Question

我创建了一个程序，用于将非终端与终端之间的编程语言分开。枚举器跟踪非终结符的标记名称。非终结ID设置为0，其他所有其他ID都枚举（应该如此）。然而，当程序找到非终结符时，它会打印一个奇怪的值1953167781，尽管有些值确实为0.看起来像1953年出现的那些......就像那样排序：open int num = 7 ，num是非终结者。 num不被视为0但是被视为1953167781.但是，如果它的声明如下：guarded class test{}非终结test为0，而不是1953167781. 是否有出现这种现象的可能原因？提前感谢您的反馈！下面是输出图片和发生过程的示例。

修改

例如，让我们创建标题来保存我们的声明：

#ifndef __TOK_H__
#define __TOK_H__

#define MAX (1024)

typedef enum {
    ID = 0,OPEN,CLOSED,GUARDED,ARTIFICIAL,STATIC,GLOBAL,CONT,SUPER,INT,FLOAT,CHAR,STRING,
    BOOL,COLLECTION,CLASS,FUNCT,METHOD,STRUCT,ENUM,IF,ELSE,_OR_,DO,UNTIL,UNLESS,FOR,
    FOREACH,IN,TRY,CATCH,EXCEPTION,RETURN,SKIP,BREAK,TERM,NEW,CALL,TRU,FALS,NIL,
    INUMBER,FNUMBER,VCHAR,VSTRING,UNDEF,LT,GT,LE,GE,EQ,NE,AND,OR,ADD,ADDE,INCRE,SUB,SUBE,
    DECRE,MULT,MULTE,DIV,DIVE,MOD,MODE,EX,EXE,INC,ASI,NOT,INHER,DOT,COM,LP,RP,LB,RB,LBR,
    RBR,APO,QUO,SEMCO,EOFT,NAT
} tokentype;

typedef struct {
    char str[MAX];
    tokentype type;
} token;

void sscan(FILE *);
token generate(FILE *fp);

#endif

这也是：

#ifndef __RESERVED_H__
#define __RESERVED_H__

char *keywords[] = { //40 keywords/reserved words
    "open","closed","guarded","artificial","static","global","cont","super",
    "int","float","char","string","bool","collection","class","funct","method",
    "struct","enum","if","else","or","do","until","unless","for","foreach","in",
    "try","catch","exception","return","skip","break","term","new","call","true",
    "false","null",
};
char *relative_operators[] = { //8 relops
    "<",">","<=",">=","==","!=","&&","||",
};
char operators[6] = { //6 operators
    "+-*/%^",
};
char delimeters[11] = { //11 delimeters
    ".,(){}[];'\"",
};
char unique_operators[] = { //5 unops
    "#!&|="
};

#endif

然后（虽然没有任何效率）扫描仪和分隔符：

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

#include "reserved.h"
#include "tok.h"

int linenum = 1;
char ch;
token TOKEN;

int isLangPunct(char ch);
int isKEY(char *word);
int isDELIM(char ch);
int isRELOP(char ch);
int isUNIQUE(char ch);
int isOPERA(char ch);
tokentype getKey(char *word);
tokentype getDel(char ch);

void sscan(FILE *fp) {
    while ((ch = fgetc(fp)) != EOF) {
        if (ch == '\n') {
            linenum++;
            continue;
        } else if (isLangPunct(ch) || isalnum(ch) || isspace(ch)) {
            continue;
        } else {
            fprintf(stderr, "Undefined character: %c at line %d.\n", ch, linenum);
        }
    }
    rewind(fp); linenum = 1;
    do {
        generate(fp);
    } while (TOKEN.type != EOFT);
}
int isLangPunct(char ch) {
    int r = 0;
    if (ch == '#' || ch == '|' || ch == '&' || ch == '=' || ch == '%' ||
        ch == '!' || ch == '+' || ch == '-' || ch == '*' || ch == '/' ||
        ch == '.' || ch == ',' || ch == '(' || ch == ')' || ch == ';' ||
        ch == '{' || ch == '}' || ch == '[' || ch == ']' || ch == '\'' ||
        ch == '"' || ch == '<' || ch == '>' || ch == '^') 
    {
        r = 1;
    } return r;
}
token generate(FILE *fp) {
    char *word = malloc(sizeof(char) * MAX);
    char *number = malloc(sizeof(char) * MAX);
    int wi = 0, ni = 0;
    while((ch = fgetc(fp)) != EOF) {
        if (ch == '\n') {
            linenum++; continue;
        } else if (ch == '\t' || ch == ' ' || ch == '\r') {
            continue;
        } else if (isalpha(ch)) {
            do {
                word[wi++] = ch;
            } while(isalpha(ch = fgetc(fp)));
            word[wi] = '\0';
            wi = 0;
            strcpy(TOKEN.str, word);
            if (isKEY(word)) {
                TOKEN.type = getKey(word);
            } else {
                TOKEN.type = ID;
            }
            fseek(fp, -1, SEEK_CUR);
            printf("%d ", (int)TOKEN.type);
            return TOKEN;
        } else if (isdigit(ch)) {
            do {
                number[ni++] = ch;
            } while(isdigit(ch = fgetc(fp)));
            if (ch == '.') {
                do {
                    number[ni++] = ch;
                } while(isdigit(ch = fgetc(fp)));
                TOKEN.type = FNUMBER;
            } else {
                TOKEN.type = INUMBER;
            }
            number[ni] = '\0';
            ni = 0;
            strcpy(TOKEN.str, number);
            printf("%s ", TOKEN.str);
            fseek(fp, -1, SEEK_CUR);
            return TOKEN;
        } else if (isDELIM(ch)) {
            TOKEN.type = getDel(ch);
            char *str = &ch;
            //working on getting strings and chars to have
            //their own value types
            strcpy(TOKEN.str, str);
            printf("%s ", TOKEN.str);
            return TOKEN;
        } else if (isRELOP(ch) || isUNIQUE(ch)) {
            switch (ch) {
                case '<':
                    if ((ch = fgetc(fp)) == '=') {
                        TOKEN.type = LE;
                        strcpy(TOKEN.str, "<=");
                    } else {
                        TOKEN.type = LT;
                        strcpy(TOKEN.str, "<");
                        fseek(fp, -1, SEEK_CUR);
                    } break;
                case '>':
                    if ((ch = fgetc(fp)) == '=') {
                        TOKEN.type = GE;
                        strcpy(TOKEN.str, ">=");
                    } else {
                        TOKEN.type = GT;
                        strcpy(TOKEN.str, ">");
                        fseek(fp, -1, SEEK_CUR);
                    } break;
                case '=':
                    if ((ch = fgetc(fp)) == '=') {
                        TOKEN.type = EQ;
                        strcpy(TOKEN.str, "==");
                    } else {
                        TOKEN.type = ASI;
                        strcpy(TOKEN.str, "=");
                        fseek(fp, -1, SEEK_CUR);
                    } break;
                case '!':
                    if ((ch = fgetc(fp)) == '=') {
                        TOKEN.type = NE;
                        strcpy(TOKEN.str, "!=");
                    } else {
                        TOKEN.type = NOT;
                        strcpy(TOKEN.str, "!");
                        fseek(fp, -1, SEEK_CUR);
                    } break;
                case '&':
                    if ((ch = fgetc(fp)) == '&') {
                        TOKEN.type = AND;
                        strcpy(TOKEN.str, "&&");
                    } else {
                        fprintf(stderr, "Token Error: & missing at line %d.\n", linenum);
                        fseek(fp, -1, SEEK_CUR);
                    } break;
                case '|':
                    if ((ch = fgetc(fp)) == '|') {
                        TOKEN.type = OR;
                        strcpy(TOKEN.str, "||");
                    } else {
                        TOKEN.type = INHER;
                        strcpy(TOKEN.str, "|");
                        fseek(fp, -1, SEEK_CUR);
                    } break;
                case '#':
                    TOKEN.type = INC;
                    strcpy(TOKEN.str, "#");
            }
            printf("%s ", TOKEN.str);
            return TOKEN;
        } else if (isOPERA(ch)) {
            switch (ch) {
                case '+':
                    if ((ch = fgetc(fp)) == '+') {
                        TOKEN.type = INCRE;
                        strcpy(TOKEN.str, "++");
                    } else if (ch == '=') {
                        TOKEN.type = ADDE ;
                        strcpy(TOKEN.str, "+=");
                        fseek(fp, -1, SEEK_CUR);
                    } else {
                        TOKEN.type = ADD ;
                        strcpy(TOKEN.str, "+");
                        fseek(fp, -1, SEEK_CUR);
                    } break;
                case '-':
                    if ((ch = fgetc(fp)) == '-') {
                        TOKEN.type = DECRE;
                        strcpy(TOKEN.str, "--");
                    } else if (ch == '=') {
                        TOKEN.type = SUBE;
                        strcpy(TOKEN.str, "-=");
                        fseek(fp, -1, SEEK_CUR);
                    } else {
                        TOKEN.type = SUB;
                        strcpy(TOKEN.str, "-");
                        fseek(fp, -1, SEEK_CUR);
                    } break;
                case '*':
                    if ((ch = fgetc(fp)) == '=') {
                        TOKEN.type = MULTE;
                        strcpy(TOKEN.str, "*=");
                    } else {
                        TOKEN.type = MULT;
                        strcpy(TOKEN.str, "*");
                        fseek(fp, -1, SEEK_CUR);
                    } break;
                case '/':
                    if ((ch = fgetc(fp)) == '=') {
                        TOKEN.type = DIVE;
                        strcpy(TOKEN.str, "/=");
                    } else if (ch == '/') {
                        do {
                            continue;
                        } while ((ch = fgetc(fp)) != '\n');
                    } else {
                        TOKEN.type = DIV;
                        strcpy(TOKEN.str, "/");
                        fseek(fp, -1, SEEK_CUR);
                    } break;
                case '%':
                    if ((ch = fgetc(fp)) == '=') {
                        TOKEN.type = MODE;
                        strcpy(TOKEN.str, "%=");
                    } else {
                        TOKEN.type = MOD;
                        strcpy(TOKEN.str, "%");
                        fseek(fp, -1, SEEK_CUR);
                    } break;
                case '^':
                    if ((ch = fgetc(fp)) == '=') {
                        TOKEN.type = EXE;
                        strcpy(TOKEN.str, "^=");
                    } else {
                        TOKEN.type = EX;
                        strcpy(TOKEN.str, "^");
                        fseek(fp, -1, SEEK_CUR);
                    } break;
            }
            printf("%s ", TOKEN.str);
            return TOKEN;
        }
    }
    free(word);
    free(number);
    TOKEN.type = EOFT;
    strcpy(TOKEN.str, "\0");
    printf("%s ", TOKEN.str);
    return TOKEN;
}
int isKEY(char *word) {
    int r = 0;
    for (int i = 0; i < 40; i++) {
        if (*word == *keywords[i]) {
            r = 1;
        }
    }
    return r;
}
int isDELIM(char ch) {
    int r = 0;
    for (int i = 0; i < 11; i++) {
        if (ch == delimeters[i]) {
            r = 1;
        }
    }
    return r;
}
int isRELOP(char ch) {
    int r = 0;
    if (ch == '<' || ch == '>') {
        r = 1;
    }
    return r;
}
int isUNIQUE(char ch) {
    int r = 0;
    for (int i = 0; i < 5; i++) {
        if (ch == unique_operators[i]) {
            r = 1;
        }
    }
    return r;
}
int isOPERA(char ch) {
    int r = 0;
    for (int i = 0; i < 6; i++) {
        if (ch == operators[i]) {
            r = 1;
        }
    }
    return r;
}
tokentype getKey(char *word) {
    tokentype type;
    if (!strcmp(word, "open")) { type = OPEN; }
    else if (!strcmp(word, "closed")) { type = CLOSED; }
    else if (!strcmp(word, "guarded")) { type = GUARDED; }
    else if (!strcmp(word, "artificial")) { type = ARTIFICIAL; }
    else if (!strcmp(word, "static")) { type = STATIC; }
    else if (!strcmp(word, "global")) { type = GLOBAL; }
    else if (!strcmp(word, "cont")) { type = CONT; }
    else if (!strcmp(word, "super")) { type = SUPER; }
    else if (!strcmp(word, "int")) { type = INT; }
    else if (!strcmp(word, "float")) { type = FLOAT; }
    else if (!strcmp(word, "char")) { type = CHAR; }
    else if (!strcmp(word, "string")) { type = STRING; }
    else if (!strcmp(word, "bool")) { type = BOOL; }
    else if (!strcmp(word, "collection")) { type = COLLECTION; }
    else if (!strcmp(word, "class")) { type = CLASS; }
    else if (!strcmp(word, "funct")) { type = FUNCT; }
    else if (!strcmp(word, "method")) { type = METHOD; }
    else if (!strcmp(word, "struct")) { type = STRUCT; }
    else if (!strcmp(word, "enum")) { type = ENUM; }
    else if (!strcmp(word, "if")) { type = IF; }
    else if (!strcmp(word, "else")) { type = ELSE; }
    else if (!strcmp(word, "or")) { type = _OR_; }
    else if (!strcmp(word, "do")) { type = DO; }
    else if (!strcmp(word, "until")) { type = UNTIL; }
    else if (!strcmp(word, "unless")) { type = UNLESS; }
    else if (!strcmp(word, "for")) { type = FOR; }
    else if (!strcmp(word, "foreach")) { type = FOREACH; }
    else if (!strcmp(word, "in")) { type = IN; }
    else if (!strcmp(word, "try")) { type = TRY; }
    else if (!strcmp(word, "catch")) { type = CATCH; }
    else if (!strcmp(word, "exception")) { type = EXCEPTION; }
    else if (!strcmp(word, "return")) { type = RETURN; }
    else if (!strcmp(word, "skip")) { type = SKIP; }
    else if (!strcmp(word, "break")) { type = BREAK; }
    else if (!strcmp(word, "term")) { type = TERM; }
    else if (!strcmp(word, "new")) { type = NEW; }
    else if (!strcmp(word, "call")) { type = CALL; }
    else if (!strcmp(word, "true")) { type = TRU; }
    else if (!strcmp(word, "false")) { type = FALS; }
    else if (!strcmp(word, "null")) { type = NIL; }
    return type;
}
tokentype getDel(char ch) {
    tokentype type;
    if (ch == '.') { type = DOT; }
    if (ch == ',') { type = COM; }
    if (ch == '(') { type = LP; }
    if (ch == ')') { type = RP; }
    if (ch == '{') { type = LB; }
    if (ch == '}') { type = RB; }
    if (ch == '[') { type = LBR; }
    if (ch == ']') { type = RBR; }
    if (ch == '"') { type = QUO; }
    if (ch == '\'') { type = APO; }
    if (ch == ';') { type = SEMCO; }
    return type;
}

在上述单独的文件或同一文件中，main()能够打开文件并读取文件，将文件流发送到sscan(FILE *)。对于最后一篇文章，请阅读：

# call Sys;
# call SysIO;

static int num = 7;
cont global float fl = 12.895;
cont char letter = 'h';

collection Program {
    cont closed string aswkey = "abdcbabcdabdbcdab";
    open class quiz {
        bool decision = true;
        artificial method.study (int time) {};
        open quiz() {
            //do something
        };
    };
    class test | quiz {
        bool descision;
        super method.study (int time) {
            if (decision == true) {
                //do something
            } or if (time == 0) {
                //do something
            } else {
                //do nothing
            };
            time = 50;
        };
        guarded test (bool n) {
            descision = n;
        };
    };
    funct.Enter () {
        quiz Quiz = new quiz();
        test Test = new test(false);
        Test.study(60);
        if (something != this) {
            //do something
        };
        term;
    };
};

Answer 1

因此，要了解出了什么问题，让我们来看看发生了什么。在getKey(word)函数中，当一个单词被发送到那里以与关键字进行比较时，它会将tokentype type返回到TOKEN.type声明。但是，如果单词与任何比较都不匹配，那么它仍然仅作为type的变量返回，该变量在我的计算机上具体为地址1953167781。因此，通过向getKey(word)添加最终的else语句，将tokentype type设置为ID，它将所有非终结符与ID匹配。然而，这并不能解释为什么像Sys和SysIO这样的单词也不会成为1953167781.而if(isKEY(word))无法返回false会让人感到有点困惑，它似乎总是返回true 。虽然它必须返回false才能转到else语句，该语句将有效地将所有非终结符设置为ID。但这仍然是一个问题，我们不会讨论，但应该思考：为什么像Sys和SysIO这样的非终结符会成为ID（注意getKey(word)函数会没有新的最终else语句）在一个永远真实的if语句中，当它必须返回false以便发生这种情况时，为什么其他类似的单词不会采用相同的过程？无论如何，它工作正常，我将把它留给专家来弄清楚为什么它做了它在前面的陈述中所做的事情。

C枚举不返回未指定的值

1 个答案: