSplitFile);
247 result.add(String.format("--input%d=%s", i, wikiSplitFile));
248 result.add(String.format("--input%dName=%s.wiktionary.org", i, wikitionaryLang.toLowerCase()));
249 result.add(String.format("--input%dFormat=%s", i, WholeSectionToHtmlParser.NAME));
250 result.add(String.format("--input%dTitleIndex=%d", i, Arrays.asList(pair).indexOf(foreignIso) + 1));
251 result.add(String.format("--input%dWiktionaryLang=%s", i, wikitionaryLang));
252 result.add(String.format("--input%dSkipLang=%s", i, foreignIso));
253 result.add(String.format("--input%dWebUrlTemplate=http://%s.wiktionary.org/wiki/%%s", i, wikitionaryLang.toLowerCase()));
257 // Deal with the pairs where one is English.
258 if (Arrays.asList(pair).contains("EN")) {
259 final String foreignIso = getOtherLang(pair, "EN");
261 String foreignRegex = WiktionaryLangs.isoCodeToEnWikiName.get(foreignIso);
262 if (foreignIso.equals("ZH")) {
263 // HACK: The missing "e" prevents a full match, causing "Cantonese" to be appended to the entries.
264 foreignRegex = "Chinese|Mandarin|Cantones";
267 result.add(String.format("--lang1=%s", lang1));
268 result.add(String.format("--lang2=%s", lang2));
269 result.add(String.format("--dictInfo=(EN)Wikitionary-based EN-%s dictionary.%s", foreignIso, getDedication(foreignIso)));
271 result.add(String.format("--input%d=%s/wikiSplit/en/%s.data", i, INPUTS, foreignIso));
272 result.add(String.format("--input%dName=ENWiktionary.%s", i, foreignIso)) ;
273 result.add(String.format("--input%dFormat=enwiktionary", i));
274 result.add(String.format("--input%dWiktionaryType=EnForeign", i));
275 result.add(String.format("--input%dLangPattern=%s", i, foreignRegex));
276 result.add(String.format("--input%dLangCodePattern=%s", i, foreignIso.toLowerCase()));
277 result.add(String.format("--input%dEnIndex=%d", i, Arrays.asList(pair).indexOf("EN") + 1));
280 result.add(String.format("--input%d=%swikiSplit/en/EN.data", i, INPUTS));
281 result.add(String.format("--input%dName=enwiktionary.english", i));
282 result.add(String.format("--input%dFormat=enwiktionary", i));
283 result.add(String.format("--input%dWiktionaryType=EnToTranslation", i));
284 result.add(String.format("--input%dLangPattern=%s", i, foreignRegex));
285 result.add(String.format("--input%dLangCodePattern=%s", i, foreignIso.toLowerCase()));
286 result.add(String.format("--input%dEnIndex=%d", i, Arrays.asList(pair).indexOf("EN") + 1));
289 if (foreignIso.equals("DE")) {
290 result.add(String.format("--input%d=%sde-en_chemnitz.txt", i, INPUTS));
291 result.add(String.format("--input%dName=chemnitz", i));
292 result.add(String.format("--input%dCharset=UTF8", i));
293 result.add(String.format("--input%dFormat=chemnitz", i));
298 // Pairs without English.
299 result.add(String.format("--lang1=%s", lang1));
300 result.add(String.format("--lang2=%s", lang2));
301 result.add(String.format("--dictInfo=Wikitionary-based %s-%s dictionary.", lang1, lang2));
303 result.add(String.format("--input%d=%swikiSplit/en/EN.data", i, INPUTS));
304 result.add(String.format("--input%dName=BETA!enwiktionary.%s-%s", i, lang1, lang2));
305 result.add(String.format("--input%dFormat=%s", i, EnTranslationToTranslationParser.NAME));
306 result.add(String.format("--input%dLangPattern1=%s", i, lang1));
307 result.add(String.format("--input%dLangPattern2=%s", i, lang2));
314 public static void main(final String[] args) throws Exception {
316 final List<String[]> allPairs = new ArrayList<String[]>();
318 allPairs.addAll(Arrays.asList(nonEnPairs));
319 // Add all the EN-XX pairs.
320 for (final String isoCode : WiktionaryLangs.isoCodeToEnWikiName.keySet()) {
321 allPairs.add(new String[] {"EN", isoCode});
325 final Set<List<String>> done = new LinkedHashSet<List<String>>();
326 for (final String[] pair : allPairs) {
328 final List<String> pairList = Arrays.asList(pair);
329 if (done.contains(pairList)) {
334 if (!pairList.contains("EN") && !pairList.contains("EL")) {
338 DictionaryBuilder.main(getMainArgs(pair).toArray(new String[0]));