J-M

@lrasmus it affects both. I've disabled support for dynamic on a branch; if it does not work and we cannot figure out what is going on it just does not...

Upcoming version 1.7.0 should have a fix for it, as a related issue occured with a version of R that was not "numbers only" - I will check with unit...

Not a priority issue compared to e.g. making R.NET work with R 3.5.x., but check if this persists with newer versions.

@pfgherardini There can only be one R engine in a process, and R itself is largely not thread-safe.

This is great; thanks for all the work. I'll try to spend some time in the coming two weeks but am busy with work deliveries and moving house so will...

Hi, This is a bit of a hunch but I bumped into a similar issue when I was specifying an R path to R 3.4.4 but the R 3.5.x DLL...

Yikes, I went too far in responding fast to an issue. Backfired; live and learn. I see the kind of unit tests we need. If you can think of further...

I have a candidate fix for this currently on the testing branch https://github.com/jmp75/rdotnet/commit/4306481cfabd213a507a3134b17f5f04fee6d8c3. Probably not the best solution for parsing, but this passes some non-trivial corner case tests with '#'...

Working from ASP.NET is likely problematic: as I recall it assumes that applications are thread safe, and R is not. rnorm not found is also a symptom for some console...

Looking into it; Unicode character may indeed not be supported yet.