Fixes mpaland#120 (and thus fixes #24): When printing am "%f" form wi…

…thin an "%e" printing, we now check whether the rounding of the fractional part according to the precision limits breaks the "%e" constraint of a single-digit integral part.

printf.c

@@ -575,7 +575,17 @@ static size_t sprint_exponential_number(out_fct_type out, char* buffer, size_t i
 
  // output the floating part
  const size_t start_idx = idx;
- idx = sprint_decimal_number(out, buffer, idx, maxlen, negative ? -abs_number : abs_number, precision, fwidth, flags, buf, len);
+ struct double_components number_ = get_components(negative ? -abs_number : abs_number, precision);
+ // For "%e" notation, the integral part must be between 1 and 9; but the rounding can potentially
+ // bring it up from, say, 9.999something to 10 - in which case we must "steal" this extra 10 in
+ // favor of the exponent
+ if (!(flags & FLAGS_ADAPT_EXP) && number_.integral >= 10) {
+ number_.integral = 1;
+ number_.fractional = 0;
+ exp10++;
+ }
+ // TODO: Do we need to check for number_.integral being 0?
+ idx = sprint_broken_up_decimal(number_, out, buffer, idx, maxlen, precision, fwidth, flags, buf, len);
 
  // output the exp10 part
  if (minwidth) {