- change the various escaping / normalization functions to better deal with
non-ascii, non-latin languages, such as Russian.
It seems. now we can match 'Тесла' or 'Аркона' without problem.
- added unit test.
- WIP -- needs more testing.