Remove any control characters in response

[LevitatingBusinessMan]

I think this should fix #43 and #7.

I couldn't find a good method to escape the control characters while retaining all other special utf-8 characters, but escaping is more prone to errors anyway as different programs interpret escape sequences differently.

The characters are removed from the lines just before being joined, encoded and send.

Currently 3 tests are failing, which seems to be caused by a mock variable not behaving the same as it's real counterpart. I am not that experienced with unittest.mock so some advice there is welcome.

I also still have to write an appropriate unit test.

[jodal]

Didn't look at the test failures as I'm on a phone.

[LevitatingBusinessMan]

@jodal I made the character mapping a constant at the top of network.py.

I am still not sure what the best way is to deal with those test errors.

[djmattyg007]

You'll need to rebase your branch on top of master. Also I took a look at the test failure last night and am genuinely bamboozled 😲

[jodal]

I think we need to replace sentinel.lines with an actual iterable, like a list, for the existing tests to be compatible with this change.

[jodal]

I updated the tests using sentinel.lines. Thanks for your contribution!