At the moment cyrillic strings are ignored.
It cannot now, as I see, because regexp in apply_string takes only Latin letters, digits and underscore. I'm know very little of Python. Is it difficult to make it process utf strings?
#20 draft fixes
Please try it, then write how it works.
Inc-redible! Thank you, it works fine.