longs?

[Ironholds]

The codebase now includes those. Do we have that supported, R-side?

[hrbrmstr]

"those"?

[hrbrmstr]

oh, "longs". perhaps a bit more exposition might help clarify the question for me.

[Ironholds]

Oop, yep. So we have a lot of functions returning say std::vector < std::long > ... or accepting those as arguments. Do we have R-side support uniformly for that? I wasn't aware base was now cool with longs as a general thing.

[hrbrmstr]

As long as it isn't an unsigned long, aye.

is.integer(2147483647L)
## TRUE

is.integer(4294967295L)
## FALSE

(I did not test that everywhere)

which means we may need to re-think the params/returns since IPv4 conversions are going to beed the full unsigned space.

[Ironholds]

Mmmn yeah. Would using doubles make more sense/give us enough space? Also means we can build in the NA support.