Implemented a better method to check whether a string should be encoded as Unicode...

Implemented a better method to check whether a string should be encoded as Unicode: We should check whether the string can be represented with the Latin-1 Codepage rather than with the user's "local" Codepage.