The _checkByline function is a little weird in a few ways:

• Despite its name and return type it does more than check
• It relies on external information to short-curcuit itself
• It calls another function to do more checks
• It checks whether the node is an element (which will always be true) before getting attributes, but unconditionally relies on variables it has set conditionally

It gets the job done, but I found it confusing to read and make sense of considering its modest length. I have consequently cleaned it up and re-organized the logic surrounding byline harvesting in general.

I included some basic documentation, but since I've never written a JavaScript doc-string before and I'm uncertain what the matchString argument is actually for, suggestions for improvement there are welcome.

For reference, one can assume that node is an element becaause the _grabArticle loop begins at documentElement, and the _getNextNode function which drives the loop always returns an element if it returns a node.