package com.hughes.android.dictionary.engine;
-import java.io.BufferedWriter;
import java.io.File;
-import java.io.FileWriter;
import java.io.IOException;
import java.io.PrintWriter;
import java.io.RandomAccessFile;
import java.util.List;
import com.hughes.android.dictionary.DictionaryInfo;
+import com.hughes.android.dictionary.parser.wiktionary.WiktionaryLangs;
public class CheckDictionariesMain {
// dictionaryInfoOut.println("# LANG_1\t%LANG_2\tFILENAME\tVERSION_CODE\tFILESIZE\tNUM_MAIN_WORDS_1\tNUM_MAIN_WORDS_2\tNUM_ALL_WORDS_1\tNUM_ALL_WORDS_2");
final File[] files = dictDir.listFiles();
+ final List<String> dictNames = new ArrayList<String>();
Arrays.sort(files);
for (final File dictFile : files) {
if (!dictFile.getName().endsWith("quickdic")) {
// Find the stats.
System.out.println("Stats...");
+ final String lang1 = WiktionaryLangs.isoCodeToWikiName.get(dictionaryInfo.indexInfos.get(0).shortName);
+ final String lang2 = WiktionaryLangs.isoCodeToWikiName.get(dictionaryInfo.indexInfos.get(1).shortName);
+ dictNames.add(String.format("%s-%s\n", lang1, lang2));
final String row = dictionaryInfo.append(new StringBuilder()).toString();
if (!zipFile.canRead()) {
System.err.println("Couldn't read zipfile: " + zipFile);
}
System.out.println(row + "\n");
+
dictionaryInfoOut.println(row);
dictionaryInfoOut.flush();
raf.close();
-
}
+ Collections.sort(dictNames);
+ System.out.println(dictNames.toString().replaceAll(",", " *"));
+
dictionaryInfoOut.close();
}
}
parser.parse(file, entrySource, pageLimit);
} else if (EnTranslationToTranslationParser.NAME.equals(inputFormat)) {
- final Pattern codePattern1 = Pattern.compile(keyValueArgs.remove(prefix + "LangPattern1"), Pattern.CASE_INSENSITIVE);
- final Pattern codePattern2 = Pattern.compile(keyValueArgs.remove(prefix + "LangPattern2"), Pattern.CASE_INSENSITIVE);
+ final String code1 = keyValueArgs.remove(prefix + "LangPattern1");
+ final String code2 = keyValueArgs.remove(prefix + "LangPattern2");
+ if (code1 == null || code2 == null) {
+ fatalError("Must specify LangPattern1 and LangPattern2.");
+ return;
+ }
+ final Pattern codePattern1 = Pattern.compile(code1, Pattern.CASE_INSENSITIVE);
+ final Pattern codePattern2 = Pattern.compile(code2, Pattern.CASE_INSENSITIVE);
new EnTranslationToTranslationParser(dictionaryBuilder.indexBuilders, new Pattern[] {codePattern1, codePattern2}).parse(file, entrySource, pageLimit);
} else {
fatalError("Invalid or missing input format: " + inputFormat);
package com.hughes.android.dictionary.engine;
+import java.util.Arrays;
import java.util.LinkedHashMap;
+import java.util.LinkedHashSet;
+import java.util.List;
import java.util.Map;
+import java.util.Set;
import junit.framework.TestCase;
+import com.hughes.android.dictionary.parser.wiktionary.EnTranslationToTranslationParser;
import com.hughes.android.dictionary.parser.wiktionary.WiktionaryLangs;
public class DictionaryBuilderMain extends TestCase {
static final String STOPLISTS = "data/inputs/stoplists/";
static final String OUTPUTS = "data/outputs/";
- public static void main(final String[] args) throws Exception {
+ public static <E> void main(final String[] args) throws Exception {
// Builds all the dictionaries it can, outputs list to a text file.
// HACK: The missing "e" prevents a full match, causing "Cantonese" to be appended to the entries.
isoToRegex.put("ZH", "Chinese|Mandarin|Cantones");
+
+ // Build the non EN ones.
+
+ final String[][] nonEnPairs = new String[][] {
+ {"AR", "DE" },
+ {"AR", "ES" },
+ {"AR", "FR" },
+ {"AR", "IT" },
+ {"AR", "JA" },
+ {"AR", "RU" },
+ {"AR", "ZH" },
+
+ {"DE", "FR" },
+ {"DE", "CA" }, // Catalan
+ {"DE", "CS" }, // Czech
+ {"DE", "EO" }, // Esperanto
+ {"DE", "ES" },
+ {"DE", "FR" },
+ {"DE", "HU" }, // Hungarian
+ {"DE", "IT" },
+ {"DE", "JA" },
+ {"DE", "LA" }, // Latin
+ {"DE", "PL" }, // Polish
+ {"DE", "RU" },
+ {"DE", "SV" }, // Swedish
+ {"DE", "ZH" },
+
+
+ {"FR", "BG" }, // Bulgarian
+ {"FR", "CS" }, // Czech
+ {"FR", "DE" },
+ {"FR", "ES" },
+ {"FR", "IT" },
+ {"FR", "JA" },
+ {"FR", "LA" },
+ {"FR", "NL" }, // Dutch
+ {"FR", "RU" },
+ {"FR", "ZH" },
+
+ {"IT", "DE" },
+ {"IT", "EL" }, // Greek
+ {"IT", "ES" },
+ {"IT", "FR" },
+ {"IT", "HU" },
+ {"IT", "JA" },
+ {"IT", "LA" }, // Latin
+ {"IT", "LV" }, // Latvian
+ {"IT", "NL" },
+ {"IT", "PL" },
+ {"IT", "RU" },
+ {"IT", "SV" },
+ {"IT", "ZH" },
+
+ {"JA", "ZH" },
+ {"JA", "AR" },
+
+ {"ZH", "AR" },
+ {"ZH", "DE" },
+ {"ZH", "ES" },
+ {"ZH", "FR" },
+ {"ZH", "IT" },
+
+
+ {"NO", "SV" },
+ {"NO", "FI" },
+ {"FI", "SV" },
+ {"AR", "HE" },
+ {"KO", "JA" },
+ {"KO", "ZH" },
+ };
+
+ final Set<List<String>> done = new LinkedHashSet<List<String>>();
+ for (final String[] pair : nonEnPairs) {
+ Arrays.sort(pair);
+ final List<String> pairList = Arrays.asList(pair);
+ if (done.contains(pairList)) {
+ continue;
+ }
+ done.add(pairList);
+
+ final String lang1 = pair[0];
+ final String lang2 = pair[1];
+
+ final String dictFile = String.format("%s/%s-%s_enwiktionary_BETA.quickdic",
+ OUTPUTS, lang1, lang2);
+ System.out.println("building dictFile: " + dictFile);
+
+ if (!isoToStoplist.containsKey(lang1)) {
+ isoToStoplist.put(lang1, "empty.txt");
+ }
+ if (!isoToStoplist.containsKey(lang2)) {
+ isoToStoplist.put(lang2, "empty.txt");
+ }
+
+ DictionaryBuilder.main(new String[] {
+ String.format("--dictOut=%s", dictFile),
+ String.format("--lang1=%s", lang1),
+ String.format("--lang2=%s", lang2),
+ String.format("--lang1Stoplist=%s", STOPLISTS + isoToStoplist.get(lang1)),
+ String.format("--lang2Stoplist=%s", STOPLISTS + isoToStoplist.get(lang2)),
+ String.format("--dictInfo=(EN)Wikitionary-based %s-%s dictionary.", lang1, lang2),
+
+ String.format("--input2=%swikiSplit/en/EN.data", INPUTS),
+ String.format("--input2Name=BETA!enwiktionary.%s-%s", lang1, lang2),
+ String.format("--input2Format=%s", EnTranslationToTranslationParser.NAME),
+ String.format("--input2LangPattern1=%s", lang1),
+ String.format("--input2LangPattern2=%s", lang2),
+ });
+ }
+ if (1==1) {
+ return;
+ }
+
+
+ // Now build the EN ones.
+
// isoToWikiName.keySet().retainAll(Arrays.asList("UK", "HR", "FI"));
//isoToWikiName.clear();
boolean go = true;
} // foreignIso
+ // Now special case German-English.
+
final String dictFile = String.format("%s/DE-EN_chemnitz_enwiktionary.quickdic", OUTPUTS);
DictionaryBuilder.main(new String[] {
"--dictOut=" + dictFile,
projects / DictionaryPC.git / commitdiff