Remove html comment tag before parsing

[eguitarz]

The commit cleans html comment tag () before parsing. For some pages like https://devcenter.heroku.com/articles/custom-domains, Readability will select content with html comments. It's not convenient to read an article when there's html tags inside that.

[cantino]

Thanks for your patch- I agree, removing comments is a good idea!

I'd recommend removing comments with Nokogiri instead of regular expressions. Could you try something like this?

@html.xpath('//comment()').each { |i| i.remove }

[eguitarz]

Thanks for pointing out. It works!

[eguitarz]

And it's been pushed on my fork. You can pull if you think it's all right.

[cantino]

Would you be up for adding a unit test? Then I'll definitely pull it.

[eguitarz]

Pushed the test case.

[cantino]

Thanks!