Andy Thomson

For a source of truth for vector names (util.c), we have the following: ``` TypeTable[] = { { "NULL", NILSXP }, /* real types */ { "symbol", SYMSXP }, {...

We are quite close to this list with Wrapper names and RType (should this be Rtype?)

Sounds reasonable...

I'm keen not to alias the std::String type, but Char or Str are options.

No, that works fine. An Rstr would wrap a single character object. I'm moving towards making Robj just a wrapper of a SEXP pointer without the enum. This will enable...

We also need to tighten up the mutablility of Robj types. I've been requiring `&mut self` for anything which can change an Robj. such as set_slot, set_attrib. This should be...

It would be worth having a "elt()", "region()" and "set_elt()" methods for each wrapper type. For ALTREP, these are quite efficient.

If you like, I can make a sample PR for "Integers" for discussion.

We do have some limited thread safety, but short of putting everything through a mutex, it will never be recommended to call R functions concurrently. The one place we do...

Best to talk to @yutannihilation or @Ilia-Kosenkov on why the methods are not documented. My guess is that the way the methods are defined in R doesn't lend itself to...