4 ms·
I dipped my feet into R a few years back, but eventually stopped it because of the way it handles integers. At the time it treated all integers internally as si
by AndyMcConachie 3y ago
I dipped my feet into R a few years back, but eventually stopped it because of the way it handles integers. At the time it treated all integers internally as signed 32-bit and if the number is too large for that it converted it to a float.
I don't know what R does now, but this was a deal breaker for me at the time because I was dealing with really large integers that regularly broke this limit.
- armchairhacker 3y agoIntegers are still only 32 bits. There’s a class which effectively represents 64-bit integers (https://www.rdocumentation.org/packages/csvread/versions/1.2.2/topics/int64 https://www.rdocumentation.org/packages/csvread/versions/1.2...) as well as arbitrary-sized (https://cran.r-project.org/web/packages/gmp/index.html https://cran.r-project.org/web/packages/gmp/index.html, https://www.rdocumentation.org/packages/gmp/versions/0.7-4/topics/bigz https://www.rdocumentation.org/packages/gmp/versions/0.7-4/t...). I will say there are a few pitfalls where the integer bits are unexpectedly converted to something else, but it’s workable.
- nutshell42 3y agoFinally a real reason. A lot of the stuff above was complaining about issues where Python is a lot worse than R, about non-issues or with a fundamental misunderstanding of the language. I'd given up hope of seeing a real weakness named as such :) There is bit64 and doubles being used as 53bit pseudo-integers - but if I needed 64bit integers, R wouldn't be my first choice, definitely.