I get number representing strings like
248.03500000066338
313.44999999979470
4.2346999999
and I need to round them to something like
248.035
313.45
4.2347
while keeping a fixed relative precision of let's say 6 significant figures. I could do it using Math.log10
, computing what absolute precision is needed, and rounding correspondingly, but I wonder if there's a simple way.
Ideally, the resulting number should be such that it does not produce the trailing nines when converted to string, but this is not needed and maybe impossible.