c - 在 C 中转换转义序列

Question

我有一个来自某些用户输入的名为 ArrayA 的字符数组，其中可能包含转义序列字符。我想将 ArrayA 逐个字符复制到 ArrayB 中。话虽如此，我将如何将 ArrayA 中的转义字符表示为 ArrayB？如果我要正确打印 ArrayB，简单地复制字符实际上不会将 \t 转换为选项卡吗？它只是将 \t 作为一个字符而不是正确的？我可以将 0x09 输入到 ArrayB 中以获得 \t 转义序列，以便在打印 ArrayB 时它实际上会在数组中的特定位置打印一个选项卡吗？

Answer 1

功能cstrlit_chr()

这个怎么样：

/* Convert C Character Literal in (str..end] (excluding surrounding quotes) */
/* to character, returning converted char or -1 if string is invalid. */

/* Convert string containing C character literal to character value */
/* Returns -1 if character literal is invalid, otherwise 0x00..0xFF */
/* Does not support extension \E for ESC \033. */
/* Does not support any extension for DEL \177. */
/* Does not support control-char notation ^A for CTRL-A \001. */
/* Accepts \z as valid z when z is not otherwise special. */
/* Accepts \038 as valid CTRL-C \003; next character starts with the 8. */
/* Accepts \x3Z as valid CTRL-C \003; next character starts with the Z. */
/* Treats invalid octal escape \8 or \9 as 8 or 9 */
int cstrlit_chr(const char *str, const char *end, char const ** const eptr)
{
    unsigned char u;
    int rv;

    if (str >= end)
        rv = -1;    /* String contains no data */
    else if ((u = *str++) != '\\')
        rv = u;
    else if (str == end)
        rv = -1;    /* Just a backslash - invalid */
    else if ((u = *str++) == 'x')
    {
        /**
        ** Hex character constant - \xHH or \xH, where H is a hex digit.
        ** Technically, can be \xHHH too, if CHAR_BIT > 8; this nicety
        ** is being studiously ignored.
        */
        int x1;
        int x2;
        if (str == end)
            rv = -1;
        else if ((x1 = basedigit(*str++, 16)) < 0)
        {
            rv = -1;        /* Invalid hex constant */
            str--;
        }
        else if (str == end)
            rv = x1;        /* Single digit hex constant */
        else if ((x2 = basedigit(*str++, 16)) < 0)
        {
            rv = x1;        /* Single-digit hex constant */
            str--;
        }
        else
            rv = (x1 << 4) | x2;    /* Double-digit hex constant */
    }
    else if (isdigit(u))
    {
        /**
        ** Octal character constant - \O or \OO or \OOO, where O is an
        ** octal digit.  Technically, the constant extends for an
        ** indefinite number of octal digits; this nicety is being
        ** studiously ignored.  Treat \8 as 8 and \9 as 9.
        */
        int o1;
        int o2;
        int o3;
        if ((o1 = basedigit(u, 8)) < 0)
            rv = u; /* Invalid octal constant (\8 or \9) */
        else if (str == end)
            rv = o1;    /* Single-digit octal constant */
        else if ((o2 = basedigit(*str++, 8)) < 0)
        {
            rv = o1;    /* Single-digit octal constant */
            str--;
        }
        else if (str == end)
            rv = (o1 << 3) | o2;    /* Double-digit octal constant */
        else if ((o3 = basedigit(*str++, 8)) < 0)
        {
            rv = (o1 << 3) | o2;    /* Double-digit octal constant */
            str--;
        }
        else if (o1 >= 4)
            rv = -1;                /* Out of range 0x00..0xFF (\000..\377) */
        else
            rv = (((o1 << 3) | o2) << 3) | o3;
    }
    else
    {
        /* Presumably \a, \b, \f, \n, \r, \t, \v, \', \", \? or \\ - or an error */
        switch (u)
        {
        case 'a':
            rv = '\a';
            break;
        case 'b':
            rv = '\b';
            break;
        case 'f':
            rv = '\f';
            break;
        case 'n':
            rv = '\n';
            break;
        case 'r':
            rv = '\r';
            break;
        case 't':
            rv = '\t';
            break;
        case 'v':
            rv = '\v';
            break;
        case '\"':
            rv = '\"';
            break;
        case '\'':
            rv = '\'';
            break;
        case '\?':
            rv = '\?';
            break;
        case '\\':
            rv = '\\';
            break;
        case '\0':  /* Malformed: solitary backslash followed by NUL */
            rv = -1;
            break;
        default:
            rv = u; /* Nominally invalid: \X but X not special; return X. */
            break;
        }
    }
    if (eptr != 0)
        *eptr = str;
    return(rv);
}

它处理 C89 字符序列；它不处理 Unicode（通用）字符（\uXXXX或\U00XXXXXX）。

功能basedigit()

/*
** Convert character to digit in given base,
** returning -1 for invalid bases and characters.
*/
int basedigit(char c, int base)
{
    int             i;

#if (('z' - 'a') != 25 || ('Z' - 'A') != 25)
#error Faulty Assumption
This code assumes the code set is ASCII, ISO 646, ISO 8859, or something similar.
#endif /* Alphabet test */
    if (base < 2 || base > 36)
        i = -1;
    else if (c >= '0' && c <= '9')
        i = c - '0';
    else if (c >= 'A' && c <= 'Z')
        i = c - 'A' + 10;
    else if (c >= 'a' && c <= 'z')
        i = c - 'a' + 10;
    else
        i = -1;
    return((i < base) ? i : -1);
}

示例使用

/* Sample usage */

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

int main(void)
{
    const char  data[] = "ab\\xFF\\03\\7\\377\\t\\?\\'\\\\yz";
    const char *end    = data + strlen(data);
    const char *start  = data;
    const char *next;
    int   c;
    while ((c = cstrlit_chr(start, end, &next)) != -1)
    {
        char buffer[20];
        snprintf(buffer, sizeof(buffer), "[[%.*s]]", (int)(next-start), start);
        printf("%3d (0x%.2X) %-10s - [[%s]]\n",  c, c & 0xFF, buffer, next);
        start = next;
    }
    return 0;
}

请注意，扫描的范围由指向开始字符的指针标识，结束由指向范围结束后的字符的指针标识（在示例中，'\0'在字符串的末尾，但该函数适用于任意数据并且不需要它以空结尾。输入字符串在源代码中具有双反斜杠，因此实际字符串包含单个反斜杠。

如果c == -1，则转换失败。否则，c包含字符，并且end是指向转换完成位置的指针。