Fix emphasis escape sequence detection

[jcheatham]

@robin850 - Fixes issues #214 and #369

[robin850]

Hey @jcheatham,

Thanks a lot for this patch! ❤️ Could you please add a changelog entry ? Matt or I will merge it then, but I will first have a look on my computer tomorrow to make sure the conformance suite passes on it (c.f. #378 (comment)).

[jcheatham]

Alright, changelog updated. Wasn't sure if I should be poking @mattr- for these too. So, this one doesn't include the backslash-hell fix, just the nokogiri test fix in order to get travis to pass (so hopefully this will pass the conformance suite on your computer with no problem ;)

[robin850]

s/while/with here, no ?

[jcheatham]

I looked at it as "during the state of emphasizing", so chose "while"

[robin850]

Your English skill is certainly better than mine anyway, let's keep while.

[robin850]

Thanks again @jcheatham! The conformance suite is green on my computer. However, #369 doesn't seem to be fixed with this patch, I'm unable to get this test green:

def test_char_escaping_when_highlighting
  markdown = "==attribute\==="
  output = render_with({highlight: true}, markdown)

  html_equal "<p><mark>attribute=</mark></p>\n", output
end

Would you mind having a look please ? :-)

[jcheatham]

Hi @robin850, looks like the problem with that case was that = wasn't included in the list of escape characters. I added it and the test (assuming the test above meant "==attribute===" :) The conformance tests failed for me at first, maybe due to some debugging statements I had in place, but I haven't had any failures since... will see what travis has to say about it.

[robin850]

@jcheatham : Yes, my test was wrong anyway. Awesome work! Merging, thanks again! ❤️ 💙 💛 💜 💚