R equivalent?

[WardBrian]

This has some up a couple times. In the R ecosystem around Stan the closest equivalent is posterior, but it is concerned with a lot more than just IO, and doesn't handle some of the trickier cases like complex numbers or tuples:

• Add a repair_variable_names function posterior#317
• Support for complex numbers posterior#319

This would be useful in e.g. BridgeStan as well, see @avehtari's issue here: roualdes/bridgestan#202

Also pinging @andrjohns as he has mentioned interest in one before

[avehtari]

posterior package aims to be agnostic for the inference engine and is relying on rstan, rstanarm, and CmdStanR to provide something that can be converted to a posterior object. It would be useful to have stanio R functions to handle Stan specific things. Alternative place for such functions would be rstantools and thus pinging also @jgabry

[WardBrian]

Putting them in rstantools would probably preclude their usage in e.g. BridgeStan since we wouldn't want to transitively pick up the rcpp dependencies