编译原理及实践教材TINY编译器代码解析

    编译原理及实践教材附带了TINY编译器,在这里对这个小型编译器的代码,做一下简单的解析.

   TINY编译器的词法分析Lex源程序是:

%{ #include "globals.h" #include "util.h" #include "scan.h" /* lexeme of identifier or reserved word */ char tokenString[MAXTOKENLEN+1]; %} digit       [0-9] number      {digit}+ letter      [a-zA-Z] identifier  {letter}+ newline     /n whitespace  [ /t]+ %% "if"            {return IF;} "then"          {return THEN;} "else"          {return ELSE;} "end"           {return END;} "repeat"        {return REPEAT;} "until"         {return UNTIL;} "read"          {return READ;} "write"         {return WRITE;} ":="            {return ASSIGN;} "="             {return EQ;} "<"             {return LT;} "+"             {return PLUS;} "-"             {return MINUS;} "*"             {return TIMES;} "/"             {return OVER;} "("             {return LPAREN;} ")"             {return RPAREN;} ";"             {return SEMI;} {number}        {return NUM;} {identifier}    {return ID;} {newline}       {lineno++;} {whitespace}    {/* skip whitespace */} "{"             { char c;                   do                   { c = input();                     if (c == EOF) break;                     if (c == '/n') lineno++;                   } while (c != '}');                 } .               {return ERROR;} %% TokenType getToken(void) { static int firstTime = TRUE;   TokenType currentToken;   if (firstTime)   { firstTime = FALSE;     lineno++;     yyin = source;     yyout = listing;   }   currentToken = yylex();   strncpy(tokenString,yytext,MAXTOKENLEN);   if (TraceScan) {     fprintf(listing,"/t%d: ",lineno);     printToken(currentToken,tokenString);   }   return currentToken; }

TINY编译器的语法分析Yacc源程序是:

%{ #define YYPARSER /* distinguishes Yacc output from other code files */ #include "globals.h" #include "util.h" #include "scan.h" #include "parse.h" #define YYSTYPE TreeNode * static char * savedName; /* for use in assignments */ static int savedLineNo;  /* ditto */ static TreeNode * savedTree; /* stores syntax tree for later return */ %} %token IF THEN ELSE END REPEAT UNTIL READ WRITE %token ID NUM %token ASSIGN EQ LT PLUS MINUS TIMES OVER LPAREN RPAREN SEMI %token ERROR   %% /* Grammar for TINY */ program     : stmt_seq                  { savedTree = $1;}             ; stmt_seq    : stmt_seq SEMI stmt                  { YYSTYPE t = $1;                    if (t != NULL)                    { while (t->sibling != NULL)                         t = t->sibling;                      t->sibling = $3;                      $$ = $1; }                      else $$ = $3;                  }             | stmt  { $$ = $1; }             ; stmt        : if_stmt { $$ = $1; }             | repeat_stmt { $$ = $1; }             | assign_stmt { $$ = $1; }             | read_stmt { $$ = $1; }             | write_stmt { $$ = $1; }             | error  { $$ = NULL; }             ; if_stmt     : IF exp THEN stmt_seq END                  { $$ = newStmtNode(IfK);                    $$->child[0] = $2;                    $$->child[1] = $4;                  }             | IF exp THEN stmt_seq ELSE stmt_seq END                  { $$ = newStmtNode(IfK);                    $$->child[0] = $2;                    $$->child[1] = $4;                    $$->child[2] = $6;                  }             ; repeat_stmt : REPEAT stmt_seq UNTIL exp                  { $$ = newStmtNode(RepeatK);                    $$->child[0] = $2;                    $$->child[1] = $4;                  }             ; assign_stmt : ID { savedName = copyString(tokenString);                    savedLineNo = lineno; }               ASSIGN exp                  { $$ = newStmtNode(AssignK);                    $$->child[0] = $4;                    $$->attr.name = savedName;                    $$->lineno = savedLineNo;                  }             ; read_stmt   : READ ID                  { $$ = newStmtNode(ReadK);                    $$->attr.name =                      copyString(tokenString);                  }             ; write_stmt  : WRITE exp                  { $$ = newStmtNode(WriteK);                    $$->child[0] = $2;                  }             ; exp         : simple_exp LT simple_exp                  { $$ = newExpNode(OpK);                    $$->child[0] = $1;                    $$->child[1] = $3;                    $$->attr.op = LT;                  }             | simple_exp EQ simple_exp                  { $$ = newExpNode(OpK);                    $$->child[0] = $1;                    $$->child[1] = $3;                    $$->attr.op = EQ;                  }             | simple_exp { $$ = $1; }             ; simple_exp  : simple_exp PLUS term                  { $$ = newExpNode(OpK);                    $$->child[0] = $1;                    $$->child[1] = $3;                    $$->attr.op = PLUS;                  }             | simple_exp MINUS term                  { $$ = newExpNode(OpK);                    $$->child[0] = $1;                    $$->child[1] = $3;                    $$->attr.op = MINUS;                  }             | term { $$ = $1; }             ; term        : term TIMES factor                  { $$ = newExpNode(OpK);                    $$->child[0] = $1;                    $$->child[1] = $3;                    $$->attr.op = TIMES;                  }             | term OVER factor                  { $$ = newExpNode(OpK);                    $$->child[0] = $1;                    $$->child[1] = $3;                    $$->attr.op = OVER;                  }             | factor { $$ = $1; }             ; factor      : LPAREN exp RPAREN                  { $$ = $2; }             | NUM                  { $$ = newExpNode(ConstK);                    $$->attr.val = atoi(tokenString);                  }             <