Mickey Mathieson a écrit :
> --- David <[EMAIL PROTECTED]> wrote:
>
>
>> Mickey Mathieson a écrit :
>>
>>>> char* reverse_r( char*, char* );
>>>>
>>>> char *RevStr(int count, char *str)
>>>> {
>>>> return reverse_r( str, str + count - 1 );
>>>> }
>>>>
>>>> char* reverse_r( char* begin, char* end )
>>>> {
>>>> if ( end <= begin ) return begin;
>>>> char c = *end; *end = *begin; *begin = c;
>>>>
>> //swap
>>
>>>> reverse_r( begin+1, end-1 );
>>>> return begin;
>>>> }
>>>>
>>>>
>>>>
>>>>
>>> Then i could do the same thing with my version. I
>>> wounder which one is faster. I am not familiar
>>>
>> with
>>
>>> the program/technique you used to compare the
>>> functions. Can you let me know and/or test them.
>>>
>>> Thanks
>>> Mickey
>>>
>>> char *RevStrB(int bcount, int count, char *str);
>>>
>>> char *RevStr(char *str)
>>> {
>>> int count = strlen(str);
>>> return(RevStrB(count, count, str));
>>>
>>> }
>>>
>>> char *RevStrB(int bcount, int count, char *str)
>>> {
>>> char Value;
>>>
>>> if (!count) return(str);
>>> Value = str[count-1];
>>> RevStrB(bcount, count-1, str);
>>> str[bcount - count] = Value;
>>> return(str);
>>> }
>>>
>>>
>> Mine wins, she twice must fast ( or more exactly
>> twice less slow)
>>
>> In fact, you just have to count the number of calls
>> for each functions.
>> Your will made strlen(str) calles since you
>> decrement count by 1 each time
>> Mine will made strlen(str)/2 calles since I
>> increment the beginning by 1
>> and decrement the end by 1.
>> (at least on the computer I have here)
>>
>> David
>>
>>
>>
>
> But how about the prcessing each function does - your
> makes 1/2 the calls but how about while in the
> function itself.
>
?
> Your is faster if just counting the number of time
> the recursion function is called. But how about total
> processing time.
>
>
> Mickey
>
>
>
"Faster" using the test suite I post.
David
