View Single Post
Old 09-22-2013, 01:50 PM   #98
DNSB
Bibliophagist
DNSB ought to be getting tired of karma fortunes by now.DNSB ought to be getting tired of karma fortunes by now.DNSB ought to be getting tired of karma fortunes by now.DNSB ought to be getting tired of karma fortunes by now.DNSB ought to be getting tired of karma fortunes by now.DNSB ought to be getting tired of karma fortunes by now.DNSB ought to be getting tired of karma fortunes by now.DNSB ought to be getting tired of karma fortunes by now.DNSB ought to be getting tired of karma fortunes by now.DNSB ought to be getting tired of karma fortunes by now.DNSB ought to be getting tired of karma fortunes by now.
 
DNSB's Avatar
 
Posts: 47,084
Karma: 169815798
Join Date: Jul 2010
Location: Vancouver
Device: Kobo Sage, Libra Colour, Lenovo M8 FHD, Paperwhite 4, Tolino epos
Quote:
Originally Posted by hwlester View Post
Yeah, this is one thing that annoys me about the Kobo app: it doesn't use absolute page numbers like my Sony reader does. I seem to remember that even sideloading an EPUB onto the Kobo app did the same thing, but I may be misremembering. Nice to hear that the Kobo devices do it correctly.
The main issue with the Adobe page numbering algorithm is that it only works with identical ebooks. While it is independent of the display page size, it is sensitive to the compression and file structure of the epub file.

As I understand it, the algorithm looks at the compressed size of the file and counts 1 page per 1024 bytes of compressed data and always rounding up -- if you redo an epub file using a different compression algorithm, the page numbering can change. A 114 byte file or a 1024 byte file would show as 1 page while a 1025 byte file would show as 2 pages. You add up all those page counts and get the total number of pages in the ebook. Once you open a file, the pages within that file are evenly distributed over the uncompressed data size. A file that is 1025 bytes will be broken into two pages that are whose character count will depend on the uncompressed data size while a 2047 byte file would be broken into two pages which would be very likely to be much longer than those from the 1025 byte file.

There are other issues such as an attempt to compensate for the differences between a DRMed and non-DRMed file which may make page numbering differ between two otherwise identical files.

Adobe did offer a non-standard extension called page map which never seemed to get any traction. It worked by attaching page numbers to HTML anchor tags.

Regards,
David
DNSB is offline   Reply With Quote