[Feature Request] A Way to retrieve non-standard characters.

[TNTErick]

...
</w:rPr>
  <w:sym w:font="STEDT" w:char="F0DC"/>
</w:r>

As the code above encodes a non-standard character of '\uF0DC' which is disregarded at extraction. This breaks my document as there are many characters encoded in the Private Use Area.