Find words using a word regex, rather than splitting on what is not

[foolip]

This helped when finding typos in HTML, like tihs. An
alternative fix is of course to include <> in the split regex.

[lyda]

I really need to run through this to address UTF8 issues in general. There are words that could be in the misspell list but are not due to UTF8 issues. I filed #16 to remind myself to dig into that some weekend.

In the meantime, I've submitted 475fe97 which should help with HTML, XML and things like it (apache config files spring to mind). Beyond your example, every single word next to a tag boundary was going unchecked which is annoying.

[foolip]

Thanks for looking into this!

[lyda]

No prob. I do want to learn more about utf8 in python so I will get to this. I hope the change I made is sufficient for the delimiters. I've released a new version that has this change so pip install --upgrade misspellings should grab it. Thanks for taking the time to give feedback!