تحويل الإدخال إلى رموز

ما هو المحلّل الرمزي؟

نوع الرمز

typedef enum {
  TOK_NUM, TOK_PLUS, TOK_MINUS,
  TOK_STAR, TOK_SLASH,
  TOK_LPAREN, TOK_RPAREN, TOK_EOF
} TokKind;

typedef struct {
  TokKind kind;
  int value; /* used when kind == TOK_NUM */
} Token;

موضع المسح

static const char *src;

static char peek(void) {
  return *src;
}

static char advance(void) {
  return *src++;
}

تجاوز المسافات البيضاء

#include <ctype.h>

static void skip_ws(void) {
  while (isspace((unsigned char)peek()))
    advance();
}

تحليل عدد معجميًا

static int lex_number(void) {
  int n = 0;
  while (isdigit((unsigned char)peek())) {
    n = n * 10 + (advance() - '0');
  }
  return n;
}

الدالة next_token

static Token next_token(void) {
  skip_ws();
  char c = peek();
  if (c == '\0') return (Token){TOK_EOF, 0};
  if (isdigit((unsigned char)c))
    return (Token){TOK_NUM, lex_number()};
  advance();
  switch (c) {
    case '+': return (Token){TOK_PLUS, 0};
    case '-': return (Token){TOK_MINUS, 0};
    case '*': return (Token){TOK_STAR, 0};
    case '/': return (Token){TOK_SLASH, 0};
    case '(': return (Token){TOK_LPAREN, 0};
    case ')': return (Token){TOK_RPAREN, 0};
  }
  return (Token){TOK_EOF, 0};
}

تشغيل المحلّل المعجمي

#include <stdio.h>
#include <ctype.h>

typedef enum { TOK_NUM, TOK_PLUS, TOK_STAR, TOK_EOF } TokKind;
typedef struct { TokKind kind; int value; } Token;

static const char *src;
static char peek(void){ return *src; }
static char advance(void){ return *src++; }

static Token next_token(void){
  while (isspace((unsigned char)peek())) advance();
  char c = peek();
  if (c=='\0') return (Token){TOK_EOF,0};
  if (isdigit((unsigned char)c)){
    int n=0; while(isdigit((unsigned char)peek())) n=n*10+(advance()-'0');
    return (Token){TOK_NUM,n};
  }
  advance();
  if (c=='+') return (Token){TOK_PLUS,0};
  return (Token){TOK_STAR,0};
}

int main(void){
  src = "12 + 3 * 4";
  Token t;
  do {
    t = next_token();
    if (t.kind==TOK_NUM) printf("NUM %d\n", t.value);
    else if (t.kind==TOK_PLUS) printf("PLUS\n");
    else if (t.kind==TOK_STAR) printf("STAR\n");
    else printf("EOF\n");
  } while (t.kind != TOK_EOF);
  return 0;
}

رمز واحد للاستبصار

static Token current;

static void init_lexer(const char *s) {
  src = s;
  current = next_token();
}

static Token cur(void) { return current; }

static void bump(void) { current = next_token(); }

الإبلاغ عن الأخطاء المعجمية

#include <stdio.h>
#include <stdlib.h>

static void lex_error(char c) {
  fprintf(stderr, "lex error: unexpected '%c'\n", c);
  exit(1);
}

العوامل متعددة المحارف

/* fragment: distinguish '=' from '==' */
if (peek() == '=') {
  advance();
  if (peek() == '=') { advance(); /* TOK_EQ */ }
  else { /* TOK_ASSIGN */ }
}

لماذا تهم الرموز

تحقّق سريع

مراجعة