Create table of contents with CSS tags

[spartanhooah]

I found the nicely-documented Auto-create a table of contents function, but that requires the chapter headings be simple <h/> tags. I've got a book where the chapter headings look like this:

<p class="calibre4">
    <span class="calibre3">
        <strong class="calibre5">
            <span lang="EN-US">Chapter 2</span>
        </strong>
        <span lang="EN-US"></span>
    </span>
</p>

Is there a way to adapt the example function to work in this instance?

[theducks]

I simply REGEX that kind of mess into
<h3 class="chapno">\1</h3>

Where .chapno (CSS) contains the consolidated styling of those 3 cascaded classes

[spartanhooah]

Oohhh, that's a good idea.

[theducks]

Visually, it looks the same, but now the Auto-tools work

BTW I use H3 for chapters, H2 for Parts (when needed), H1 for Titles in Anthologies
Since you are using stylesheets, the H level is stylistically meaningless

[spartanhooah]

I just realized another possible snag: this e-book is split across four HTML files. Is there any reason I shouldn't combine them all before running the table of contents function?

[theducks]

Originally Posted by spartanhooah

I just realized another possible snag: this e-book is split across four HTML files. Is there any reason I shouldn't combine them all before running the table of contents function?

Nope.
The beauty of using an e-book editor is that it is aware of the structure.
All that is needed is the files are in the proper order within the book (file) browser

[DNSB]

Originally Posted by spartanhooah

I just realized another possible snag: this e-book is split across four HTML files. Is there any reason I shouldn't combine them all before running the table of contents function?

Personally, I would suggest merging the files and then splitting the one file into multiple files at chapter boundaries.

[theducks]

I believe these are already split properly as the OP said Line 14 of each, so this is not needed in THIS case

[DNSB]

Is that a reference to estevam's posts in the Sigil regex thread? That's the only line 14 reference that comes to mind in the recent past. The OP in this thread has 3 posts all in this thread.

[hobnail]

Another way to accomplish what theducks said about a regular expression is to use the plugin Diaps editing toolbag. With it you can delete the outer p, span, and strong, and then convert the span around the chapter title to h3. When cleaning up books from Gutenberg I nuke all of their classes and whatnot, merge the files like DNSB said, copy everything between the body tags and paste it into a new file, then in that split on the h3 or whatever tags.

[spartanhooah]

Alright, I'll go the merge route. Any gotchas to look out for that may not be obvious?

[DNSB]

Originally Posted by spartanhooah

Alright, I'll go the merge route. Any gotchas to look out for that may not be obvious?

I'd make sure that you have no significant errors in the files before merging. If you don't have the epubcheck plugin, I'd install and use it as well the the builtin checker. Some errors won't affect the merge process but I've had a couple of times where merging files with errors in those files produced total garbage as the output file.