Use logical page numbers (if available) for reference extraced pdf annotations about zotfile HOT 3 CLOSED

It would be great if you manage to take a look at it. I won't be able to do it anytime soon. The annotation extraction based on pdf.js was mainly implemented by devietti so that one of the pull requests or an email might be a good place to ask a question about extraction code (e.g. #8). I can help out with the main zotfile code.

from zotfile.

I contacted devietti. I'll post what he wrote here for reference:

Adding support for logical page numbers would require parsing out the
"page label dictionary" (if present) from the PDF. This data
structure is discussed in 8.3.1 of the PDF 1.7 spec
(http://www.adobe.com/devnet/pdf/pdf_reference.html). Some PDF
documents may have logical page numbers that aren't encoded in a page
label dictionary, and I have no idea how to get at those.

Assuming there's a page label dictionary present in the PDF document,
I'm not really sure how hard it would be to get at it - the existing
pdf.js code should be able to parse this object easily, because it
seems like a standard PDF object. I'd say it's possible and medium
difficulty, mainly due to finding out how to extract the page label
dictionary and interpret it.

from zotfile.

Closing this as duplicate of #190

from zotfile.