[ldolse]

Preprocess doesn't work on mobi right now, but that should be coming soon. By itself it won't create a TOC anyway.

What you need is to look at the actual html of the mobi file using the debug output. In the conversion options specify a temp directory to send debug output to. Then run a conversion, this will populate that directory with a bunch of files, one of which will be the main file you need for figuring this out.

The file will be in the 'structure' folder, and it will be a large html file, the name may change based on the book - when you view this in a browser you'll see the raw text of your entire book. Open the file in a text editor, and find a chapter heading.

At this point it gets a bit more complicated, but your welcome to come back here for help. Once you've found the chapter heading you need to figure out what is unique about it, and then write an Xpath to use in the structure detection panel of the conversion options.

If Xpath is way over your head, just copy a few lines of text - a dozen or so above, below, and including the chapter heading, and paste that into a code block here. We should be able to help you with the xpath then.