詞法分析器

上週幫一個剛剛找到本身方向的好友作了一個編譯原理的課程實驗，要求是作一個詞法分析器，具體要求以下：

對下述文法和單詞表定義的語言設計編制一個語法分析器。

（1）單詞符號及種別表

單詞符號	種別編碼	單詞值
main	1
int	2
float	3
double	4
char	5
if	6
else	7
do	8
while	9
l(l|d)*	10	內部字符串
( +|-|ε ) dd*(.dd* | ε)( e ( +|-|ε ) dd*|ε)	20	二進制數值表示
=	21
+	22
-	23
*	24
/	25
(	26
)	27
{	28
}	29
,	30
;	31
>	32
>=	33
<	34
&lt;=	35
==	36
!=	37

（2）語法結構定義

&lt;表達式> ::= <項>{ +<項>|-<項>}

<項> ::= <因子>{*<因子>|/<因子>}

<因子> ::=ID|num|(<表達式>)

num::= ( +|-|ε ) 數字數字*(.數字數字* | ε)( e ( +|-|ε ) 數字數字*|ε)

ID::=字母(字母|數字)*

字母::=a|b|c…|z|A|B|C…|Z

數字::=0|1|2…|9

原本不是個很難的程序，但我仍是寫的較爲複雜，有這方面心得的朋友能夠和我聯繫，一塊兒探討一下吧～～～下面我把本身簡陋的代碼貼在下面：

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

#define INVALUECHAR   '.'  //非法字符，用於給token賦初值
#define MAXCHARNUM    80   //接受字符串輸入的長度
#define TOKENLENGTH   15   //詞組的最大長度
#define RETABLENGTH   9    //關鍵字的個數
#define TRUE          0
#define FALSE         -1

int  syn, sum = 0;
char token[TOKENLENGTH] = { '\0' }; //記錄找到的詞組
char *rwtab[RETABLENGTH] = { "main", "int", "float", "double", "char", "if", "else", "do", "while" }; //記錄關鍵字

int IsNumber(char num); //判斷num是否爲數字
int GetSum(int i); //求token前i個數的加權和
int HoldE(int *local, char *string, int num, int *i); //讀取浮點數時進行遇到符號e的操做
int NumberOP(int *local, char *string, int num, int *i, char c); //讀取浮點數時進行遇到非數字字符的操做
int Analyze(char *string, int num); //搜尋最長字串的函數


int main()
{
    int p = 0, i, j;
    char string[MAXCHARNUM] = { '\0' };

    printf("please input a string:\n");
    //讀取輸入的字符串，超過長度限定就讓用戶從新輸入
    do {
        scanf("%c", &string[p++]);
        if (MAXCHARNUM <= p) {
            printf("\nYou have input more than 80 characters!\n");
            printf("Please input again:\n");
            string[0] = '\0';
            p = 0;
        }
    }while ('\n' != string[p-1]); //字符串以'\n'結束

    //遍歷輸入的字符串，忽略回車和空格
    for (i = 0; i < p; i++) {
        if ((' ' == string[i]) || ('\n' == string[i]))
            continue;
        else {
            j = Analyze(string, i); //當出現錯誤時Analyze返回FALSE，正確時返回找到字串的最後一個字符的位置
            if (FALSE == j) {
                printf("\nYour input is wrong!\n");
                return 1;
            }
            i = j; //將i賦值爲查找到字串的最後一個字符的位置

            switch (syn) {
                //-1爲出現非法字符，0爲正常結束，11爲整型數的sum輸出
                case -1: { printf("\nYou have input illegal characters!\nSo it ended!\n");
                           break; }
                case 0: { break; }
                case 20: { if (0 != sum) {
                               printf("( %-5d%15d )\n", syn, sum);
                               sum = 0;
                           }
                           else
                               printf("( %-5d%15s )\n", syn, token);
                           break; }
                default: { printf("( %-5d%15s )\n", syn, token);
                           break; }
            }
        }
    }
    return 0;
}

//判斷num是否爲數字
int IsNumber(char num)
{
   if (('0' <= num) && ('9' >= num))
       return TRUE;
   else
       return FALSE;
}

//求token前i個數字的加權和
int GetSum(int i)
{
    int sum, j;
    char a[TOKENLENGTH];
    for (j = 0; j < i; j++)
        a[j] = token[j];
    a[j] = '\0';
    sum = atoi(a);
    return sum;
}

//讀取浮點數時遇到e的處理
int HoldE(int *local, char *string, int num, int *i)
{
    char ch = 'e';
    token[(*i)++] = ch;
    ch = string[++num];
    //e的後一位爲數字
    if (TRUE == IsNumber(ch)) {
        token[(*i)++] = ch;
        ch = string[++num];
        while (TRUE == IsNumber(ch)) {
            token[(*i)++] = ch;
            ch = string[++num];
        }
        *local = num - 1; //改變string中下標須要指向的位置
        return TRUE;
    }
    //e的後一位爲'+'或'-'
    else if (('+' == ch) || ('-' == ch)) {
        token[(*i)++] = ch;
        ch = string[++num];
        if (TRUE != IsNumber(ch))
            return FALSE;
        while (TRUE == IsNumber(ch)) {
            token[(*i)++] = ch;
            ch = string[++num];
        }
        *local = num - 1;
        return TRUE;
    }
    //其餘狀況就出現錯誤
    else
        return FALSE;
}

//讀取浮點數時遇到第一個非數字的字符的處理
int NumberOP(int *local, char *string, int num, int *i, char c)
{
    int result = TRUE;
    char ch = c;
    //這個非數字字符爲小數點
    if ('.' == ch) {
        token[(*i)++] = ch;
        ch = string[++num];
        //小數點後一位若是不是數字就出現錯誤
        if (TRUE != IsNumber(ch))
            return FALSE;
        while (TRUE == IsNumber(ch)) {
            token[(*i)++] = ch;
            ch = string[++num];
        }
        //小數點以後一串數字事後遇到e
        if ('e' == ch) {
            result = HoldE(local, string, num, i);
            if (FALSE == result)
                return FALSE;
        }
        else
            *local = num - 1;
    }
    //這個字符爲e
    else if ('e' == ch) {
        result = HoldE(local, string, num, i);
        if (FALSE == result)
            return FALSE;
    }
    //其餘狀況就不用再讀取了，直接修改local並返回
    else
        *local = num - 1;
    syn = 20;
    return TRUE;
}

//查詢最長字串
int Analyze(char *string, int num)
{
    int m, i = 0, local = num, n, flag = 0;
    int result = TRUE;
    char ch;

    for (m = 0; m < TOKENLENGTH; m++)
        token[m++] = INVALUECHAR;

    ch = string[num];
    //第一種狀況爲字符
    if (((ch <= 'z') && (ch >= 'a')) || ((ch <= 'Z') && (ch >= 'A'))) {
        while(((ch <= 'z') && (ch >= 'a')) || (( ch <= 'Z') && (ch >= 'A')) || (TRUE == IsNumber(ch))) {
            token[i++] = ch;
            //token長度限定
            if (i >= (TOKENLENGTH - 1))
                return FALSE;
            else
                ch = string[++num];
        }
        local = num - 1;
        //syn預設爲l（l | d）*的形式
        syn = 10;
        token[i] = '\0';
        //遍歷關鍵字數組，若是找到就修改syn並
        for (n = 0; n < RETABLENGTH; n++)
            if (0 == strcmp(token, rwtab[n])) {
                syn = n + 1;
                break;
            }
    }
    //第二種狀況爲遇到'+'、'-'或數字
    else if (('+' == ch) || ('-' == ch) || (TRUE == IsNumber(ch))) {
        token[i++] = ch;
        //若是爲數字就用flag標記一下
        if (TRUE == IsNumber(ch))
            flag = 1;
        ch = string[++num];
        //下一位也爲數字
        if (TRUE == IsNumber(ch)) {
            token[i++] = ch;
            ch = string[++num];
            while (TRUE == IsNumber(ch)) {
                token[i++] = ch;
                ch = string[++num];
            }
            //遇到第一個
            result = NumberOP(&local, string, num, &i, ch);
            if (FALSE == result)
                return FALSE;
        }
        //下一位不爲數字要分狀況，當第一位爲數字時執行下列操做，
        //第一位爲'+'、'-'時，下一位必須爲數字
        else if (1 == flag) {
            result = NumberOP(&local, string, num, &i, ch);
            if (FALSE == result)
                return FALSE;
        }
        //當第一位爲'+'、'-'時，下一位不爲數字證實'+'、'-'已是最長的字串了
        else {
            if ('+' == token[i-1])
                syn = 22;
            else
                syn = 23;
        }
    }
    else {
        switch (ch) {
            case '=': { token[i++] = ch;
                        ch = string[++num];
                        if ('=' == ch) {
                            syn = 36;
                            token[i++] = ch;
                            local = num;
                            break;
                        }
                        syn = 21;
                        break; }
            case '*': { syn = 24;
                        token[i++] = ch;
                        break; }
            case '/': { syn = 25;
                        token[i++] = ch;
                        break; }
            case '(': { syn = 26;
                        token[i++] = ch;
                        break; }
            case ')': { syn = 27;
                        token[i++] = ch;
                        break; }
            case '{': { syn = 28;
                        token[i++] = ch;
                        break; }
            case '}': { syn = 29;
                        token[i++] = ch;
                        break; }
            case ',': { syn = 30;
                        token[i++] = ch;
                        break; }
            case ';': { syn = 31;
                        token[i++] = ch;
                        break; }
            case '>': { token[i++] = ch;
                        ch = string[++num];
                        if ('=' == ch) {
                            syn = 33;
                            token[i++] = ch;
                            local = num;
                            break;
                        }
                        syn = 32;
                        break; }
            case '<': { token[i++] = ch;
                        ch = string[++num];
                        if ('=' == ch) {
                            syn = 35;
                            token[i++] = ch;
                            local = num;
                            break;
                        }
                        syn = 34;
                        break; }
            case '!': { token[i++] = ch;
                        ch = string[++num];
                        if ('=' != ch) {
                            return FALSE;
                        }
                        syn = 37;
                        token[i++] = ch;
                        local = num;
                        break; }
            case '\n': { syn = 0;
                         token[i++] = ch;
                         break; }
            default: { syn = -1;
                       break; }
        }
    }
    token[i] = '\0';
    return local;
}