[Plugin] KindleImport Sigil plugin
 

Import Amazon Kindlebooks into Sigil

** NOTE: this plugin periodically checks for updated versions by connecting to github (where the source is maintained). Read the full post for more details **

If you're using Sigil v0.9.0 or later on OSX or Windows, all dependencies should already be met so long as you're using the bundled Python interpreter (default).

Linux users will have to make sure that the Tk or PyQt5 graphical python module is present if it's not already. On Debian-based flavors this can be done with "sudo apt-get install python-tk (or python-pyqt5)" for python 2.7 or "sudo apt-get install python3-tk (or python3-pyqt5)" for Python 3.4. On Arch distributions it can be done with pacman -S tk and/or pacman -S python-pyqt5).

*Note: Do not rename any Sigil plugin zip files before attempting to install them *

This plugin will work with either Python 3.4+ or Python 2.7+ (defaults to 3.x if both are present).
The absolute minimum version of Sigil required is v0.8.3 (Python must be installed separately prior to v0.9.0)


In addition to being able to unpack/open the KF8 sections of files, it also takes a pretty serious stab at opening the older style mobi-only files (thanks to KevinH for contributions/suggestions on that front). Also, if the original source package is a part of the file (think kindlegen output) and the original source is an epub, then the plugin will open that instead of using the unpacked version.

* Please note that there's some atrociously bad markup in many older mobi files (even retail ones), so while the plugin tries its best to whip that markup into some semblance of validity, it's still in your best interest to allow Sigil to correct any issues it may detect upon first opening. *

Configurable preferences (available after first run in the plugin's corresponding json prefs file) are:

• azw3_epub_version : a string value that determines whether KF8 books will be imported as epub version 2.0 or epub version 3.0. Defaults to "2" but can be changed manually to "3". Changing this value to "3" will probably cause you nothing but misery before Sigil v0.9.3. NOTE: mobi7 formatted books will be unaffected by this preference. They will always be opened as v2.0 epubs. Use the ePub3-itizer plugin if you want those older mobis as ePub3s.
  
• use_hd_images : a true or false (boolean) value (defaults to true). If HD versions of images are present in KF8 books, use those in the resulting epub instead of their lo-rez counterparts.
  
• use_src_from_dual_mobi : a true or false (boolean) value (defaults to true). If the source package is included in the dual-format kindlegen output (and the source is an epub), open that source epub instead of unpacking the KF8 binary. Use false to always unpack.

• asin_for_kindlegen_plugin: a true or false (boolean) value (defaults to false). Create a dc:identifier tag from the KindleUnpack metadata that the kindlegen Sigil plugin can then use to force the KF8 portion of a kindlegen-produced, dual-format mobi to have an ASIN.
  
• preserve_kindleunpack_meta: a true or false (boolean) value (defaults to false). KindleUnpack (the core of the KindleImport Plugin) creates a lot of reference metadata that is commented out in the OPF file it creates. Starting with Sigil 0.8.9XX, these comments will not survive when opened with Sigil. Setting this value to "true" will cause this section to be uncommented, so that the metadata tags exist in the OPF file's metadata section in Sigil. It will be up to the users to keep/delete the values they want/don't want.

For those who don't know where to find the plugin's json preferences file, the easiest way is to open the Sigil Preferences location from within Sigil. This can be achieved by 'Edit->Preferences' (or F5) and looking for the "Open Preferences Location' button in the bottom-left portion of the Sigil Preferences dialog. After clicking that button, you open the 'plugins_prefs' folder, and then the 'KindleImport' folder. There you should find a file named 'KindleImport.json', which is the one you want. If that file is missing, you probably haven't used the plugin yet. It needs to be run once before the file will be created.

KindleImport's code is hosted/maintained on Github.

This plugin's update checker will only try to check for updates after at least 12 hours has elapsed since its last check. The plugin will only notify you about updates once for each new version released.

Changes

Feature request ;):
Epub file not new.epub but BookTitle.epub?

That's out of the Input plugin developer's hands.

This is for testing and only for DD :).

Not really understanding that, I will try to persist :).

Hi varlog,

I do think that it is controllable by the plugin. The problem is that you would not want to use the book title as the file name without cleaning up all of the possibly nasty characters, control chars, forward and backward slashes, any character with special meaning in the shell, spaces at the end, etc that can come from a book title used to create a file name. So using "new.epub" is safer and since the file itself is passed into Sigil via temp directories/scratchfiles, the user will need to rename it and save it someplace anyway ... which makes "new.epub" as good a temporary file name as any.

KevinH

It's a good thing you decided to persist ... because I clearly didn't know what I was talking about. :o

Should be doable. :)

YES:D!

I've come across an error trying to import a Mobi eBook from Amazon. I am using Windows 7 32-bit and Python 32-bit (version 2.7.6) from Python.org with the 32-bit Sigil.

Simpleton as I am: if the book has some KindleUnpack recognizable metadata the title must be there. Converting it to sane string should be possible?
The overkill would be to include the name of the original file :).

Hey Jon,

By any chance, does that same book fail when trying to unpack using KindleUnpack (plugin or stand-alone)? If so, we may want to take this to the KindleUnpack thread. That's what the KindleImport plugin uses.

You do realize that you are talking about naming a temporary file that will be needed to have save-as done on it once it makes it to Sigil, right?

Yes, running it through a file cleaning routine is indeed possible and KindleUnpack has one I think. Giving it the same basename as the original file might in fact be easier to do since no file name cleaning routine would be needed at all since a valid base name has to exist.

Hi JSWolf,
please run DumpMobiHeader_v018.py on that mobi and copy what values it has for cover offset and thumbnail offset in the EXTH metadata for the old Mobi header. I am betting that the ThumbOffset is 0xFFFFFFFF which means it's missing. If so I can give you a new mobi_opf.py to fix that issue. DumpMobiHeader is an extra Python script that I have added to the KindleUnpack zip archives on occasion. Any recent version would work.

If not, then please post a test case to the KindleUnpack forum and I will try to track it down.

KevinH

yes, it fails with the KindleUnpack 0.75 plugin.

Yes, you are correct about ThumbOffset being 0xffffffff. I'm ready to run your test version of mobi_opf.py when you are ready to get it to me.

Here is the output from DumpMobiHeader from 0.75.