atoi函数的实现（考虑不同进制、溢出）

2013-07-08 16:21:15

atio就是将ASCII码表示的字符为int型数据，看似简单的问题，实际上需要考虑的很多。对于简单的问题，考虑是否全面就显得特别重要。

小结：

对于该函数的实现要考虑一下几个方面：

1. 输入字符串为NULL；
2. 输入开始是否包含符号'+'、'-'
3. 输入的字符是否合法（对十进制'0' ~'9'为合法的输入）；
4. 计算出的 数值必须为unsigned Int，才能判断溢出；
5. 数据溢出的处理（上溢出时，返回最大正数；下溢出时，返回最大负数）；
6. 输入的字符包含前导的空格；
7. 进制的考虑，考虑16进制以及8进制，通过开头的0或0x判断进制；
8. 对于不同的进制判断的范围不同
   对16进制0123456789ABCDEF或0123456789abcdef
   对8进制01234567
   对10进制0123456789

其中，前4点是最基本的功能，后面两点则算是功能的扩展，根据不同的要求还可以进行扩展，比如加上各种状态标志，表明是否溢出、输入是否合法，具体在函数_atoi_3中有体现。

另外，在编写代码时，还有几点需要注意：

判断是否溢出时，用到的最大数必须为unsigned long int，这样才能表示绝对值最大的负数。

在考虑进制时，比较繁琐，按照下面步骤来做：

根据首个非空字符判断符号：

if (0 == sign)  //判断数值的符号
  {
   if ('+' == *str)
   {
    sign = 1;
    ++str;
   }
   else if('-' == *str)
   {
    sign = -1;
    ++str;
   }
   else if ( isdigit(*str) )
   {
    sign = 1;
   }
  }

接下来，判断进制：

if (0 == base) //判断数值的进制
    {
     if ( '0' == *str && ( 'x' == *(str + 1) || 'X' == *(str + 1) ) )
     {
      base = 16;
      str = str + 2;
     }
     else if ( '0' == *str && !( 'x' == *(str + 1) || 'X' == *(str + 1) ) )
     {
      base = 8;
      str = str + 1;
     }
     else
     {
      base = 10;   //不更新str
     }
    }

 计算出数据后，判断溢出：

if (IsOverflow (IntNumber,sign) == 1)
     {
      return MaxPositive;
     }

if (IsOverflow (IntNumber,sign) == -1)
     {
      return -( ( MaxPositive + 1) );
     }

参考：http://blog.csdn.net/tianshuai11/article/details/7798698

---

代码（测试暂无发现问题，如发现问题，欢迎指正！）：

 #include <iostream>
 #include <cmath>
 using namespace std;

 #define SIZE 100

 //没有区别输入0与\0
 //只能对十进制数进行转换
 //没有处理溢出
 int _atoi_1(char *str)
 {
     if (NULL == str)    //输入合法性检查
     {
         cout<<"NULL input"<<endl;
         exit();
     }

     int sign = ;
     int IntNumber = ;

     while ( *str )
     {
         if ( == sign)
         {
             if ('-' == *str)
             {
                 sign = -;
                 ++str;
             }
             else if ('+' == *str)
             {
                 sign = ;
                 ++str;
             }
             else if (*str >= '' && *str <= '')
             {
                 sign = ;  //不更新str
             }
             else
             {
                 cout<<"invalid input!"<<endl;
                 exit();
             }
         }
         else if (*str >= '' && *str <= '')
         {
             IntNumber = IntNumber* + *str - '';
             ++str;
         }
         else
         {
             cout<<"invalid input!"<<endl;
             exit();
         }
     }

     return (sign * IntNumber);
 }

 //加上溢出处理
 //并可对10进制、16进制以及8进制转换
 //状态标志
 bool IsNotNULL = true;
 bool IsDigit = true;
 bool IsNotOverflow = true;

 int IsOverflow (unsigned long int IntNumber,int sign)
 {
     unsigned long int MaxPositive = (long int) pow( (float),(int)( *sizeof(long int) -  ) ) - ; 

     if ( (sign == ) && !(IntNumber >=  && IntNumber <= MaxPositive ) )   //判断是否溢出
     {
         IsNotOverflow = true;
         return ;
     }

     if ( (sign == -) && !(IntNumber >  && IntNumber <= (MaxPositive + ) ) )   //判断是否溢出
     {
         IsNotOverflow = true;
         return -;
     }

     return ;
 }

 long int _atoi_2(char *str)
 {
     if (NULL == str)
     {
         IsNotNULL = false;
         return ;
     }

     int sign = ;
     unsigned long int MaxPositive = (long int) pow( (float),(int)( *sizeof(long int) -  ) ) - ;
     unsigned long int IntNumber = ;   //要表示最小负数，须用unsigned long int，若用long int，其范围为-2^31到2^31 -1,不能表示2^31
     unsigned int base = ;

     while ( *str )
     {
         while ( ' ' == *str || '\t' == *str)     //忽略前导的空格或TAB
             ++str;

         if ( == sign)  //判断数值的符号
         {
             if ('+' == *str)
             {
                 sign = ;
                 ++str;
             }
             else if('-' == *str)
             {
                 sign = -;
                 ++str;
             }
             else if ( isdigit(*str) )
             {
                 sign = ;
             }
         }
         else
         {
             if ( isdigit(*str) )
             {
                 if ( == base)    //判断数值的进制
                 {
                     if ( '' == *str && ( 'x' == *(str + ) || 'X' == *(str + ) ) )
                     {
                         base = ;
                         str = str + ;
                     }
                     else if ( '' == *str && !( 'x' == *(str + ) || 'X' == *(str + ) ) )
                     {
                         base = ;
                         str = str + ;
                     }
                     else
                     {
                         base = ;   //不更新str
                     }
                 }
                 else
                 {
                     IntNumber = IntNumber * base + *str - '';

                     if (IsOverflow (IntNumber,sign) == )
                     {
                         return MaxPositive;
                     }

                     if (IsOverflow (IntNumber,sign) == -)
                     {
                         return -( ( MaxPositive + ) );
                     }

                     ++str;
                 }
             }
             else
             {
                 IsDigit = false;
                 return ;
             }
         }
     }

     return (sign * IntNumber);   //如果为空串，也即第一个字符即为\0，返回值为0
 }
 //对不同的进制，判断的范围不同
 //对16进制0123456789ABCDEF
 //对8进制01234567
 //对10进制0123456789
 long int _atoi_3(char *str)
 {
     if (NULL == str)    //输入合法性检查
     {
         IsNotNULL = false;
         return ;
     }

     int sign = ;
     unsigned long int IntNumber = ;   //要表示最小负数，须用unsigned long int，若用long int，其范围为-2^31到2^31 -1,不能表示2^31
     unsigned long int MaxPositive = (long int) pow( (float),(int)( *sizeof(long int) -  ) ) - ;
     unsigned int base = ;

     while ( *str )
     {
         while ( ' ' == *str || '\t' == *str)     //忽略前导的空格或TAB
             ++str;

         if ( == sign)  //判断数值的符号
         {
             if ('+' == *str)
             {
                 sign = ;
                 ++str;
             }
             else if('-' == *str)
             {
                 sign = -;
                 ++str;
             }
             else if ( isdigit(*str) )
             {
                 sign = ;
             }
             else
             {
                 IsDigit = false;
                 return ;
             }
         }
         else
         {
             if ( == base)    //判断数值的进制
             {
                 if ( '' == *str && ( 'x' == *(str + ) || 'X' == *(str + ) ) )
                 {
                     base = ;
                     str = str + ;
                 }
                 else if ( '' == *str && !( 'x' == *(str + ) || 'X' == *(str + ) ) )
                 {
                     base = ;
                     str = str + ;
                 }
                 else
                 {
                     base = ;   //不更新str
                 }
             }
             else
             {
                 if ( == base)
                 {
                     if ( isdigit(*str) )
                     {
                         IntNumber = IntNumber * base + *str - '';
                         if (IsOverflow (IntNumber,sign) == )
                         {
                             return MaxPositive;
                         }

                         if (IsOverflow (IntNumber,sign) == -)
                         {
                             return -( ( MaxPositive + ) );
                         }
                         ++str;
                     }
                     else
                     {
                         IsDigit = false;
                         return ;
                     }
                 }
                 else if ( == base)
                 {
                     if ( isdigit(*str) || (*str >= 'a'&& *str <= 'f') || (*str >= 'A'&& *str <= 'F'))
                     {
                         if ( isdigit(*str))
                         {
                             IntNumber = IntNumber * base + *str - '';
                         }
                         else if (*str >= 'a'&& *str <= 'f')
                         {
                             IntNumber = IntNumber * base + *str - 'a' + ;
                         }
                         else if (*str >= 'A'&& *str <= 'F')
                         {
                             IntNumber = IntNumber * base + *str - 'A' + ;
                         }

                         if (IsOverflow (IntNumber,sign) == )
                         {
                             return MaxPositive;
                         }

                         if (IsOverflow (IntNumber,sign) == -)
                         {
                             return -( ( MaxPositive + ) );
                         }
                         ++str;
                     }
                     else
                     {
                         IsDigit = false;
                         return ;
                     }
                 }
                 else
                 {
                     if ( *str>= '' && *str <= '')
                     {
                         IntNumber = IntNumber * base + *str - '';
                         if (IsOverflow (IntNumber,sign) == )
                         {
                             return MaxPositive;
                         }

                         if (IsOverflow (IntNumber,sign) == -)
                         {
                             return -( ( MaxPositive + ) );
                         }
                         ++str;
                     }
                     else
                     {
                         IsDigit = false;
                         return ;
                     }
                 }
             }
         }
     }

     return (sign * IntNumber);   //如果为空串，也即第一个字符即为\0，返回值为0
 }

 //测试程序
 int main()
 {
     char str[SIZE];

     cout<<"size of type int is : "<<sizeof(int)<<endl;   //
     cout<<"size of type long int is : "<<sizeof(long int)<<endl;    //
     cout<<"size of type float is : "<<sizeof(float)<<endl;    //
     cout<<"size of type long double is : "<<sizeof(long double)<<endl;    //
     /*char str[SIZE] = "100278";
     char str[SIZE] = "+1267
     char str[SIZE] = "-12";
     char str[SIZE] = "+";
     char str[SIZE] = "-";
     char str[SIZE] = "";
     char str[SIZE] = "z98";
     char str[SIZE] = "-9s8";*/

     //test _atoi_1...
     cout<<"test _atoi_1..."<<endl;
     cout<<"please enter the string :"<<endl;
     while(cin>>str)
     {
         cout<<"the string is :"<<str<<endl;
         cout<<"the int number is : "<<_atoi_1(str)<<endl;
         cout<<"please enter the string :"<<endl;
     }

     cin.clear();
     cin.sync();
     cout<<endl;

     //test _atoi_2...
     cout<<"test _atoi_2..."<<endl;
     cout<<"please enter the string :"<<endl;
     while(cin>>str)
     {
         cout<<"the string is :"<<str<<endl;
         cout<<"the int number is : "<<_atoi_2(str)<<endl;
         cout<<"please enter the string :"<<endl;
     }

     cin.clear();
     cin.sync();
     cout<<endl;

     //test _atoi_3...
     cout<<"test _atoi_3..."<<endl;
     cout<<"please enter the string :"<<endl;
     while(cin>>str)
     {
         cout<<"the string is :"<<str<<endl;
         cout<<"the int number is : "<<_atoi_3(str)<<endl;
         cout<<"please enter the string :"<<endl;
     }

     return ;
 }

对代码的测试也要注意开头中提到的几点，运行结果：

size of type int is :
size of type long int is :
size of type float is :
size of type long double is :
test _atoi_1...
please enter the string :

the string is :
the int number is :
please enter the string :
+
the string is :+
the int number is :
please enter the string :
-
the string is :-
the int number is : -
please enter the string :

the string is :
the int number is :
please enter the string :

the string is :
the int number is :
please enter the string :

the string is :
the int number is : -
please enter the string :
-
the string is :-
the int number is :
please enter the string :
^Z

test _atoi_2...
please enter the string :

the string is :
the int number is :
please enter the string :
-
the string is :-
the int number is : -
please enter the string :
-
the string is :-
the int number is : -
please enter the string :
0x1a
the string is :0x1a
the int number is :
please enter the string :

the string is :
the int number is :
please enter the string :
0x89
the string is :0x89
the int number is :
please enter the string :
0xaa
the string is :0xaa
the int number is :
please enter the string :
^Z

test _atoi_3...
please enter the string :

the string is :
the int number is :
please enter the string :
-
the string is :-
the int number is : -
please enter the string :
0x1a
the string is :0x1a
the int number is :
please enter the string :
-0xaa
the string is :-0xaa
the int number is : -
please enter the string :

the string is :
the int number is :
please enter the string :

the string is :
the int number is :
please enter the string :
^Z
请按任意键继续. . .