Quote:
Originally Posted by woodhead52
Thanks to both of you for your help on this. I really did not have to do the conversion. It was more for learning than practicality.
So, If I'm understanding what you both are saying, it's not possible to do a conversion and retain page number recognition.
|
I don't think you can with calibre, anyway. Maybe not with anything. If you have an apnx file with "real" page numbers (real as opposed to ball-park estimations), you can give that apnx file to the stand-alone version of KindleUnpack and it will output an epub that contains a pagemap.xml file. That epub can, in turn, be fed to kindlegen/kindle previewer which will produce a dual-purpose mobi/KF8 file that has the page information stored in a special section of its header file (that section gets stripped out on Amazon's end to create the apnx files for retail books) and can be used by devices/apps that support it. You could then use KindleUnpack to split that file into its mobi/azw3 pieces. I have no idea, though, if the mobi-only portion will retain that internal "actual" page info, or if it will just return to estimating the page info based on the byte-count algorithm that mobi always used before KF8 came to be. I suspect the latter.