The detection of a content- and childless tag does not account for namespaces

[mr-stephan]

When using the html formatter for xml I found that if the elements have a namespace prefix, the element after a self-closing-tag is in the same line as the self-closing-tag.

For example:

<ns1:Apple><ns1:Banana><ns1:CitrusFruit/></ns1:Banana></ns1:Apple>

is formatted to something like:

<ns1:Apple>
  <ns1:Banana>
    <ns1:CitrusFruit/></ns1:Banana>
  </ns1:Apple>

There is a simple fix:
Replace token_text.match(/\w+/) with token_text.match(/[\w:]+/) and
replace .match(/<\s*(\w+)/) with .match(/<\s*([\w:]+)/)

The example is contrived and the output may not be absolutely correct, but the point is, that the pattern does not account for a namespace prefix.

---

If you give me enough time, I will send a pull request