Help with TOC formatting?


I’m trying to figure out how to format my section headings in the compiled pdf as well as how to format the table of contents. What I’d like is something very much like the pdf of the scrivener manual, actually, where the table of contents might look something like this:

IV. Teaching
__ A. Class One
__ B. Class Two
____ 1. Lab for Class Two
____ 2. Lecture for Class Two

etc. only without the underscores (sorry, couldn’t figure out how to indent them in this post)

and then in the main document, I’d like the corresponding title to be:

I. Teaching

I.A. Class One

I’ve managed to get the very first number by going into formatting during compile, the choosing “Section Layout” and updating the Title Prefix to: <$n>. But I’m not sure of how to add the next numbers or access those values using markdown. I’ve also tried changing the subfields that have subfields from files into folders. That changed some things, but didn’t get the right result. In the TOC I’m getting things labeled like:

  1. Teaching
    __2. Class One
    __3. Class Two
    ____ 4. Lab for Class Two
    ____ 5. Lecture for Class Two

Any advice? I’ve looked through manuals and forums but seem to just be missing what I need. Any help or pointers in the right direction would be greatly appreciated! Thanks in advance!

If you’re using Markdown then there may be a better solution to all of this that doesn’t involve using Scrivener’s numbering tokens. The user manual, for example, uses MultiMarkdown as well, and I get all of those heading numbers and the ToC itself as a part of the LaTeX typesetting system. I don’t use any of Scrivener’s numbering routines, and LaTeX’s are generally better anyway. If you aren’t going to LaTeX there are still other and better options in my opinion—particularly if you’re going to ODT. That will be stylesheet based, meaning you can use the word processor’s stylesheet functions to handle numbering as well as ToC generation.

Scrivener’s numbering is pretty good, don’t get me wrong, but it does have a few blind spots, and using a separate numbering scheme between the heading display itself, and how it is referred to in cross-references (and the ToC, which is just based on the same) cannot be different from that. If you have “I.A. Class One” in the text, that is exactly what it will look like in the ToC. Another blind spot is that it doesn’t handle three+ depth numbering very well. There is no way to reference the grandparent’s (or greater) token value in an efficient manner.