How to import really large numbers into R?

Asked Apr 24 '17 at 14:27

Active Apr 24 '17 at 14:27

Viewed 35 times

I tend to read a large number from a string input and then convert it into numeric format. However, I always get wrong output.

#the input string is: "31415926535897932384626433832795"

My code is like this:

input <- readLines("stdin", warn = F)
options(scipen = 10^6)
nums <- as.numeric(input)
cat(nums)

the output is like this, the original number has been modified. I don't get what happened when importing and converting the string into numeric?

31415926535897933290036940242944

asked Apr 24 '17 at 14:27

Elsa Li

The reason can be found here: http://stackoverflow.com/questions/9508518/why-are-these-numbers-not-equal but I don't want to close as duplicate because the answer to importing your numbers isn't in that thread. – Dason Apr 24 '17 at 14:33
Load the package `bit64`. That may fix your issue. – Kristofersen Apr 24 '17 at 14:36
@Dason the linked-to qustion isn't relevant because that's about floating point numbers and OP has an integer. – Robert Dodier Apr 24 '17 at 14:41
1

@QianLi see also the `gmp` package for R. – Robert Dodier Apr 24 '17 at 14:43
1

@RobertDodier And yet it is relevant because they aren't storing it as integer. – Dason Apr 24 '17 at 14:47
@ Kristofersen , can you explain a bit more? I loaded the package bit64, and used the function as.integer64(), still it cannot correctly convert the string into a number. Thanks. – Elsa Li Apr 24 '17 at 14:48
@Dason I think you're needlessly jumping to conclusions here. – Robert Dodier Apr 24 '17 at 16:56
@Robert Dodier `gmp` works perfectly for this. Problem solved! Thanks. – Elsa Li Apr 25 '17 at 13:45

0 Answers0