From f3c7983749e44555700a176fab9965793c13f452 Mon Sep 17 00:00:00 2001
From: LiXiaoQi <lxq877250099@mail.ustc.edu.cn>
Date: Thu, 22 Sep 2022 09:29:48 +0800
Subject: [PATCH] Now it can pass a few tests. A clear bug is the regex for
 comment

---
 src/parser/lexical_analyzer.l | 64 +++++++++++++++++++++--------------
 src/parser/syntax_analyzer.y  |  2 +-
 tests/parser/easy/expr.cminus |  2 +-
 3 files changed, 41 insertions(+), 27 deletions(-)

diff --git a/src/parser/lexical_analyzer.l b/src/parser/lexical_analyzer.l
index 1bb15aa..e6c1728 100644
--- a/src/parser/lexical_analyzer.l
+++ b/src/parser/lexical_analyzer.l
@@ -7,12 +7,24 @@
 #include "syntax_tree.h"
 #include "syntax_analyzer.h"
 
-int lines;
+int lines = 1;
 int pos_start;
 int pos_end;
 
 void pass_node(char *text){
-     yylval.value = new_syntax_tree_node(text);
+     yylval.node = new_syntax_tree_node(text);
+}
+void comment_helper(char *comment, unsigned int len) 
+{
+    pos_end += 2;
+    for (int i = 2; i < len-2; i++){
+        if (comment[i] == '\n' || comment[i] == '\r'){
+            lines++;
+            pos_end = 0;
+        } else
+            pos_end++;
+    }
+    pos_end += 2;
 }
 
 /*****************声明和选项设置  end*****************/
@@ -23,8 +35,11 @@ letter [a-zA-Z]
 digit [0-9]
 ID {letter}+
 INTEGER {digit}+
-FLOAT {digit}+\. | {digit}*\.{digit}+
+FLOAT {digit}+\.|{digit}*\.{digit}+
+COMMENT "/*".*"*/"
 
+NEWLINE \r\n|\r|\n
+WHITESPACE [ \t]
 /*
 
 %token <node> _IF _ELSE _WHILE _RETURN _INT _FLOAT _VOID 
@@ -35,31 +50,30 @@ FLOAT {digit}+\. | {digit}*\.{digit}+
 */
 
 %%
-if {pos_start = pos_end; pos_end += 2; pass_node("if"); return _IF;}
-else {pos_start = pos_end; pos_end += 4; pass_node("else"); return _ELSE;}
-while {pos_start = pos_end; pos_end += 5; pass_node("while"); return _WHILE;}
-return {pos_start = pos_end; pos_end += 6; pass_node("return"); return _RETURN;}
-int {pos_start = pos_end; pos_end += 3; pass_node("int"); return _INT;}
-float {pos_start = pos_end; pos_end += 5; pass_node("float"); return _FLOAT;}
-void {pos_start = pos_end; pos_end += 4; pass_node("void"); return _VOID;}
+{COMMENT} { comment_helper(yytext, yyleng); }
+if { pos_start = pos_end; pos_end += 2; pass_node("if"); return _IF;}
+else { pos_start = pos_end; pos_end += 4; pass_node("else"); return _ELSE;}
+while { pos_start = pos_end; pos_end += 5; pass_node("while"); return _WHILE;}
+return { pos_start = pos_end; pos_end += 6; pass_node("return"); return _RETURN;}
+int { pos_start = pos_end; pos_end += 3; pass_node("int"); return _INT;}
+float { pos_start = pos_end; pos_end += 5; pass_node("float"); return _FLOAT;}
+void { pos_start = pos_end; pos_end += 4; pass_node("void"); return _VOID;}
 
-{ID} {pos_start = pos_end; pos_end += yyleng; pass_node(yytext); return _ID;}
-{INTEGER} {pos_start = pos_end; pos_end += yyleng; pass_node(yytext); return _INTEGER;}
-{FLOAT} {pos_start = pos_end; pos_end += yyleng; pass_node(yytext); return _FLOATPOINT;}
+{ID} { pos_start = pos_end; pos_end += yyleng; pass_node(yytext); return _ID;}
+{INTEGER} { pos_start = pos_end; pos_end += yyleng; pass_node(yytext); return _INTEGER;}
+{FLOAT} { pos_start = pos_end; pos_end += yyleng; pass_node(yytext); return _FLOATPOINT;}
 
-\= 	{pos_start = pos_end; pos_end += 1; pass_node("="); return _ASSIGN;}
-"<=" | ">=" | "<" | ">" | "==" | "!=" {pos_start = pos_end; pos_end += yyleng; pass_node(yytext); return _RELOP;}
-"+" | "-" {pos_start = pos_end; pos_end += 1; pass_node(yytext); return _ADD_OP;}
-"*" | "/" {pos_start = pos_end; pos_end += 1; pass_node(yytext); return _MUL_OP;}
+"=" { pos_start = pos_end; pos_end += 1; pass_node("="); return _ASSIGN;}
+"<="|">="|"<"|">"|"=="|"!=" { pos_start = pos_end; pos_end += yyleng; pass_node(yytext); return _RELOP;}
+"+"|"-" { pos_start = pos_end; pos_end += 1; pass_node(yytext); return _ADD_OP;}
+"*"|"/" { pos_start = pos_end; pos_end += 1; pass_node(yytext); return _MUL_OP;}
 
-\[ | \] {pos_start = pos_end; pos_end += 1; pass_node(yytext); return yytext[0] == '[' ? _L_SQUARE : _R_SQUARE;}
-\( | \) {pos_start = pos_end; pos_end += 1; pass_node(yytext); return yytext[0] == '(' ? _L_PARE : _R_PARE;}
-\{ | \} {pos_start = pos_end; pos_end += 1; pass_node(yytext); return yytext[0] == '{' ? _L_BRACKET : _R_BRACKET;}
+"["|"]" { pos_start = pos_end; pos_end += 1; pass_node(yytext); return yytext[0] == '[' ? _L_SQUARE : _R_SQUARE;}
+"("|")" { pos_start = pos_end; pos_end += 1; pass_node(yytext); return yytext[0] == '(' ? _L_PARE : _R_PARE;}
+"{"|"}" { pos_start = pos_end; pos_end += 1; pass_node(yytext); return yytext[0] == '{' ? _L_BRACKET : _R_BRACKET;}
 
-"," | ";" {pos_start = pos_end; pos_end += 1; pass_node(yytext); return yytext[0] == ',' ? _COMMA : _SEMI;}
+","|";" { pos_start = pos_end; pos_end += 1; pass_node(yytext); return yytext[0] == ',' ? _COMMA : _SEMI;}
 
-" " | \t { pos_end++; }
-\r\n | \n | \r { lines++; pos_end = 0;}
-/* . { pos_start = pos_end; pos_end++; return ERROR; } */
-/****请在此补全所有flex的模式与动作  end******/
+{WHITESPACE} { pos_end++; }
+{NEWLINE} { lines++; pos_end = 0;}
 %%
diff --git a/src/parser/syntax_analyzer.y b/src/parser/syntax_analyzer.y
index 1beaaf5..1ae75be 100644
--- a/src/parser/syntax_analyzer.y
+++ b/src/parser/syntax_analyzer.y
@@ -32,7 +32,7 @@ syntax_tree_node *node(const char *node_name, int children_num, ...);
    Hint: See pass_node(), node(), and syntax_tree.h.
          Use forward declaring. */
 %union {
-    node value;
+    struct _syntax_tree_node *node;
 }
 
 /* TODO: Your tokens here. */
diff --git a/tests/parser/easy/expr.cminus b/tests/parser/easy/expr.cminus
index 787f1ca..2c99f4e 100644
--- a/tests/parser/easy/expr.cminus
+++ b/tests/parser/easy/expr.cminus
@@ -13,4 +13,4 @@ int main(void){
     x = a/x;
 
     return 0;
-}
\ No newline at end of file
+}
-- 
GitLab