Fixed combining marks on Unicode regexes.

[DictionaryPC.git] / src / com / hughes / android / dictionary / parser / DictFileParser.java

diff --git a/src/com/hughes/android/dictionary/parser/DictFileParser.java b/src/com/hughes/android/dictionary/parser/DictFileParser.java
index 67ca43293cbdbf9bde04bd27de5ff03c7f55ce50..991ed8ad417b29e4941f68fb07e4a1eaf4b4248b 100644 (file)
--- a/src/com/hughes/android/dictionary/parser/DictFileParser.java
+++ b/src/com/hughes/android/dictionary/parser/DictFileParser.java
@@ -30,6 +30,7 @@ import java.util.regex.Matcher;
 import java.util.regex.Pattern;
 
 import com.hughes.android.dictionary.engine.DictionaryBuilder;
+import com.hughes.android.dictionary.engine.EntrySource;
 import com.hughes.android.dictionary.engine.IndexedEntry;
 import com.hughes.android.dictionary.engine.EntryTypeName;
 import com.hughes.android.dictionary.engine.IndexBuilder;
@@ -37,7 +38,7 @@ import com.hughes.android.dictionary.engine.Language;
 import com.hughes.android.dictionary.engine.PairEntry;
 import com.hughes.android.dictionary.engine.PairEntry.Pair;
 
-public class DictFileParser {
+public class DictFileParser implements Parser {
   
   static final Logger logger = Logger.getLogger(DictFileParser.class.getName());
 
@@ -49,17 +50,15 @@ public class DictFileParser {
   public static final Pattern PIPE = Pattern.compile("\\|");
   
   static final Pattern SPACES = Pattern.compile("\\s+");
-//  static final Pattern DE_NOUN = Pattern.compile("([^ ]+) *\\{(m|f|n|pl)\\}");
-//  static final Pattern EN_VERB = Pattern.compile("^to ([^ ]+)");
   
   static final Pattern BRACKETED = Pattern.compile("\\[([^]]+)\\]");
   static final Pattern PARENTHESIZED = Pattern.compile("\\(([^)]+)\\)");
   static final Pattern CURLY_BRACED = Pattern.compile("\\{([^}]+)\\}");
   
-  static final Pattern NON_CHAR_DASH = Pattern.compile("[^-'\\p{L}0-9]+");
-  public static final Pattern NON_CHAR = Pattern.compile("[^\\p{L}0-9]+");
+  static final Pattern NON_CHAR_DASH = Pattern.compile("[^-'\\p{L}\\p{M}\\p{N}]+");
+  public static final Pattern NON_CHAR = Pattern.compile("[^\\p{L}\\p{M}\\p{N}]+");
 
-  static final Pattern TRIM_PUNC = Pattern.compile("^[^\\p{L}0-9]+|[^\\p{L}0-9]+$");
+  static final Pattern TRIM_PUNC = Pattern.compile("^[^\\p{L}\\p{M}\\p{N}]+|[^\\p{L}\\p{M}\\p{N}]+$");
 
   final Charset charset;
   final boolean flipCols;
@@ -71,6 +70,8 @@ public class DictFileParser {
   final IndexBuilder[] langIndexBuilders;
   final IndexBuilder bothIndexBuilder;
   
+  EntrySource entrySource;
+  
   // final Set<String> alreadyDone = new HashSet<String>();
     
   public DictFileParser(final Charset charset, boolean flipCols,
@@ -86,11 +87,16 @@ public class DictFileParser {
     this.bothIndexBuilder = bothIndexBuilder;
   }
 
-  public void parseFile(final File file) throws IOException {
+  @Override
+  public void parse(final File file, final EntrySource entrySouce, final int pageLimit) throws IOException {
+    this.entrySource = entrySouce;
     final BufferedReader reader = new BufferedReader(new InputStreamReader(new FileInputStream(file), charset));
     String line;
     int count = 0;
     while ((line = reader.readLine()) != null) {
+      if (pageLimit >= 0 && count >= pageLimit) {
+        return;
+      }
       if (count % 10000 == 0) {
         logger.info("count=" + count + ", line=" + line);
       }
@@ -131,10 +137,20 @@ public class DictFileParser {
       subfields[1] = new String[] { fields[1] };
     }
     
-    final PairEntry pairEntry = new PairEntry();
+    final PairEntry pairEntry = new PairEntry(entrySource);
     for (int i = 0; i < subfields[0].length; ++i) {
       subfields[0][i] = subfields[0][i].trim();
       subfields[1][i] = subfields[1][i].trim();
+      if (subfields[0][i].length() == 0 && subfields[1][i].length() == 0) {
+        logger.warning("Empty pair: " + line);
+        continue;
+      }
+      if (subfields[0][i].length() == 0) {
+        subfields[0][i] = "__";
+      }
+      if (subfields[1][i].length() == 0) {
+        subfields[1][i] = "__";
+      }
       pairEntry.pairs.add(new Pair(subfields[0][i], subfields[1][i]));
     }
     final IndexedEntry entryData = new IndexedEntry(pairEntry);