Compiler/MSP430FR2433: Issue with sprintf with unsigned int and float

maanideep kumar

Part Number: MSP430FR2433

Tool/software: TI C/C++ Compiler

Hi,

If I use --printf_support=nofloat i'm not able to use float value. If I use minimal mode I'm not able to use unsigned integer. If I use full mode getting some memory usage error. But my requirement is to display float and unsigned interger using UART on Teraterm.

Thankyou

over 7 years ago

0 Bruce McKenney over 7 years ago

Guru 95935 points

One fairly simple trick I've used is to scale the float into an integer, then split it into Left and Right of the decimal point for formatting. Schematically, something like:

int intval = floatval * 100;        // Save two decimal places
int intpart = intval / 100;         // Integer part (left of decimal point)
int decpart = intval % 100;         // Decimal digits (right)
sprintf(str,"%d.%02d", intpart, decpart);   // Use snprintf(), not sprintf()

I can't find any of that code, but I seem to recall a few corner cases (-0, maybe?). That's the essence, anyway.

0 maanideep kumar over 7 years ago in reply to Bruce McKenney

Intellectual 265 points

0 Bruce McKenney over 7 years ago in reply to maanideep kumar

Guru 95935 points

I thought you were using "nofloat" rather than "minimal". I'm pretty sure "nofloat" supports field widths.

0 Tomasz Kocon over 7 years ago in reply to maanideep kumar

Guru 26200 points

Line 4:
sprintf(str,"%d.%02d", intpart, decpart); // Use snprintf(), not sprintf()
snsprintf() is not used.

if (decpart < 10)
sprintf(str,"%d.0%d", intpart, decpart);
else
sprintf(str,"%d.%d", intpart, decpart);

Probably less code than implementing / using modifiers.

**Attention** This is a public forum

MSP low-power microcontrollers

MSP low-power microcontroller forum

Compiler/MSP430FR2433: Issue with sprintf with unsigned int and float