Passing DataFrame with large integer value in it throws cryptic error

[davidanthoff]

Here is the repo:

using DataFrames, RCall

df = DataFrame(x=[2^63 - 1, 2])

R"$df"

[randy3k]

It is expected due to the limitation of integer size, R has a limitation of 2^32-1 2^31-1. It also happens for plain julia integer RObject(2^63 - 1).
For those large integer values, you should convert them to float numbers/float vectors.

[davidanthoff]

Ah, interesting! Maybe RCall could throw a better error message that points people in the right direction if this happens?

[simonbyrne]

Maybe we could define our own RConversionException?

[randy3k]

we need to first investigate the time/resources required for checking integer overflow or any other conversion validity.

However, making our own exceptions is not a bad idea.