Quote:
Originally Posted by ZodWallop
I just dump .azw files into Calibre and let it convert to ePub with the built in converter. So far I have never spotted an error caused by the conversion.
What are the advantages of KindleUnpack? Is it just less prone to errors, or are there small things I might notice on every converted ePub?
|
If available, KindleUnpack will extract the epub source file stored inside the azw3 container. Calibre has become much better but it still likes modifying the CSS, removing inline styles, flattening the CSS, etc. There are relatively small differences but enough to sway me to use KindleUnpack. As a test, I just unpacked one azw3 and also converted it. The converted one had 27 added css entries where inline styles were converted to stylesheet entries. 24 of them were paragraph indents ranging from 12 to 23 pixels plus the occasional italic or bold. In the unpacked one, I simply looked for the style= with a bit of regex while the calibre converted one took 4 passes to get the same result.
Quote:
Originally Posted by ZodWallop
Same question for Modify ePub.
|
Much the same. When I edit the book, I tend to leave unused spans as <span> and then strip the empty spans using ModifyEpub. Basically, ModifyEpub does everything that Polish does other than adding soft hyphens and adds more functions that I find useful such as removing inline javascript and files.
Quote:
Originally Posted by ZodWallop
I'm curious as I found ePubs on my Nook almost universally have nicer layout than kepubs on my Kobo*. But I'm assuming that has to do with the kepub layout engine rather than errors in the converted ePub. But I'd be happy to be proven wrong.
(*I'm assuming ePubs on the Kobo would look as nice as ePubs on the Nook.)
|
I seldom use the kepub renderer on my Kobo since being a bit of a typography nerd, I prefer the font handling in RMSDK. I do make exceptions for ebooks that require kepub such as epub3 FLO books and the occasional book where there are enough high resolution images to make the image zoom worth having (I think I might need one hand to count those books).