Cannot encode/decode ?private? Unicode characters

IPCC reports have characters (rows on dots) in Table of Contents that seem to be outside normal Unicode ranges (maybe private).

Example - dots in p4 of test/resources/ipcc/wg2/spm/fulltext.pdf

Have currently trapped this at a page level so pages with these characters are not written but don't break the flow.
The Exception mess is simply IO_ENCODER