Description of the new feature/enhancement

"wordDelimiter" lists a couple of stop characters, such as ASCII quotation mark ", apostrophe ', hyphen/minus - and such.

However, special Unicode quotation marks like “, ”, apostrophes like ’ dashes –, —, box drawing characters, non-breaking spaces and so on an so forth remain word characters (selected on a double click) which is most likely not the best behavior, and adding such characters one by one to the set as the user encounters them is cumbersome.

Proposed technical implementation details

I think the default behavior should be based on Unicode character categories. On top of this there could be a way to add/remove certain characters to/from the set as exceptions. (Maybe even a way to add/remove entire character categories at once, although that might be an overkill.)