diff --git a/.gitignore b/.gitignore
index 7dfdb96..36b7417 100644
--- a/.gitignore
+++ b/.gitignore
@@ -3,6 +3,7 @@ out/
 temp/
 NlpParser/*
 !NlpParser/Nlp.g4
+!NlpParser/NlpRunner.java
 !NlpParser/testbench.txt
 !NlpParser/compile.bat
 
diff --git a/NlpParser/Nlp.g4 b/NlpParser/Nlp.g4
index 002e784..b543bd6 100644
--- a/NlpParser/Nlp.g4
+++ b/NlpParser/Nlp.g4
@@ -1,14 +1,14 @@
 grammar Nlp;
 
-nlp_document: LANG_HEADER (section)* ;
+document: LANG_HEADER (section)* ;
 
-section: SECTION_HEAD (sub_section | entry)* ;
+section: SECTION_HEAD (subSection | entry)* ;
 
-sub_section: SUB_SECTION_HEAD (entry)* ;
+subSection: SUB_SECTION_HEAD (entry)* ;
 
-entry: ENTRY_STRING                             # entry_string_t
-| ENTRY_STRING (LINE_CONCAT ENTRY_STRING)+      # entry_concated_string_t
-| ENTRY_INTEGER                                 # entry_integer_t
+entry: ENTRY_STRING                             # entryString
+| ENTRY_STRING (LINE_CONCAT ENTRY_STRING)+      # entryConcatedString
+| ENTRY_INTEGER                                 # entryInteger
 ;
 
 LANG_HEADER: 'Language:' [a-zA-Z]+ ;
diff --git a/NlpParser/NlpRunner.java b/NlpParser/NlpRunner.java
new file mode 100644
index 0000000..63fb006
--- /dev/null
+++ b/NlpParser/NlpRunner.java
@@ -0,0 +1,164 @@
+import org.antlr.v4.runtime.*;
+import org.antlr.v4.runtime.tree.*;
+import java.util.Stack;
+import java.util.List;
+import java.lang.StringBuilder;
+
+public class NlpRunner {
+	public static class ConvertToJson extends NlpBaseListener {
+		public ConvertToJson() {
+			mIndent = 0;
+			
+			mEntryIndex = 0;
+			mEntryIndexStack = new Stack<Integer>();
+			mIsFirst = true;
+			mIsFirstStack = new Stack<Boolean>();
+		}
+		
+		private String cutLangHead(String strl) {
+			return strl.substring("Language:".length());
+		}
+		private String cutSectionHead(String strl) {
+			return strl.substring(1, strl.length() - 1);
+		}
+		private String cutString(String strl) {
+			return strl.substring(1, strl.length() - 1);
+		}
+		private String joinConcatedString(List<TerminalNode> ls) {
+			StringBuilder sb = new StringBuilder();
+			for (TerminalNode node : ls) {
+				sb.append(cutString(node.getText()));
+			}
+			
+			return sb.toString();
+		}
+		
+		int mEntryIndex;
+		boolean mIsFirst;
+		Stack<Integer> mEntryIndexStack;
+		Stack<Boolean> mIsFirstStack;
+		private void pushSection() {
+			mEntryIndexStack.push(mEntryIndex);
+			mEntryIndex = 0;
+			mIsFirstStack.push(mIsFirst);
+			mIsFirst = true;
+		}
+		private void popSection() {
+			mEntryIndex = mEntryIndexStack.pop();
+			mIsFirst = mIsFirstStack.pop();
+		}
+		private void printComma() {
+			// only the first entry do not need comma
+			if (mIsFirst) {
+				mIsFirst = false;
+			} else {
+				System.out.print(',');
+			}
+		}
+		
+		int mIndent;
+		private void printIndent() {
+			for(int i = 0; i < mIndent; ++i) {
+				System.out.print('\t');
+			}
+		}
+		private void printEOL() {
+			System.out.print('\n');
+		}
+		private void printIndentLn(String strl) {
+			// call this when writting tail bracket
+			printEOL();
+			printIndent();
+			System.out.print(strl);
+		}
+		private void printIndentCommaLn(String strl) {
+			// call this when writting anything else.
+			printComma();
+			printEOL();
+			printIndent();
+			System.out.print(strl);
+		}
+		
+		private void printStrEntry(String val) {
+			printIndentCommaLn(String.format("\"%s\": \"%s\"", mEntryIndex++, val));
+		}
+		private void printIntEntry(int val) {
+			printIndentCommaLn(String.format("\"%s\": %d", mEntryIndex++, val));
+		}
+
+		
+		@Override
+		public void enterDocument(NlpParser.DocumentContext ctx) {
+			printIndentCommaLn("{");
+			pushSection();
+			++mIndent;
+			
+			printIndentCommaLn(String.format("\"Language\": \"%s\"", cutLangHead(ctx.LANG_HEADER().getText())));
+			
+			printIndentCommaLn("\"document\": {");
+			pushSection();
+			++mIndent;
+		}
+		@Override
+		public void exitDocument(NlpParser.DocumentContext ctx) {
+			--mIndent;
+			popSection();
+			printIndentLn("}");
+			
+			--mIndent;
+			popSection();
+			printIndentLn("}");
+		}
+		
+		@Override 
+		public void enterSection(NlpParser.SectionContext ctx) { 
+			printIndentCommaLn(String.format("\"%s\": {", cutSectionHead(ctx.SECTION_HEAD().getText())));
+			pushSection();
+			++mIndent;
+		}
+		@Override 
+		public void exitSection(NlpParser.SectionContext ctx) { 
+			--mIndent;
+			popSection();
+			printIndentLn("}");
+		}
+		
+		@Override 
+		public void enterSubSection(NlpParser.SubSectionContext ctx) { 
+			printIndentCommaLn(String.format("\"%s\": {", cutSectionHead(ctx.SUB_SECTION_HEAD().getText())));
+			pushSection();
+			++mIndent;
+		}
+		@Override 
+		public void exitSubSection(NlpParser.SubSectionContext ctx) {
+			--mIndent;
+			popSection();
+			printIndentLn("}");
+		}
+		
+		@Override 
+		public void enterEntryString(NlpParser.EntryStringContext ctx) {
+			printStrEntry(cutString(ctx.ENTRY_STRING().getText()));
+		}
+		@Override 
+		public void enterEntryConcatedString(NlpParser.EntryConcatedStringContext ctx) {
+			printStrEntry(joinConcatedString(ctx.ENTRY_STRING()));
+		}
+		@Override 
+		public void enterEntryInteger(NlpParser.EntryIntegerContext ctx) { 
+			printIntEntry(Integer.parseInt(ctx.ENTRY_INTEGER().getText()));
+		}
+	}
+	
+	public static void main(String[] args) throws Exception {
+		ANTLRInputStream input = new ANTLRInputStream(System.in);
+		NlpLexer lexer = new NlpLexer(input);
+		CommonTokenStream tokens = new CommonTokenStream(lexer);
+		NlpParser parser = new NlpParser(tokens);
+		
+		ParseTree tree = parser.document();
+		ParseTreeWalker walker = new ParseTreeWalker();
+		walker.walk(new ConvertToJson(), tree);
+		System.out.println();
+	}
+}
diff --git a/NlpParser/compile.bat b/NlpParser/compile.bat
index 47af20c..4305e92 100644
--- a/NlpParser/compile.bat
+++ b/NlpParser/compile.bat
@@ -1,3 +1,4 @@
 antlr4 Nlp.g4
 javac Nlp*.java
-grun Nlp nlp_document < testbench.txt
\ No newline at end of file
+;grun Nlp document -tree < testbench.txt
+java NlpRunner < ../NlpSrc/VT50.txt > result.txt
\ No newline at end of file