\tableofcontents undefined in quarto book (PDF)

[friendly]

Description

My book, Visualizing Multivariate Data and Models in R is nearly done, but I've run into a serious, and mysterious problem in producing the PDF version. That is, the XeLaTeX compilation terminates with the error Undefined control sequence: \tableofcontents with this text in the console log:

[11])
\tf@toc=\write6
\openout6 = `index.toc'.


! Undefined control sequence.
<argument> ...@edef \Hy@testname {\Hy@tocdestname 
                                                  }\ifx \Hy@testname \@empty...
l.531 \tableofcontents
                      
The control sequence at the end of the top line
of your error message was never \def'ed. If you have
misspelled it (e.g., `\hobx'), type `I' and the correct
spelling (e.g., `I\hbox'). Otherwise just continue,
and I'll forget about whatever was undefined.

\tableofcontents IS defined in the krantz.cls class file (lines 1017-1034), which is what my publisher wants. I have no idea where this ! Undefined control sequence comes from.

Curiously, the output file, index.pdf from this pass does contain the table of contents, but nothing further is produced in this attempt to build via Render book.

To track this down, I open the resulting index.tex file, generated by Quarto in TeXStudio,
In this file, I see the following, generated by Quarto:

\renewcommand*\contentsname{Table of contents}
{
\hypersetup{linkcolor=}
\setcounter{tocdepth}{2}
\tableofcontents
}

TeXStudio flags \tableofcontents as an error, but I've set this up to use nonstopmode, which causes it to ignore this. I have to do another pass in TexStudio to build the book index (Tools -> Index), but then I get a usable. complete PDF.

Can someone here help me solve this mystery and relieve me of some misery?

I'm not suggesting that this necessarily a Quarto problem, but there is nothing I can see in my files that would cause
\tableofcontents to become undefined. If anyone is brave enough, you can look at the index.tex that I'm working with.

[friendly]

To add more context for this problem: In the RStudio console log, I see this output from the first two XeLaTeX passes to compile and resolve cross references -- it produces index.pdf of the entire book (485 pages), except for the final book Index:

  Output written on index.pdf (485 pages).
  Transcript written on index.log.
  
  Re-compiling document for index
  This is XeTeX, Version 3.141592653-2.6-0.999997 (TeX Live 2025) (preloaded format=xelatex)
   restricted \write18 enabled.
  entering extended mode
  (./index.tex

But then it runs makeindex to produce the index .idx file:

  (c:/Users/friendly/AppData/Roaming/TinyTeX/texmf-dist/tex/latex/caption/ltcaption.sty)) (c:/Users/friendly/AppData/Roaming/TinyTeX/texmf-dist/tex/latex/url/url.sty) (c:/Users/friendly/AppData/Roaming/TinyTeX/texmf-dist/tex/latex/tabularray/tabularray.sty (c:/Users/friendly/AppData/Roaming/TinyTeX/texmf-dist/tex/latex/ninecolors/ninecolors.sty)) (c:/Users/friendly/AppData/Roaming/TinyTeX/texmf-dist/tex/latex/float/float.sty) (c:/Users/friendly/AppData/Roaming/TinyTeX/texmf-dist/tex/latex/graphics/rotating.sty (c:/Users/friendly/AppData/Roaming/TinyTeX/texmf-dist/tex/latex/base/ifthen.sty)) (c:/Users/friendly/AppData/Roaming/TinyTeX/texmf-dist/tex/generic/ulem/ulem.sty) (c:/Users/friendly/AppData/Roaming/TinyTeX/texmf-dist/tex/latex/siunitx/siunitx.sty) (c:/Users/friendly/AppData/Roaming/TinyTeX/texmf-dist/tex/latex/base/makeidx.sty)
  Writing index file index.idx

This is when the error occurs, and the end of the console log tells me that only 7 pages were produced.

  l.524 \tableofcontents
                        
  Output written on index.pdf (7 pages).
  Transcript written on index.log.

[cscheid]

Without digging fully into the issue, there are a couple of things I notice in krantz.cls: lines 1015 and 1016 appear to be conditionals. Have you confirmed that these conditionals are satisfied?

\newif\if@break
\newif\if@firstauthor
\newcommand\tableofcontents{%
    \if@twocolumn
      \@restonecoltrue\onecolumn
    \else
      \@restonecolfalse
    \fi
    \chapter*{\contentsname
% ...

[friendly]

AFAICS, those conditionals only determine if the index is set in one column or two. But \tableofcontents IS defined there.

From what I observed above, it somehow becomes undefined after the first XeLaTeX pass, which completes w/o problems.

[cscheid]

The only thing I can think of is to copy the .tex file into a separate directory and start reducing the test case (deleting lines from the file) until you find the specific culprit...

[mcanouil]

From #13497, I can confidently say that the issue is from krantz which likely conflicts with something from the template used by Quarto/Pandoc.

1. quarto create project book.
2. add krantz.cls.
3. quarto render.
4. Get the error.

Now, you need to look at the tex file as suggested by Carlos, to strip further down what from krantz and partials/templates (https://github.com/quarto-dev/quarto-cli/tree/main/src/resources/formats/pdf/pandoc) is the source of the conflict

[mcanouil]

@friendly Did you modify Krantz.cls?

For instance, the following used as a Quarto CRC book example works: https://github.com/bgreenwell/quarto-crc/blob/main/krantz.cls
Yours is very different from the ones I can find online (including https://github.com/yihui/bookdown-crc/blob/master/krantz.cls)

[friendly]

Thanks so much for follow up on this. My version of krantz.cls has exactly the same header as the one on https://github.com/bgreenwell/quarto-crc/blob/main/krantz.cls, so I assumed they were identical. I'll download these and do a diff to see where/if they differ.

%% This is file `Krantz.cls'
%%% Created by Shashi Kumar / ITC [August 2008]


\NeedsTeXFormat{LaTeX2e}[1995/12/01]
\ProvidesClass{krantz}
              [2005/09/16 v1.4f
 Standard LaTeX document class]

[friendly]

SUCCESS!
I downloaded both of your suggested copies -- they're identical, except that one is missing a newline at the end.
But they are VERY DIFFERENT than the "copy" I've been using all this time, in spite of having the same date, vers. number in the header.

It also solves the workflow issue I had in #13497

Thank you so very much for taking the time to track this down.