We use the right default encoding according to the runtime and we detect the coding
comments at the top of a file and use the right encoding for it.
Checklist:
Did you find errors in the documentation? Do you have improvements to suggest? Suggest edits!