added LRParser

lab/16-18/src/main/java/play/parser/ASTNode.java

@@ -27,6 +27,12 @@ public class ASTNode {
         this.type = type;
     }
 
+    protected ASTNode (String type, String text){
+        this.type = type;
+        this.text = text;
+    }
+
+
     public String getType() {
         return type;
     }

lab/16-18/src/main/java/play/parser/DFAState.java

@@ -23,13 +23,13 @@ public class DFAState extends State {
     }
 
     /**
-     * 提供一个字符，看能否迁移到下一个状态
-     * @param ch
+     * 提供一个对象作为迁移条件，看能否迁移到下一个状态
+     * @param obj，做词法分析
      * @return
      */
-    protected DFAState getNextState(char ch) {
+    protected DFAState getNextState(Object obj) {
         for (Transition transition : transitions()) {
-            if (transition.match(ch)) {
+            if (transition.match(obj)) {
                 return (DFAState) getState(transition);
             }
         }
@@ -39,7 +39,7 @@ public class DFAState extends State {
     @Override
     public String toString() {
         String rtn = super.toString();
-        rtn += "\t(";
+        rtn += "\tNFA states: ";
         int i = 0;
         for (State state : states) {
             if (i++ > 0) {
@@ -50,7 +50,7 @@ public class DFAState extends State {
                 rtn += "(" + state.getGrammarNode().getName() + ")";
             }
         }
-        rtn += ")\n";
+        rtn += "\n";
         return rtn;
     }
 

lab/16-18/src/main/java/play/parser/Lexer.java

@@ -45,7 +45,7 @@ public class Lexer extends Regex{
         List<DFAState> dfaStates = NFA2DFA(nfaStates[0],CharSet.ascii);
         List<Token> tokens = tokenize(str,dfaStates.get(0),lexerGrammar);
         //加上结束符号
-        tokens.add(new Token("$", "$"));
+        tokens.add(Token.EOF);
         return tokens;
     }
 

lab/16-18/src/main/java/play/parser/SampleGrammar.java

@@ -197,6 +197,7 @@ public class SampleGrammar {
         return exp;
     }
 
+
     /**
      * 带有左递归的简化版的语法规则：
      * add	: mul | add '+' mul ;
@@ -230,6 +231,78 @@ public class SampleGrammar {
         return add;
     }
 
+    /**
+     * 带有左递归的语法规则：
+     * expression	: assign ;
+     * assign	: equal | assign '=' equal ;
+     * equal	: rel | equal ('==' | '!=') rel ;
+     * rel		: add | rel ('>=' | '>' | '<=' | '<') add ;
+     * add		: mul | add ('+' | '-') mul ;
+     * mul		: pri | mul ('*' | '/') pri ;
+     * pri		: ID | INT_LITERAL | LPAREN expression RPAREN ;
+     *
+     * @return
+     */
+    public static GrammarNode fullLeftRecursiveExpressionGrammar() {
+        //expression
+        GrammarNode exp = new GrammarNode("expression", GrammarNodeType.And);
+
+        //assign
+        GrammarNode assign = exp.createChild("assign", GrammarNodeType.Or);
+        GrammarNode equal = assign.createChild("equal", GrammarNodeType.Or);
+        GrammarNode assign_2 = assign.createChild(GrammarNodeType.And);
+        assign_2.addChild(assign);  //左递归
+        GrammarNode assignOp = assign_2.createChild( new Token("ASSIGN","="));
+        assign_2.addChild(equal);   //TODO 这里是否可以改为expresssion
+
+        //equal
+        GrammarNode rel = equal.createChild("rel", GrammarNodeType.Or);
+        GrammarNode equal_2 = equal.createChild(GrammarNodeType.And);
+        equal_2.addChild(equal);  //左递归
+        GrammarNode equalOp = equal_2.createChild( GrammarNodeType.Or);
+        equalOp.createChild(new Token("EQUAL","=="));
+        equalOp.createChild(new Token("NOTEQUAL", "!="));
+        equal_2.addChild(rel);
+
+        //rel
+        GrammarNode add = rel.createChild("add", GrammarNodeType.Or);
+        GrammarNode rel_2 = rel.createChild(GrammarNodeType.And);
+        rel_2.addChild(rel);  //左递归
+        GrammarNode relOp = rel_2.createChild( GrammarNodeType.Or);
+        relOp.createChild(new Token("GE",">="));
+        relOp.createChild(new Token("GT", ">"));
+        relOp.createChild(new Token("LE", "<="));
+        relOp.createChild(new Token("LT", "<"));
+        rel_2.addChild(add);
+
+        //add
+        GrammarNode mul = add.createChild("mul", GrammarNodeType.Or);
+        GrammarNode add_2 = add.createChild(GrammarNodeType.And);
+        add_2.addChild(add);  //左递归
+        GrammarNode addOp = add_2.createChild( GrammarNodeType.Or);
+        addOp.createChild(new Token("ADD","+"));
+        addOp.createChild(new Token("SUB", "-"));
+        add_2.addChild(mul);
+
+        //mul
+        GrammarNode pri = mul.createChild("pri", GrammarNodeType.Or);
+        GrammarNode mul_2 = mul.createChild(GrammarNodeType.And);
+        mul_2.addChild(mul);
+        GrammarNode mulOp = mul_2.createChild(GrammarNodeType.Or);
+        mulOp.createChild(new Token("MUL", "*"));
+        mulOp.createChild(new Token("DIV", "/"));
+        mul_2.addChild(pri);
+
+        //pri
+        pri.createChild(new Token("ID"));
+        pri.createChild(new Token("INT_LITERAL"));
+        GrammarNode pri_3 = pri.createChild(GrammarNodeType.And);
+        pri_3.createChild(new Token("LPAREN"));
+        pri_3.addChild(exp);
+        pri_3.createChild(new Token("RPAREN"));
+
+        return exp;
+    }
 
     /**
      * 创建一个示例用词法规则，支持：

lab/16-18/src/main/java/play/parser/Token.java

@@ -21,6 +21,9 @@ public class Token {
     private int startColumn = 0;
     private int endColumn = 0;
 
+    //程序结束符号
+    protected static Token EOF = new Token("$");
+
     public Token(String type){
         this.type = type;
     }

lab/16-18/src/main/java/play/parser/TokenReader.java

@@ -10,12 +10,18 @@ import java.util.List;
 public class TokenReader {
 
     List<Token> tokens = null;
+
+    //当前指针位置。
     int pos = 0;
 
     public TokenReader(List<Token> tokens) {
         this.tokens = tokens;
     }
 
+    /**
+     * 读取一个Token，并移动指针。
+     * @return 如果已经读完，则返回null。
+     */
     public Token read() {
         if (pos < tokens.size()) {
             return tokens.get(pos++);
@@ -23,6 +29,10 @@ public class TokenReader {
         return null;
     }
 
+    /**
+     * 预读一个Token。
+     * @return 如果已经读完，则返回null。
+     */
     public Token peek() {
         if (pos < tokens.size()) {
             return tokens.get(pos);
@@ -30,16 +40,27 @@ public class TokenReader {
         return null;
     }
 
+    /**
+     * 回溯一个Token。
+     */
     public void unread() {
         if (pos > 0) {
             pos--;
         }
     }
 
+    /**
+     * 获取当前指针位置。
+     * @return
+     */
     public int getPosition() {
         return pos;
     }
 
+    /**
+     * 设置指针位置。用于回溯。
+     * @param position
+     */
     public void setPosition(int position) {
         if (position >=0 && position < tokens.size()){
             pos = position;