如何让sscanf读到'\0‘个字符

Question

我希望名称中的所有字符都保留在行中，直到'\0'。

#include <stdio.h>

int main(){
    char line[] = "1999-08-01,14.547,0.191,United Kingdom";
    unsigned int year, month, day;
    float temp, uncertainty;
    char name[100];
    sscanf(line, "%u - %u - %u, %f , %f , %s", &year, &month,
                       &day, &temp, &uncertainty, name);
    printf("%u-%u-%u,%lf,%lf,%s\n", year, month, day, temp, uncertainty, name);
}

我可以让它这样工作：

#include <stdio.h>
#include <string.h>
#include <stdlib.h>
int main(){
    char line[] = "1999-08-01,14.547,0.191,United Kingdom";
    char* newline = malloc(strlen(line) + 2);
    strcpy(newline, line);
    newline[strlen(newline)] = '\n';
    newline[strlen(newline)] = '\0';
    unsigned int year, month, day;
    float temp, uncertainty;
    char name[100];
    sscanf(line, "%u - %u - %u, %f , %f , %[^\n]", &year, &month,
                       &day, &temp, &uncertainty, name);
    printf("%u-%u-%u,%lf,%lf,%s\n", year, month, day, temp, uncertainty, name);
}

但我觉得这很不雅观。

Answer 1

这应该是可行的：

char line[] = "1999-08-01,14.547,0.191,United Kingdom";
unsigned int year, month, day;
float temp, uncertainty;
char name[100];
sscanf(line, "%u - %u - %u, %f , %f , %99[^\n]", &year, &month,
                   &day, &temp, &uncertainty, name);
printf("%u-%u-%u,%lf,%lf,%s\n", year, month, day, temp, uncertainty, name);

将不会找到'\n'，但由于不会达到99的限制，sscanf将继续读取，直到字符串结束标记。

Answer 2

sscanf不是最优雅的界面，但它有很多特性。其中之一是能够找出您在输入字符串中的位置，这使您可以提取(或只是指向)“输入的其余部分”。

例如，after；

int nchar = -1;
int nfield = sscanf(line, "%u - %u - %u, %f , %f , %n", &year, &month,
                    &day, &temp, &uncertainty, &nchar);

nchar将在name字段的line中包含偏移量(除非它仍然是-1，表示sscanf无法匹配格式字符串)。如果该字段扩展到line的末尾，则可以直接使用它(line + nchar)，或者在检查它不是太长之后将其复制到不同的字符串中。

如果line与其名称相反，包含多行，并且您希望提取直到换行符的字符串，则可以使用两种%n格式，并在其间使用%*[^\n] (星号会抑制副本，以避免溢出问题)：

char name[NAME_MAX + 1];
int nstart = -1, nend = -1;
int nfield = sscanf(line, "%u - %u - %u, %f , %f , %n%*[^\n]%n", &year, &month,
                    &day, &temp, &uncertainty, &nstart, &nend);
if (nend > 0) {
  if (nend - nstart <= NAME_MAX) {
    memcpy(name, line + nstart, nend - nstart);
    name[nend - nstart] = 0;
  }
  else {
    /* name is too long */
  }
}
else if (nstart > 0) {
  /* Name was 0 bytes long. Sscanf requires that %[ match at least
   * one character; if not, it fails the scan.
   */
  name[0] = 0; /* Perhaps you wanted to signal an error
}
else {
  /* Line didn't match format */
}

显然，当我知道它有多大时，我可以通过动态分配缓冲区来避免使用固定长度的缓冲区和检查溢出：

char* name = NULL;
// ...
if (nend > 0) 
  name = strndup(line + nstart, nend - nstart);

// or, if you don't like strndup
//   name = malloc(nend - nstart + 1);
//   memcpy(name, line + nstart);
//   name[nend - nstart] = 0;

如果您真正想要的是一个动态分配的字符串，并且您有一个与Posix兼容的sscanf，那么可以通过使用m长度修饰符来避免这个麻烦，这是最全面、最简单的解决方案。

char* name = NULL;
int nfield = sscanf(line, "%u - %u - %u, %f , %f , %m[^\n]", &year, &month,
                    &day, &temp, &uncertainty, &name);

有关详细信息，请参阅您的sscanf手册页。在动态分配name的所有情况下，不要忘记在使用完它后释放它。

Answer 3

几天前，我在阅读2003年版的"Unix系统编程:通信，并发，线程“一书中的第二章，我研究了一个例子，这个例子导致用自定义分隔符(它可以是，或_或空格或其他)将字符串拆分成标记。它使用了strtok() C库函数。下面的示例以某种方式进行了调整以满足您的需求。我会给出两个文件：

makeargv.c

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

/* frees all the memory that was allocated by makeargv */
void freemakeargv(char **argv)
{
   if (argv == NULL)
      return;
   if (*argv != NULL)
      free(*argv);
   free(argv);
}
/* Now the function that breaks string s into tokens */ 
int makeargv(const char *s, const char *delimiters, char ***argvp) 
{
   int error;  int i;
   int numtokens;
   const char *snew;
   char *t;
   if ((s == NULL) || (delimiters == NULL) || (argvp == NULL)) 
    { errno = EINVAL; return -1; }
   *argvp = NULL; /* so that a failed call to malloc,will leave it NULL */
      /* now we consume any initial delimiters characters of input s */
   snew = s + strspn(s, delimiters);   /* snew is real start of string */
   if ((t = malloc(strlen(snew) + 1)) == NULL)
      return -1;
   strcpy(t, snew);
   numtokens = 0;
   if (strtok(t, delimiters) != NULL)  /* count number of tokens in s */
      for (numtokens = 1; strtok(NULL, delimiters) != NULL; numtokens++) ;
      /* next,create argument array for ptrs to the tokens */
   if ((*argvp = malloc((numtokens + 1)*sizeof(char *))) == NULL) 
    {
      error = errno;
      free(t);
      errno = error;
      return -1;
    }
      /* now insert pointers-to-tokens into the argument array */
   if (numtokens == 0)  free(t);
    else 
    {
      strcpy(t, snew);
      **argvp = strtok(t, delimiters);
      for (i = 1; i < numtokens; i++)
          *((*argvp) + i) = strtok(NULL, delimiters);
    }
    *((*argvp) + numtokens) = NULL;      /* append final NULL pointer */
    return numtokens;
}

main.c

#include <stdio.h>
#include <stdlib.h>
int makeargv(const char *s, const char *delimiters, char ***argvp);
void freemakeargv(char **argv);

int main()
{
    char delim[] = ",";
    int i, numtokens;
    char **myargv;  /* memory will be allocated dynamicaly, and has to bee freed before exit */

    char line[] = "1999-08-01,14.547,0.191,United Kingdom-UK";
    if ((numtokens = makeargv(line, delim, &myargv)) == -1) /* nakeargv() allocates memory for myargv, it may fail */
    {
      fprintf(stderr, "Failed to construct an argument array for %s\n", line);
      return 1;
    }
   printf(" The argument array contains:\n");
   for (i = 0; i < numtokens; i++)
      printf("%d:%s\n", i, myargv[i]);
   freemakeargv(myargv);    /* do not forget to free the memory! */
   return 0;
}

gcc -Wall -std=c99 -o令牌器main.c makeargv.c

并运行它

./tokenizer