Build fr_de dictionary from enwiktionary, yeah!

[DictionaryPC.git] / src / com / hughes / android / dictionary / engine / DictionaryBuilder.java

diff --git a/src/com/hughes/android/dictionary/engine/DictionaryBuilder.java b/src/com/hughes/android/dictionary/engine/DictionaryBuilder.java
index 04b72b4d8265fd976e05f3d36c2bc11544a4b9a2..7aee4428655fc02335479cf77d617acf73dcba68 100644 (file)
--- a/src/com/hughes/android/dictionary/engine/DictionaryBuilder.java
+++ b/src/com/hughes/android/dictionary/engine/DictionaryBuilder.java
@@ -31,20 +31,22 @@ import javax.xml.parsers.ParserConfigurationException;
 import org.xml.sax.SAXException;
 
 import com.hughes.android.dictionary.parser.DictFileParser;
-import com.hughes.android.dictionary.parser.EnWiktionaryXmlParser;
+import com.hughes.android.dictionary.parser.Parser;
+import com.hughes.android.dictionary.parser.wiktionary.EnForeignParser;
+import com.hughes.android.dictionary.parser.wiktionary.EnToTranslationParser;
+import com.hughes.android.dictionary.parser.wiktionary.EnTranslationToTranslationParser;
 import com.hughes.util.Args;
 import com.hughes.util.FileUtil;
 
-
 public class DictionaryBuilder {
   
   public final Dictionary dictionary;
   public final List<IndexBuilder> indexBuilders = new ArrayList<IndexBuilder>();
   
-  public DictionaryBuilder(final String dictInfo, final Language lang0, final Language lang1, final String normalizerRules1, final String normalizerRules2, final Set<String> lang1Stoplist, final Set<String> lang2Stoplist) {
-    dictionary = new Dictionary(dictInfo);
-    indexBuilders.add(new IndexBuilder(this, lang0.getSymbol(), lang0.getSymbol() + "->" + lang1.getSymbol(), lang0, normalizerRules1, lang1Stoplist, false));
-    indexBuilders.add(new IndexBuilder(this, lang1.getSymbol(), lang1.getSymbol() + "->" + lang0.getSymbol(), lang1, normalizerRules2, lang2Stoplist, true));
+  public DictionaryBuilder(final String dictInfoString, final Language lang0, final Language lang1, final String normalizerRules1, final String normalizerRules2, final Set<String> lang1Stoplist, final Set<String> lang2Stoplist) {
+    dictionary = new Dictionary(dictInfoString);
+    indexBuilders.add(new IndexBuilder(this, lang0.getIsoCode(), lang0.getIsoCode() + "->" + lang1.getIsoCode(), lang0, normalizerRules1, lang1Stoplist, false));
+    indexBuilders.add(new IndexBuilder(this, lang1.getIsoCode(), lang1.getIsoCode() + "->" + lang0.getIsoCode(), lang1, normalizerRules2, lang2Stoplist, true));
   }
   
   void build() {
@@ -55,13 +57,18 @@ public class DictionaryBuilder {
   }
   
   public static void main(final String[] args) throws IOException, ParserConfigurationException, SAXException {
+    System.out.println("Running with arguments:");
+    for (final String arg : args) {
+      System.out.println(arg);
+    }
+    
     final Map<String,String> keyValueArgs = Args.keyValueArgs(args);
     
-    final Language lang1 = Language.lookup(keyValueArgs.remove("lang1"));
-    final Language lang2 = Language.lookup(keyValueArgs.remove("lang2"));
-    if (lang1 == null || lang2 == null) {
+    if (!keyValueArgs.containsKey("lang1") || !keyValueArgs.containsKey("lang2")) {
       fatalError("--lang1= and --lang2= must both be specified.");
     }
+    final Language lang1 = Language.lookup(keyValueArgs.remove("lang1"));
+    final Language lang2 = Language.lookup(keyValueArgs.remove("lang2"));
 
     final Set<String> lang1Stoplist = new LinkedHashSet<String>();
     final Set<String> lang2Stoplist = new LinkedHashSet<String>();
@@ -121,29 +128,47 @@ public class DictionaryBuilder {
         if (inputName == null) {
           fatalError("Must specify human readable name for: " + prefix + "Name");
         }
+        String pageLimitString = keyValueArgs.remove(prefix + "PageLimit");
+        if (pageLimitString == null) {
+          pageLimitString = "-1";
+        }
+        final int pageLimit = Integer.parseInt(pageLimitString);
 
-        final EntrySource entrySource = new EntrySource(dictionaryBuilder.dictionary.sources.size(), inputName, dictionaryBuilder.dictionary.pairEntries.size());
+        final EntrySource entrySource = new EntrySource(dictionaryBuilder.dictionary.sources.size(), inputName, 0);
         System.out.println("");
         
         String inputFormat = keyValueArgs.remove(prefix + "Format");
-        if ("dictcc".equals(inputFormat)) {
-          new DictFileParser(charset, false, DictFileParser.TAB, null, dictionaryBuilder, dictionaryBuilder.indexBuilders.toArray(new IndexBuilder[0]), null).parseFile(file);
+        if ("tab_separated".equals(inputFormat)) {
+          final boolean flipColumns = "true".equals(keyValueArgs.remove(prefix + "FlipColumns"));
+          new DictFileParser(charset, flipColumns, DictFileParser.TAB, null, dictionaryBuilder, dictionaryBuilder.indexBuilders.toArray(new IndexBuilder[0]), null).parse(file, entrySource, pageLimit);
         } else if ("chemnitz".equals(inputFormat)) {
-          new DictFileParser(charset, false, DictFileParser.DOUBLE_COLON, DictFileParser.PIPE, dictionaryBuilder, dictionaryBuilder.indexBuilders.toArray(new IndexBuilder[0]), null).parseFile(file);
+          final boolean flipColumns = "true".equals(keyValueArgs.remove(prefix + "FlipColumns"));
+          new DictFileParser(charset, flipColumns, DictFileParser.DOUBLE_COLON, DictFileParser.PIPE, dictionaryBuilder, dictionaryBuilder.indexBuilders.toArray(new IndexBuilder[0]), null).parse(file, entrySource, pageLimit);
         } else if ("enwiktionary".equals(inputFormat)) {
-          final Pattern langPattern = Pattern.compile(keyValueArgs.remove(prefix + "LangPattern"));
+          final String type = keyValueArgs.remove(prefix + "WiktionaryType");
+          final Pattern langPattern = Pattern.compile(keyValueArgs.remove(prefix + "LangPattern"), Pattern.CASE_INSENSITIVE);
           final Pattern langCodePattern = Pattern.compile(keyValueArgs.remove(prefix + "LangCodePattern"));
           final int enIndex = Integer.parseInt(keyValueArgs.remove(prefix + "EnIndex")) - 1;
-          String pageLimit = keyValueArgs.remove(prefix + "PageLimit");
-          if (pageLimit == null) {
-            pageLimit = "-1";
-          }
             
           if (enIndex < 0 || enIndex >= 2) {
             fatalError("Must be 1 or 2: " + prefix + "EnIndex");
           }
-          new EnWiktionaryXmlParser(dictionaryBuilder.indexBuilders.get(enIndex), dictionaryBuilder.indexBuilders.get(1-enIndex),
-              langPattern, langCodePattern, enIndex != 0).parse(file, Integer.parseInt(pageLimit));
+          final Parser parser;
+          if ("EnToTranslation".equals(type)) {
+            parser = new EnToTranslationParser(dictionaryBuilder.indexBuilders.get(enIndex), dictionaryBuilder.indexBuilders.get(1-enIndex),
+                langPattern, langCodePattern, enIndex != 0);
+          } else if ("EnForeign".equals(type)) {
+            parser = new EnForeignParser(dictionaryBuilder.indexBuilders.get(enIndex), dictionaryBuilder.indexBuilders.get(1-enIndex),
+                langPattern, langCodePattern, enIndex != 0);
+          } else {
+            fatalError("Invalid WiktionaryType (use EnToTranslation or EnForeign): " + type);
+            return;
+          }
+          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);
+          new EnTranslationToTranslationParser(dictionaryBuilder.indexBuilders, new Pattern[] {codePattern1, codePattern2}).parse(file, entrySource, pageLimit);
         } else {
           fatalError("Invalid or missing input format: " + inputFormat);
         }
@@ -176,6 +201,8 @@ public class DictionaryBuilder {
   
   private static void fatalError(String string) {
     System.err.println(string);
+    
+    
     System.exit(1);
   }