写入char *数组的字符变异

Question

这应该是一个简单的函数（它计算字符串中的唯一字符数），但是我遇到了一个奇怪的问题。请注意，我的代码使用期望只有ASCII字母a-z和A-Z。

int unique_chars(char* my_str) { 
//printf("starting unique_chars\n"); 
    char seen_buffer[52]; // max 52 letters a-z & A-Z 
    int seen_count = 1; // not ever expecting my_str to be NULL 
    int i, j; 
    char next; 
//printf("first char is %c\n", my_str[0]); 
    seen_buffer[0] = my_str[0]; // first char must be unique 

    for (i=1; i<strlen(my_str); i++) { // walk along the rest of my_str 
    next = my_str[i]; 

    if (next >= 97) { 
     next = next - 32; // the next char will always be capital, for convenience 
    } 

    for (j=0; j<seen_count; j++) { // compare next to all the unique chars seen before 
//printf("current char is %c, checking against %c\n", next, seen_buffer[j]); 
     if ((next==seen_buffer[j]) || (next+32==seen_buffer[j])) { 
//printf("breaking\n"); 
     break; // jump to the next char in my_str if we find a match 
     } 
     if (j==seen_count-1) { // at this point, we're sure that next hasn't been seen yet 
//printf("new unique char is %c\n", next); 
     seen_count++; 
     seen_buffer[seen_count] = next; 
//printf("new char val is %c, should be %c\n", seen_buffer[seen_count], next); 
     break; 
     } 
    } 
    } 
    return seen_count; 
} 

int main(int argc, char* argv[]){ 
    char* to_encode = argv[1]; 
    printf("unique chars: %d\n", unique_chars(to_encode)); 
} 

当我用某些字符串调用时，我得到不正确的结果。例如，尝试：

./a.out gghhiijj 

这将产生（和printf的取消注释）：

starting unique_chars 
first char is g 
current char is G, checking against g 
breaking 
current char is H, checking against g 
new unique char is H 
new char val is H, should be H 
current char is H, checking against g 
current char is H, checking against 
new unique char is H 
new char val is H, should be H 
current char is I, checking against g 
current char is I, checking against 
current char is I, checking against H 
new unique char is I 
new char val is I, should be I 
current char is I, checking against g 
current char is I, checking against 
current char is I, checking against H 
current char is I, checking against H 
new unique char is I 
new char val is I, should be I 
current char is J, checking against g 
current char is J, checking against 
current char is J, checking against H 
current char is J, checking against H 
current char is J, checking against I 
new unique char is J 
new char val is J, should be J 
current char is J, checking against g 
current char is J, checking against 
current char is J, checking against H 
current char is J, checking against H 
current char is J, checking against I 
current char is J, checking against I 
new unique char is J 
new char val is J, should be J 

所以，我不断收到在我seen_buffer重复，因为一些空白字符存储，而不是字母字符存在应在那里！然而，当我在写入到seen_buffer后进行比较（即新的字符值是％c，应该是％c \ n）时，显示正确的字符！

任何帮助表示赞赏！

Answer 1

这里，您有一个差一错误：

seen_count++; 
    seen_buffer[seen_count] = next; 

第一个char进去seen_buffer[0]和seen_count设置为1，这意味着，未来新字符进去seen_buffer[2]seen_count增加到2。什么事也没进去seen_buffer[1]（这是空白字符后您继续在你的printfs中看到），当检查一个字符与seen_buffer时，你不会检查你刚才输入的最后一个字符。

交换这些行，它应该工作。

Answer 2

你是不是检查，如果输入的值是到z

A到Z和之间还有你目前的代码可以在焦炭seen_buffer缓冲区溢出[52]所以加边界的代码检查

我觉得你的代码可以更简单

下面是一个简单的算法为您

unsigned int returnUniqueChar (const char *input) 
{ 
    int count[52] = {0}; // initialize all the memory with zero 
    int unique = 0; 

    while (*input != '\0') 
    { 
      if ((input >= 'A' && input <= 'Z') 
      { 
       count[input -'A']++; 
      } 

      else if (input >= 'a' && input <= 'z') 
      { 
       count[input -'a'+ 26]++; 
      } 
      input++; 
    } 

    for (int i = 0; i < 56 && (count[i] == 1) ; i++) 
      unique++; 

    return unique; 

}