duplicates:

• SR-3052 [Foundation] Incorrect result of string methods while working with non latin symbols

Issue Description:

Hello

I've noticed that `stringByAddingPercentEncodingWithAllowedCharacters` returns truncated result if the original string contains non-latin UTF characters. The result is encoded properly, but the last few symbols are missing, seems like a problem of calculating the length of resulting string

I'm having problems with Polish and Russian strings like: ''Łódź", "Москва" (I've added some digits on the ends of strings just for padding to see them disappearing in the resulting strings)

This reproduces on ubuntu15.10 and latest Swift development snapshot 2016-03-24. I've created a small example project demonstrating the issue here: https://github.com/kipelovets/SwiftUrlEncode

I also checked IBM sandbox and the issue reproduces on all available versions (3.0 Mar 1, 3.0 Feb 25, 2.2 Jan 11, 2.2 Jan 6). See the sandbox here: http://swiftlang.ng.bluemix.net/#/repl/3b7cd84828b080f7007ffbbb70db850e144aa3ab5e703433a7357a34f13609c3