I believe automatically inferring lifetimes in the general case is equivalent to solving the halting problem (for every sub-program, determine if a given refer

I believe automatically inferring lifetimes in the general case is equivalent to solving the halting problem (for every sub-program, determine if a given reference is stored to the heap before the sub-program terminates), so transipiling JS to Rust would presumably need manual lifetime annotation or else require significant limitations to simplify lifetime analysis.