@jensmaurer If you rebase this, I think this can land -- it has no visual effect, right?

@tkoeppe: Rebased. Confirmed no visual effect, except one strange false positive / spacing diff shown by diffpdf, and (of course) except for the additional index entries.

@tkoeppe, we should consider squashing this.

I extraced a nesting fix into b0a253c; please rebase.

Well, I can actually just fix this myself, so leave it with me.

Hm... That situation is actually fine in my copy. But apparently, I need to rebase.

@tkoeppe, rebased.

Hm, you lost the tcode around the closing paren. I'd reinstate \tcode{\keyword{int}).

Are we talking about the same thing? I have

a function of \tcode{(\keyword{int}}, pointer to pointer to \tcode{\keyword{char})} returning \keyword{int}

(These are parens around a parameter-type-list; the closing paren is at the end.)

The "function of" issue was a problem with your PR, which you have now fixed. The const_cast nesting issue was pre-existing, and your PR had fixed it, but I extraced it into a separate commit (since it has nothing to do with \keyword).

Yes, and I've rebased/force-pushed and fixed the "function of" issue. Are there any other issues?

No, that should be it.