c - 在 C 中拆分不带引号的字符串

Question

我正在编写一个将字符串拆分为指向指针的指针的函数，如果分隔符是空格，我只想拆分不在引号内的单词。例如 Hello world "not split" 应该返回

Hello
world
"not split"

函数以某种方式拆分了引号内的单词而不拆分引号外的单词。

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

int is_quotes(char *s)
{
    int i;
    int count;

    i = 0;
    count = 0;
    while (s[i])
    {
        if (s[i] == '"')
            count++;
        i++;
    }
    if (count == 0)
        count = 1;
    return (count % 2);
}

int count_words(char *s, char sep)
{
    int check;
    int i;
    int count;

    check = 0;
    if (sep == ' ')
      check = 1;
    i = 0;
    count = 0;
    while (*s && *s == sep)
        ++s;
    if (*s)
        count = 1;
    while (s[i])
    {
        if (s[i] == sep)
        {
          if (!is_quotes(s + i) && check)
          {
            i += 2;
            while (s[i] != 34 && s[i])
                i++;
          }
          count++;
        }
        i++;
    }
    return (count);
}

char    *ft_strsub(char const *s, unsigned int start, size_t len)
{
    char    *sub;

    sub = malloc(len + 1);
    if (sub)
        memcpy(sub, s + start, len);
    return (sub);
}

char        **ft_strsplit(char const *s, char c)
{
    int     words;
    char    *start;
    char    **result;
    int     i;

    words = count_words((char *)s, c);
    if (!s || !c || words == 0)
        return (NULL);
    i = 0;
    result = (char **)malloc(sizeof(char *) * (words + 1));
    start = (char *)s;
    while (s[i])
    {
        if (s[i] == c)
        {
            if (is_quotes((char *)s + i) == 0 && c == ' ')
            {
                i += 2;
                while (s[i] != '"' && s[i])
                    i++;
                i -= 1;
            }
            if (start != (s + i))
                *(result++) = ft_strsub(start, 0, (s + i) - start);
            start = (char *)(s + i) + 1;
        }
        ++i;
    }
    if (start != (s + i))
        *(result++) = ft_strsub(start, 0, (s + i) - start);
    *result = NULL;
    return (result - words);
}

int main(int argc, char **argv)
{
    if (argc > 1)
    {
        char **s;
        s = ft_strsplit(argv[1], ' ');
        int i = 0;
        while (s[i])
            printf("%s\n", s[i++]);
    }
  return 0;
}

当我用 hello world "hello hello" 运行这段代码时，我得到以下结果

hello world
"hello
hello"

Answer 1

你需要一个有两种状态的状态机，on quote 和 off quote。当您点击报价时，翻转状态。当你打一个空格时，如果不引用则转换为换行符，而不是如果引用。(您很快就会想要让它更精细以允许字符串转义等，状态机方法可以扩展到那个)。