diff --git a/lucene/analysis/kuromoji/build.xml b/lucene/analysis/kuromoji/build.xml
index 7afa31d3e92..2045d42b329 100644
--- a/lucene/analysis/kuromoji/build.xml
+++ b/lucene/analysis/kuromoji/build.xml
@@ -31,19 +31,32 @@
   <property name="ivy.default.configuration" value="default"/>
   <import file="../analysis-module-build.xml"/> 
 
-  <!-- default configuration: uses mecab-ipadic -->
+  <!-- default configuration: uses mecab-ipadic
   <property name="ipadic.type" value="ipadic"/>
   <property name="ipadic.version" value="mecab-ipadic-2.7.0-20070801" />
+  -->
 
   <!-- alternative configuration: uses mecab-naist-jdic
   <property name="ipadic.type" value="naist"/>
   <property name="ipadic.version" value="mecab-naist-jdic-0.6.3b-20111013" />
   -->
-  
+
+  <!-- alternative configuration: uses UniDic -->
+  <property name="ipadic.version" value="unidic-mecab-2.1.2_src"/>
+
+  <!--
   <property name="dict.src.file" value="${ipadic.version}.tar.gz" />
+  -->
+  <property name="dict.src.file" value="${ipadic.version}.zip" />
   <property name="dict.src.dir" value="${build.dir}/${ipadic.version}" />
+  <property name="dict.local.dir" value="/Users/johtani/tmp/dictionary" />
+  <!--
   <property name="dict.encoding" value="euc-jp"/>
   <property name="dict.format" value="ipadic"/>
+  -->
+  <property name="dict.encoding" value="utf-8"/>
+  <property name="dict.format" value="unidic"/>
+
   <property name="dict.normalize" value="false"/>
   <property name="dict.target.dir" location="${resources.dir}"/>
 
@@ -60,10 +73,15 @@
 
   <target name="compile-core" depends="jar-analyzers-common, common.compile-core" />
   <target name="download-dict" depends="ivy-availability-check,ivy-fail,ivy-configure" unless="dict.available">
-     <ivy:retrieve pattern="${build.dir}/${dict.src.file}" conf="${ipadic.type}" symlink="${ivy.symlink}"/>
-     <!-- TODO: we should checksum too -->
-     <gunzip src="${build.dir}/${dict.src.file}"/>
-     <untar src="${build.dir}/${ipadic.version}.tar" dest="${build.dir}"/>
+    <!--
+    <ivy:retrieve pattern="${build.dir}/${dict.src.file}" conf="${ipadic.type}" symlink="${ivy.symlink}"/>
+    -->
+    <!-- TODO: we should checksum too -->
+    <copy file="${dict.local.dir}/${dict.src.file}" tofile="${build.dir}/${dict.src.file}"/>
+    <unzip src="${build.dir}/${dict.src.file}" dest="${build.dir}"/>
+    <!--
+    <untar src="${build.dir}/${ipadic.version}.tar" dest="${build.dir}"/>
+    -->
   </target>
 
   <target name="patch-dict" depends="download-dict">
@@ -71,7 +89,7 @@
            originalfile="${dict.src.dir}/Noun.proper.csv"/>
   </target>
 
-  <target name="build-dict" depends="compile, patch-dict">
+  <target name="build-dict" depends="compile">
     <sequential>
       <delete verbose="true">
         <fileset dir="${resources.dir}/org/apache/lucene/analysis/ja/dict" includes="**/*"/>
diff --git a/lucene/analysis/kuromoji/src/java/org/apache/lucene/analysis/ja/util/DictionaryBuilder.java b/lucene/analysis/kuromoji/src/java/org/apache/lucene/analysis/ja/util/DictionaryBuilder.java
index 5004b62b3ca..4df26864a0e 100644
--- a/lucene/analysis/kuromoji/src/java/org/apache/lucene/analysis/ja/util/DictionaryBuilder.java
+++ b/lucene/analysis/kuromoji/src/java/org/apache/lucene/analysis/ja/util/DictionaryBuilder.java
@@ -53,7 +53,7 @@ public class DictionaryBuilder {
         .build(inputDir)
         .write(outputDir);
 
-    new UnknownDictionaryBuilder(encoding)
+    new UnknownDictionaryBuilder(format, encoding)
         .build(inputDir)
         .write(outputDir);
 
diff --git a/lucene/analysis/kuromoji/src/java/org/apache/lucene/analysis/ja/util/TokenInfoDictionaryBuilder.java b/lucene/analysis/kuromoji/src/java/org/apache/lucene/analysis/ja/util/TokenInfoDictionaryBuilder.java
index 4bb8d59b9ef..371fcd64200 100644
--- a/lucene/analysis/kuromoji/src/java/org/apache/lucene/analysis/ja/util/TokenInfoDictionaryBuilder.java
+++ b/lucene/analysis/kuromoji/src/java/org/apache/lucene/analysis/ja/util/TokenInfoDictionaryBuilder.java
@@ -73,8 +73,10 @@ class TokenInfoDictionaryBuilder {
         while ((line = reader.readLine()) != null) {
           String[] entry = CSVUtil.parse(line);
 
-          if (entry.length < 13) {
+          if (this.format == DictionaryFormat.IPADIC && entry.length < 13) {
             throw new IllegalArgumentException("Entry in CSV is not valid (13 field values expected): " + line);
+          } else if (this.format == DictionaryFormat.UNIDIC && entry.length < 21) {
+            throw new IllegalArgumentException("Entry in CSV is not valid (21 field values expected): " + line);
           }
 
           lines.add(formatEntry(entry));
@@ -149,9 +151,10 @@ class TokenInfoDictionaryBuilder {
    * 3   - word cost
    * 4-9 - pos
    * 10  - base form reading
-   * 11  - base form
+   * 11  - lexeme - not used
    * 12  - surface form
    * 13  - surface reading
+   * 14  - orth form
    */
   
   private String[] formatEntry(String[] features) {
@@ -169,7 +172,7 @@ class TokenInfoDictionaryBuilder {
       features2[7] = features[7];
       features2[8] = features[8];
       features2[9] = features[9];
-      features2[10] = features[11];
+      features2[10] = features[14];
       
       // If the surface reading is non-existent, use surface form for reading and pronunciation.
       // This happens with punctuation in UniDic and there are possibly other cases as well
diff --git a/lucene/analysis/kuromoji/src/java/org/apache/lucene/analysis/ja/util/UnknownDictionaryBuilder.java b/lucene/analysis/kuromoji/src/java/org/apache/lucene/analysis/ja/util/UnknownDictionaryBuilder.java
index 4316add3a6c..3b530904ad7 100644
--- a/lucene/analysis/kuromoji/src/java/org/apache/lucene/analysis/ja/util/UnknownDictionaryBuilder.java
+++ b/lucene/analysis/kuromoji/src/java/org/apache/lucene/analysis/ja/util/UnknownDictionaryBuilder.java
@@ -28,14 +28,18 @@ import java.util.Comparator;
 import java.util.List;
 
 import org.apache.lucene.analysis.ja.dict.CharacterDefinition;
+import org.apache.lucene.analysis.ja.util.DictionaryBuilder.DictionaryFormat;
 
 class UnknownDictionaryBuilder {
   private static final String NGRAM_DICTIONARY_ENTRY = "NGRAM,5,5,-32768,記号,一般,*,*,*,*,*,*,*";
 
   private final String encoding;
+  private final DictionaryFormat format;
 
-  UnknownDictionaryBuilder(String encoding) {
+
+  UnknownDictionaryBuilder(DictionaryFormat format, String encoding) {
     this.encoding = encoding;
+    this.format = format;
   }
 
   public UnknownDictionaryWriter build(Path dir) throws IOException {
@@ -61,7 +65,12 @@ class UnknownDictionaryBuilder {
       while ((line = lineReader.readLine()) != null) {
         // note: unk.def only has 10 fields, it simplifies the writer to just append empty reading and pronunciation,
         // even though the unknown dictionary returns hardcoded null here.
-        final String[] parsed = CSVUtil.parse(line + ",*,*"); // Probably we don't need to validate entry
+        final String[] parsed;
+        if (this.format == DictionaryFormat.UNIDIC) {
+          parsed = CSVUtil.parse(line + ",*,*,*"); // UniDic needs one more column
+        } else {
+          parsed = CSVUtil.parse(line + ",*,*"); // Probably we don't need to validate entry
+        }
         lines.add(parsed);
       }
     }
diff --git a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/CharacterDefinition.dat b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/CharacterDefinition.dat
index 4b8bd4b5b84..397d601767f 100644
Binary files a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/CharacterDefinition.dat and b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/CharacterDefinition.dat differ
diff --git a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/ConnectionCosts.dat b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/ConnectionCosts.dat
index 7679f14d8e3..d99de6d25b2 100644
Binary files a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/ConnectionCosts.dat and b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/ConnectionCosts.dat differ
diff --git a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$buffer.dat b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$buffer.dat
index 09f1e46680b..3189486e969 100644
Binary files a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$buffer.dat and b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$buffer.dat differ
diff --git a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$fst.dat b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$fst.dat
index 9328c53ee38..a9e0c70b374 100644
Binary files a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$fst.dat and b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$fst.dat differ
diff --git a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$posDict.dat b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$posDict.dat
index e727d90c40a..a4e509fc4a4 100644
Binary files a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$posDict.dat and b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$posDict.dat differ
diff --git a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$targetMap.dat b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$targetMap.dat
index 13d09bc2eb6..ed34cacdeb8 100644
Binary files a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$targetMap.dat and b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/TokenInfoDictionary$targetMap.dat differ
diff --git a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/UnknownDictionary$buffer.dat b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/UnknownDictionary$buffer.dat
index 16f0a822cce..18a364deffa 100644
Binary files a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/UnknownDictionary$buffer.dat and b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/UnknownDictionary$buffer.dat differ
diff --git a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/UnknownDictionary$posDict.dat b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/UnknownDictionary$posDict.dat
index e709dccab5b..139bdf5e504 100644
Binary files a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/UnknownDictionary$posDict.dat and b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/UnknownDictionary$posDict.dat differ
diff --git a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/UnknownDictionary$targetMap.dat b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/UnknownDictionary$targetMap.dat
index e8db0b363b2..9299529eea4 100644
Binary files a/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/UnknownDictionary$targetMap.dat and b/lucene/analysis/kuromoji/src/resources/org/apache/lucene/analysis/ja/dict/UnknownDictionary$targetMap.dat differ