Uncracking .mobi for fine-tuning

[twedigteam]

This one has been beaten to death, I'm sure...so apologize if this is reposting...

What's the best way to tweak/fine-tune a .mobi file after conversion from epub? I'm just starting out in the Kindle world (1/2 year experience w/ ePubs) and see a couple ways this is possible (I realize it is not technically possible as with ePub):

A. using scripts or user-created unpacking tools to convert back to XHTML, then edit, then repack/rezip with KindleGen

B. Converting back to ePub and editing in something like Sigil

C. Converting back to ePub, then editing XHTML within a text editor (avoiding any markup auto-correct/clean-up issues with Sigil), followed by re-converting through Calibre or KindleGen

I foresee small tweaks after conversion (only tried via Calibre so far) e.g. a header with too much bottom margin...but seems quite time/labour intensive to run tools or scripts in order to get there! Anyone else overwhelmed by the difficulty in doing so? If yes, is there any advice or expertise in this realm?

[ATDrake]

Originally Posted by twedigteam

What's the best way to tweak/fine-tune a .mobi file after conversion from epub?

If it was an ePub to begin with, start with your original ePub, make any modifications you want, and then convert to Mobi. That's what I do with all of mine (and I make a point of picking up the ePub versions just so that I can).

Trust me, it's by far easier to go back and tweak your modified ePub to output a new Mobi than it is to make any alterations beyond metadata to an existing Mobi, and far less frustrating as well. And it also saves time, not having to go through any intermediate re-conversion steps.

My personal workflow is:

1. unzip ePub into working folder,
2. text editor to tweak OPF and NCX to produce chapter markers and proper table of contents linkage, and if necessary modify css & tags for the sort of display I like (I'm lazy and not-too-picky about this, so as long as there are proper scenebreaks and no-indent on the first paragraph of a section and the chapters start on new pages, I'm reasonably content)
3. run it through KindleGen and kindlestrip.py to get rid of the bloat,
4. check new Mobi in KindlePreviewer to see if anything's gone horribly wrong with the XHTML+CSS to MobiHTML "translation" (happens fairly often for incredibly trivial reasons),
5. Mobi2Mobi to fix the inevitable metadata errors that KindleGen produces,
6. read on Kindle and highlight noticeable typos/formatting errors, etc.,
7. text editor to fix errors in a copy of the previously-modified unzipped ePub files (just in case I manage to screw things up in a way that involves restarting from scratch),
8. rinse and repeat steps 1-7 if necessary.

Hope this helps.

[twedigteam]

This is absolutely helpful. It's taking a little while to get my head wrapped around this, but in a way, it's a fun challenge. I've become so accustomed to the convenience of ePub (+ ePub editors) to correct issues...and this isn't much different aside from a little more forethought.

The .opf/.ncx dilemma seems complex, but for the moment, I am happy producing a relatively well-formatted .mobi (like you mentioned, as long as breaks/indents are proper, then good to go!).

Greatly appreciate the assistance.