Passing character vectors from R native memory to .NET arrays of strings seems to cause memory leaks. Memory footprint can be tricky thing (working set vs total commit size is confusing me), but this does not look like a false positive.
See a basis for a unit test[there](https://github.com/jmp75/rdotnet-support/blob/master/usersupport/deldir/Program.cs)
Comments: Note in passing: the call to the indexer property calls protect/unprotect very often and at rapid cadence on the charactervector (this). This is not necessary. Can it be factored out. In fact, why would the character vecgor in R be modified or garbage collected when reading? writing to the vector at an index is another matter. ``` C# public IEnumerator<T> GetEnumerator() { for (int index = 0; index < Length; index++) { yield return this[index]; } } ```