See http://stackoverflow.com/a/22755675/96656#comment34706729_22755675 by Jukka K. Korpela:
I’d expect the number 110,122 mentioned in the standard. It seems that your code includes not only ISO control codes (which might be called “control characters”, rather debatably) but also Private Use codepoints (to which the standard specifically does not assign any characters) and surrogates (which are definitely not Unicode characters at all).
Ref. mathiasbynens/unicode-data#4.