Extract table of contents from mobi file

[oecherprinte]

Hi,

are you aware of any way to extract a tabe of contents from a mobil file? I would need the following information:

- chapter/section name
- location of chapter/section

Here is what I would love to do:

I would like to process the "My Clippings" file for kindle and regroup the annotations to a book such that they are grouped under the chapters. Since I know the positions of the annotations I could group them according to chapters when I know the chapter positions and titles.

I would be fine with any programming language (python, ...).

Thanks,

Jens

P.S.: Of course it would be great if I could also extract the toc for DRMed kindle books.

[susan_cassidy]

TOCs don't work by referencing locations. They are HTML-based, and reference a tag on the chapter heading. If you run mobi2html (Perl program from MobiPerl), and look at the HTML output, there is usually just a chunk that is the list of hyperlinks forming the TOC.

[HarryT]

Originally Posted by susan_cassidy

TOCs don't work by referencing locations. They are HTML-based, and reference a tag on the chapter heading. If you run mobi2html (Perl program from MobiPerl), and look at the HTML output, there is usually just a chunk that is the list of hyperlinks forming the TOC.

You can find out where it is by looking at the "Guide" section of the Mobi's OPF file. The table of contents is referenced by the "toc" guide item.

[oecherprinte]

Thanks for the answer. But excuse my ignorance:

What is the opf file? Is that clear text xml?

Thanks,

Jens

[HarryT]

The OPF file is the file that contains all the "instructions" on how to make a Mobipocket book. It contains all the metadata, the list of files that make up the book, and what order they should be arranged in, the "Guide" section (which contains links to the cover image, the table of contents, and the point at which the book should be initially opened) and so on.

You can unpack a Mobi file to its OEB source (which will include the OPF file) using the "ebook-convert" tool that's a part of Calibre.

[oecherprinte]

That's cool. Thank you very much. So I could write a script to extract the toc and use that later on.

[bwdrennan]

'You can unpack a Mobi file to its OEB source...using the "ebook-convert" tool that's a part of Calibre.'

Hi, When I performed this on a .mobi file, the resulting styles.css file contained Calibre's styles. HTML files were laced with Calibre's styles. File names for images were altered from the original. Is there a way to unpack to the original files, bit for bit? Or, is all that original information left aside as the conversion software packs the file into a .mobi archive? Thanks!

[DiapDealer]

Is there a way to unpack to the original files, bit for bit? Or, is all that original information left aside as the conversion software packs the file into a .mobi archive? Thanks!

MOBI isn't really an "archive." It's basically a binary database that's compiled from source files. Unless the source is still included with the package (kindlegen produces files which include the source files), it's simply impossible to rebuild/recreate the exact source files "bit for bit." Mobi_Unpack will allow you to extract the markup code that's contained within a mobi file with very little change, but it still won't represent the original source. Using the debug option, you can actually dump the raw contents of the mobi file, if you so choose.

Mobi_unPack.py