prices = ["1.6", "0.15", "1.8"]
prices.sum { |price| price.to_f }
But this returns 3.5500000000000003, not 3.55.
Any idea?
Add a comment
|
1 Answer
Floating-point numbers cannot precisely represent all real numbers, and floating-point operations cannot precisely represent true arithmetic operations, this leads to many surprising situations.
I advise reading: https://en.wikipedia.org/wiki/Floating_point#Accuracy_problems
You may want to use BigDecimal to avoid such problems:
require 'bigdecimal'
prices = ["1.6", "0.15", "1.8"]
prices.sum { |price| BigDecimal(price) }
#=> 3.55
4 Comments
Remy Wang
BigDecimal(nil) throws error.
spickermann
Of course. Where is the
nil coming from? In the array in your original question there is no nil.
steenslag
["1.6", "0.15", "1.8", nil].sum(&:to_r).to_f seems to work with nil.
Stefan
You can
require 'bigdecimal/util' and use ["1.6", "0.15", "1.8", nil].sum(&:to_d) instead which treats nil values as 0.0, just like to_f.