natsort with ns.LOCALE error: ValueError: character U+110000 is not in range [U+0000; U+10ffff]

[SethMMorton]

This isn't really a bug in natsort, but rather with the UTF-8 implementation on BSD systems (like OSX). Below are some references:

• http://stackoverflow.com/questions/33459384/unicode-character-not-in-range-when-calling-locale-strxfrm: Explanation of issue
• http://bugs.python.org/issue23195: A related python bug report
• How to get accented chars to group with their unaccented versions? #21 : natsort issue describing how to install the fix
• http://natsort.readthedocs.io/en/master/locale_issues.html#the-locale-module-from-the-stdlib-has-issues: natsort documentation explaining the issue

I am creating this issue and leaving it open so that users that see this problem know that they need to use PyICU to solve this problem if they see it, and have easy access to all the relevant documentation.