Use of Sprintf increases code size!!

Look at the documentation of sprintf in the manual: just the description of all the formats and options runs to several pages - and then there's the handling of the variable-length arguments, parsing the format string, etc...

All this versatility & mighty functionality doesn't come cheap - it obviously take a lot of code to implement all that stuff!

"Is there any alternate for Sprintf ??"

Sure - if you just want to create the signed decimal string representation of an 'int', then write a function which does just that and nothing else!

Hi Andy,

Thanks for the explanation.
"Sure - if you just want to create the signed decimal string representation of an 'int', then write a function which does just that and nothing else!"

Is there any standard function available for conversion from integer to string other than sprintf.

I want to take a 3 digit decimal integer value in a string and display the three decimals.

Thanks

Aniket

There are a variety of routines in the standard C library that convert strings to binary (atoi, strtol, sscanf). But there's a dearth of routines that go the other way, integer to string. Routiens like itoa() are common extensions, but not actually part of the C89 standard, and not supplied with the Keil libraries.

It's not that hard to write. Off the top of my head:

// do not call on empty strings; len must be >= 1
void StrReverse (char* s, U8 len)
    {
    char tmpChar;
    U8 i;

    for (i = 0; i < len / 2; ++i)
        {
        tmpChar = s[i];
        s[i] = s[len - 1 - i];
        s[len - 1 - i] = tmpChar;
        }

    } // StrReverse



U8 U16toString (U16 val, char* s)
    {
    U8 len;

    len = 0;
    do
        {
        s[len++] = '0' + val % 10;
        val /= 10;
        }
    while (val);
    s[len] = 0; // null terminate string

    StrReverse (s, len);

    return len;
    } // U16toString

You could also build the string most-significant-digit first, but it seemed to me like there would be a lot more division and multiplication involved that way, and that reversing the result would be faster.

(Once upon a time, probably years ago now, Jon Ward asked if there was any reason people would use both quotient and remainder of a mod operation. Here's another example where both are used. The compiler could usefully retain both results and avoid having to divide by ten twice.)