Open progval opened 6 months ago
I would.
2 is definitely too little (8/9 = 1.125) so I wouldn't feel comfortable with 3 either.
The complexity is because it tries to fix physics calculous rules, ie. output should have as many significant digits as the input
I'm honestly surprised that this code is so complicated. Would anyone actually complain if we used a fixed (or configurable?) max precision like 2 or 3 decimal places?