View Single Post
Old 01-09-2020, 09:30 PM   #11
j.p.s
Grand Sorcerer
j.p.s ought to be getting tired of karma fortunes by now.j.p.s ought to be getting tired of karma fortunes by now.j.p.s ought to be getting tired of karma fortunes by now.j.p.s ought to be getting tired of karma fortunes by now.j.p.s ought to be getting tired of karma fortunes by now.j.p.s ought to be getting tired of karma fortunes by now.j.p.s ought to be getting tired of karma fortunes by now.j.p.s ought to be getting tired of karma fortunes by now.j.p.s ought to be getting tired of karma fortunes by now.j.p.s ought to be getting tired of karma fortunes by now.j.p.s ought to be getting tired of karma fortunes by now.
 
Posts: 5,808
Karma: 103362673
Join Date: Apr 2011
Device: pb360
Quote:
Originally Posted by hobnail View Post
Kindlegen creates a dual mobi, which does contain the book in azw3 format but it also contains the book in the old mobipocket format as well as containing the original epub so it's a bit bloated. If you give it the -dont_append_source flag it will exclude the original epub, but it still contains the unnecessary (in the OP's case) mobipocket format book.

What about simply using Calibre to convert the epub to azw3? That has the advantage of being able to tweak the output azw3 without needing to fiddle with the CSS in the epub.
I never tweak the formatting, but if I had to change the CSS, I would mount the epub as a disk drive and use the text editor of my choice to change the CSS, then unmount the epub. No need to fire up calibre.

As to splitting out the azw3, I regularly start with an azw3, use kindleunpack to obtain an epub, extract the paging information directly from the HTML, use that to generate a correct page-map.xml, put that in the epub, use kindlegen, then use kindleunpack again to split out the azw3 and apnx. Some may find that onerous, I do not.
j.p.s is online now   Reply With Quote