This only catches errors inside the statically-linked bytecode objects, not those triggered via loadfile (in import or the init script itself). But still, it's a bit of an improvement since it no longer prints the entire string passed as the chunk name...
This only catches errors inside the statically-linked bytecode objects, not those triggered via loadfile (in import or the init script itself). But still, it's a bit of an improvement since it no longer prints the entire string passed as the chunk name...