SEGGER Forum

Hey...

I'm trying to use fscanf & fprintf in code compiled with Embedded Studio and run on a Nordic nRF52840.

I managed to derive that fprintf ultimately calls __write with a file handle as a parameter, and so I'm able to differentiate between stdout and stderr and send them to the correct UART.

But it is unclear how to create the correct code for scanf (or fscanf).

I created the workaround function as suggested in the manual (see below) and it works - but I had to make up my external declaration for __vfscanf and at the end of the day I'd just as soon use the built-in fscanf rather than a workaround.

Is there a way to find/populate the system equivalent of the iod below?


int uart0_scanf(const char *fmt, ...)
{
__stream_scanf_t iod;
va_list a;
int n;
va_start(a, fmt);
iod.is_string = 0;
iod.getc_fn = uart0_getc;
iod.ungetc_fn = uart0_ungetc;
n = __vfscanf((__stream_scanf_t *)&iod, (const unsigned char *)fmt, a);
va_end(a);
return n;
}

Thanks for the quick reply, Nino.


I've done some further research of my own, and have found the solution.


I discovered last week, using the debugger, that fprintf ultimately calls a function called __write that is defined in __libc.h



It turns out that there is a corresponding function called __read that is called by fscanf