Steps to reproduce

1. Send a message like "👩‍👩‍👧‍👧http://google.com" (that is, a family emoji (or really any complicated unicode codepoint) followed by a link)
2. The link should appear in a MessageEntity in update.message.entities with the type 'url'.
3. Try to use the length or offset attribute for basically any purpose... Fx.

entity = update.message.entities[0]
link = update.message.text[entity.offset:entity.offset + entity.length]

Expected behaviour

In the example above you'd expect to have "http://google.com" in the link variable.

Actual behaviour

link contains "://google.com".

Why

This happens because telegram servers calculate lengths in 'UTF-16' codepoints, while python clearly doesn't (and really that's good since utf-16 is bad...). This means that python sees the family emoji as 8 characters while telegram sees it as 12 characters...

Solution

Either:

1. Patch the length and offset inside MessageEntity's such that it matches what python thinks...
2. Add some sort of util function to convert to and from utf-8 and utf-16
3. Add some sort of custom slicing util function that does the conversion internally

Configuration

Operating System:
Windows 10 Education

Version of Python, python-telegram-bot & dependencies:
python-telegram-bot 5.0.0
urllib3 1.16
certifi 2016.08.08
future 0.15.2
Python 3.5.2 |Continuum Analytics, Inc.| (default, Jul 5 2016, 11:41:13) [MSC v.1900 64 bit (AMD64)]

Note
The behaviour of wide strings (weird unicode stuff) apparently changed as of python 3.3 (it now has a certain behaviour always where before it was a build-flag or something like that), so to support all python versions, we'd probably have to do some sort of wizardry... (See discussion at https://stackoverflow.com/questions/30775689/python-length-of-unicode-string-confusion for more info (kinda))