I have been parsing some PDF files (using the AutoDetectParser) and noticed that in the returned metadata map there was an XMP field called xmpTPg:NPages which contains the number of pages. However, the Metadata.PAGE_COUNT property, where I would expect this sort of information, was null.
I did a bit of googling regarding the xmpTPg:NPages property and stumbled across the PagedText.N_PAGES constant within the org.apache.tika.metadata package which seems to serve no other purpose but to map to that particular XMP property. To my further confusion, the PagedText class was not even mentioned in the API docs. Could someone clear this up for me? :) Thank you!
