scandoc.cpp

编译原理--词法扫描

// scanDoc.cpp : implementation of the CScanDoc class
//

#include "stdafx.h"
#include "scan.h"

#include "scanDoc.h"

//#include "sscan.h"  //add by user

#ifdef _DEBUG
#define new DEBUG_NEW
#undef THIS_FILE
static char THIS_FILE[] = __FILE__;
#endif

//external var

extern CTraceBar g_TraceBar;
CEdit* g_pEditCtrl;             //edit control


//extern SYMTAB_NODE_PTR symtab_root;
//+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
#include <iostream.h>
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <ctype.h>

#ifndef FALSE
#define FALSE 0
#endif

#ifndef TRUE
#define TRUE 1
#endif

#define MAXRESERVED 8

const int namelen=40;
typedef struct symtab_node           //二叉树的结点结构
{
  struct symtab_node *left;          //指向左孩子
  struct symtab_node *right;         //指向右孩子
  char type[namelen+1];
  char name[namelen+1];              //指向标识符的名字串
  int lineno;						//line number
}SYMTAB_NODE,*SYMTAB_NODE_PTR;

SYMTAB_NODE_PTR symtab_root=NULL; 

typedef enum 
   {_ENDFILE,_ERROR,
    IF,THEN,ELSE,END,REPEAT,UNTIL,READ,WRITE,
    ID,NUM,
    ASSIGN,EQ,LT,PLUS,MINUS,TIMES,OVER,LPAREN,RPAREN,SEMI
   } tTokenType;

FILE* source; /* source code text file */
int lineno; /* source line number for listing */

int nStartChar=-1;
int nFilePos=-1;

#define MAXCHILDREN 3
#define MAXTOKENLEN 40

/* tokenString array stores the lexeme of each token */
char tokenString[MAXTOKENLEN+1];

/* states in scanner DFA */
typedef enum
   { START,INASSIGN,INCOMMENT,INNUM,INID,DONE }
   StateType;

/* BUFLEN = length of the input buffer for
   source code lines */
#define BUFLEN 256

static char lineBuf[BUFLEN]; /* holds the current line */
static int linepos = 0; /* current position in LineBuf */
static int bufsize = 0; /* current size of buffer string */
static int EOF_flag = FALSE; /* corrects ungetNextChar behavior on EOF */

/* lookup table of reserved words */
static struct
    { char* str;
      tTokenType tok;
    } reservedWords[MAXRESERVED]
   = {{"if",IF},{"then",THEN},{"else",ELSE},{"end",END},
      {"repeat",REPEAT},{"until",UNTIL},{"read",READ},
      {"write",WRITE}};


//在由np指向的树中查找标识符*name
SYMTAB_NODE_PTR search_symtab(char *name,SYMTAB_NODE_PTR np)
{
  int cmp;
  while(np!=NULL)
  {
    cmp=strcmp(name,np->name);
    if(cmp==0) return(np);
    np=cmp<0 ? np->left : np->right;
  }
  return(NULL);
}

void visit_symtab(const SYMTAB_NODE_PTR np)
{
	if(np)
	{
		visit_symtab(np->left);
		cout<<np->name<<endl;
		visit_symtab(np->right);
	}
}

//在由npp指向的树中加入名为*name,值为value的标识符结点
SYMTAB_NODE_PTR enter_symtab(const char *name,const char *type , int line,SYMTAB_NODE_PTR &npp)
{
  int cmp;
  SYMTAB_NODE_PTR np1,np2;
  SYMTAB_NODE_PTR new_nodep;
  new_nodep=new symtab_node;
  strncpy(new_nodep->name,name,namelen);
  strncpy(new_nodep->type,type,namelen);
  new_nodep->lineno=line;
  new_nodep->left=new_nodep->right=NULL;
  np1=npp;
  while(np1 != NULL)
  {
    cmp=strcmp(name,np1->name);
    np2=np1;
	np1 =cmp <0 ? (np1->left) : (np1->right);
  }
  np1=new_nodep;
  if(npp==NULL)
	  npp=new_nodep;
  else cmp < 0 ? np2->left=np1 : np2->right=np1;
  return(new_nodep);
}



/* getNextChar fetches the next non-blank character
   from lineBuf, reading in a new line if lineBuf is
   exhausted */
static int getNextChar(void)
{ 
    nFilePos ++;   //add for edit control ,start from 0;
	if (!(linepos < bufsize))
  { 
	lineno++;
    if (fgets(lineBuf,BUFLEN-1,source))
    { 
//    if (EchoSource) fprintf(listing,"%4d: %s",lineno,lineBuf);
      bufsize = strlen(lineBuf);
      linepos = 0;
      return lineBuf[linepos++];
    }
    else
    { EOF_flag = TRUE;
      return EOF;
    }
  }
	else return lineBuf[linepos++];
}

/* ungetNextChar backtracks one character
   in lineBuf */
static void ungetNextChar(void)
{ 
	if (!EOF_flag) linepos-- ;
	nFilePos--;
}


/* lookup an identifier to see if it is a reserved word */
/* uses linear search */
static tTokenType reservedLookup (char * s)
{ int i;
  for (i=0;i<MAXRESERVED;i++)
    if (!strcmp(s,reservedWords[i].str))
      return reservedWords[i].tok;
  return ID;
}

/****************************************/
/* the primary function of the scanner  */
/****************************************/
/* function getToken returns the 
 * next token in source file
 */
tTokenType getToken(void)
{  /* index for storing into tokenString */
   int tokenStringIndex = 0;
   /* holds current token to be returned */
   tTokenType currentToken;
   /* current state - always begins at START */
   StateType state = START;
   /* flag to indicate save to tokenString */
   int save;
   while (state != DONE)
   { 
    int c = getNextChar();
     save = TRUE;
     switch (state)
     { case START:
         if (isdigit(c))
           state = INNUM;
         else if (isalpha(c))
           state = INID;
         else if (c == ':')
           state = INASSIGN;
         else if ((c == ' ') || (c == '\t') || (c == '\n'))
		 {
			 save = FALSE;

		 /* in edit component ,SetSel() takes '\n' as two chars,
		  * so nFilePos must plus 2 
		  */
		 if(c=='\n')  nFilePos++;
		 }
         else if (c == '{')
         { save = FALSE;
           state = INCOMMENT;
         }
         else
         { state = DONE;
           switch (c)
           { case EOF:
               save = FALSE;
               currentToken = _ENDFILE;
               break;
             case '=':
               currentToken = EQ;
               break;
             case '<':
               currentToken = LT;
               break;
             case '+':
               currentToken = PLUS;
               break;
             case '-':
               currentToken = MINUS;
               break;
             case '*':
               currentToken = TIMES;
               break;
             case '/':
               currentToken = OVER;
               break;
             case '(':
               currentToken = LPAREN;
               break;
             case ')':
               currentToken = RPAREN;
               break;
             case ';':
               currentToken = SEMI;
               break;
             default:
               currentToken = _ERROR;
               break;
           }
         }
         break;
       case INCOMMENT:
         save = FALSE;
         if (c == EOF)
         { state = DONE;
           currentToken = _ENDFILE;
         }
		 /* in edit component ,SetSel() takes '\n' as two chars,
		  * so nFilePos must plus 2 
		  */
		 else if(c=='\n') nFilePos++;
         else if (c == '}') state = START;
         break;
       case INASSIGN:
         state = DONE;
         if (c == '=')
           currentToken = ASSIGN;
         else
         { /* backup in the input */
           ungetNextChar();
           save = FALSE;
           currentToken = _ERROR;
         }
         break;
       case INNUM:
         if (!isdigit(c))
         { /* backup in the input */
           ungetNextChar();
           save = FALSE;
           state = DONE;
           currentToken = NUM;
         }
         break;
       case INID:
         if (!isalpha(c))
         { /* backup in the input */
           ungetNextChar();
           save = FALSE;
           state = DONE;
           currentToken = ID;
         }
         break;
       case DONE:
       default: /* should never happen */
 //        fprintf(listing,"Scanner Bug: state= %d\n",state);
         state = DONE;
         currentToken = _ERROR;
         break;
     }
     if ((save) && (tokenStringIndex <= MAXTOKENLEN))
       tokenString[tokenStringIndex++] = (char) c;
     if (state == DONE)
     { tokenString[tokenStringIndex] = '\0';
       if (currentToken == ID)
         currentToken = reservedLookup(tokenString);
     }
   }
/*   if (TraceScan) {
     fprintf(listing,"\t%d: ",lineno);
     printToken(currentToken,tokenString);
   }
*/   return currentToken;
} /* end getToken */


void maint()