Why does pandoc not generate a table of content in a LXC container?

Question

I have a problem with pandoc not generating a table of content correclty.

Here is my workflow:

1. I create a Markdown file with the content of a web page.
2. I version my project with Git.
3. I git push to my self-hosted GitLab server.
4. Using CI/CD, my server runs 3 scripts in Ubuntu/bionic LXC containers.
   1. The first one extracts metadata.
   2. The second one calls pandoc to generate an HTML page with header, footer, table of contents and CSS.
   3. The third one pushes the new HTML page to my production server.

Everything works perfectly except a small detail: when pandoc generates the HTML, the table of contents is missing. What's wierd is that it works when I run it on my local machine (Fedora 34).

Here is the script I am running in the second container (which has nothing installed beforehand):

apt-get update
apt-get install -y python3 pandoc pandoc-citeproc
mkdir -p build
python3 .deploy/convert.py

You can find the result of the script down below.

NOTE: in an LXC container, the user is root, no need for sudo.

Content of .deploy/convert.py:

import sys
import os
import re
from pathlib import Path

if __name__ == '__main__':
    with open('data/type', 'r') as f:
        type = f.readline()
    with open('data/number', 'r') as f:
        number = f.readline()
    with open('data/name', 'r') as f:
        name = f.readline()

    result = list(Path(".").rglob("*.md"))
    for file in result:
        parts = str(file).split('.')
        parts.pop()
        filename = '.'.join(parts)
        if filename != 'README':
            parts = str(filename).split('/')
            parts.pop()
            directory = '/'.join(parts)
            os.system('mkdir -p build/' + directory)

            # This is the important line!
            os.system('pandoc --toc --metadata title=' + type + '-' + number +
                      ' --metadata type=' + type +
                      ' --metadata number=' + number +
                      ' --metadata name="' + name +
                      '" -s --template .deploy/template.html ' +
                      str(file) +
                      ' -o build/' + filename + '.html'
            )

            with open('build/' + filename + '.html', 'r') as f:
                filedata = f.read()
            filedata = re.sub('href="(.*)\.md"', 'href="\\1.html"', filedata)
            with open('build/' + filename + '.html', 'w') as f:
                f.write(filedata)

The 3 containers are called sequentially, leaving what is called artifacts in between: they are files that can be passed to the next container. I checked the artifacts before the convert container and after it, the only missing part is the table of content.

Did I miss something?

Here is the result I get from the script. Again, even if you see $ instead of #, know that I am root!

$ apt-get update
Hit:1 http://security.ubuntu.com/ubuntu bionic-security InRelease
Hit:2 http://archive.ubuntu.com/ubuntu bionic InRelease
Hit:3 http://archive.ubuntu.com/ubuntu bionic-updates InRelease
Hit:4 http://archive.ubuntu.com/ubuntu bionic-backports InRelease
Hit:5 https://packagecloud.io/github/git-lfs/ubuntu bionic InRelease
Reading package lists...
$ apt-get install -y python3 pandoc
Reading package lists...
Building dependency tree...
Reading state information...
python3 is already the newest version (3.6.7-1~18.04).
python3 set to manually installed.
The following package was automatically installed and is no longer required:
  libfreetype6
Use 'apt autoremove' to remove it.
The following additional packages will be installed:
  liblua5.1-0 libluajit-5.1-2 libluajit-5.1-common pandoc-data
Suggested packages:
  texlive-latex-recommended texlive-xetex texlive-luatex pandoc-citeproc
  texlive-latex-extra context wkhtmltopdf
The following NEW packages will be installed:
  liblua5.1-0 libluajit-5.1-2 libluajit-5.1-common pandoc pandoc-data
0 upgraded, 5 newly installed, 0 to remove and 13 not upgraded.
Need to get 7103 kB of archives.
After this operation, 53.7 MB of additional disk space will be used.
Get:1 http://archive.ubuntu.com/ubuntu bionic/universe amd64 liblua5.1-0 amd64 5.1.5-8.1build2 [100 kB]
Get:2 http://archive.ubuntu.com/ubuntu bionic/universe amd64 libluajit-5.1-common all 2.1.0~beta3+dfsg-5.1 [44.3 kB]
Get:3 http://archive.ubuntu.com/ubuntu bionic/universe amd64 libluajit-5.1-2 amd64 2.1.0~beta3+dfsg-5.1 [227 kB]
Get:4 http://archive.ubuntu.com/ubuntu bionic/universe amd64 pandoc-data all 1.19.2.4~dfsg-1build4 [40.1 kB]
Get:5 http://archive.ubuntu.com/ubuntu bionic/universe amd64 pandoc amd64 1.19.2.4~dfsg-1build4 [6692 kB]
debconf: unable to initialize frontend: Dialog
debconf: (TERM is not set, so the dialog frontend is not usable.)
debconf: falling back to frontend: Readline
debconf: unable to initialize frontend: Readline
debconf: (This frontend requires a controlling tty.)
debconf: falling back to frontend: Teletype
Fetched 7103 kB in 1s (5904 kB/s)
dpkg-preconfigure: unable to re-open stdin: 
Selecting previously unselected package liblua5.1-0:amd64.
(Reading database ... 29084 files and directories currently installed.)
Preparing to unpack .../liblua5.1-0_5.1.5-8.1build2_amd64.deb ...
Unpacking liblua5.1-0:amd64 (5.1.5-8.1build2) ...
Selecting previously unselected package libluajit-5.1-common.
Preparing to unpack .../libluajit-5.1-common_2.1.0~beta3+dfsg-5.1_all.deb ...
Unpacking libluajit-5.1-common (2.1.0~beta3+dfsg-5.1) ...
Selecting previously unselected package libluajit-5.1-2:amd64.
Preparing to unpack .../libluajit-5.1-2_2.1.0~beta3+dfsg-5.1_amd64.deb ...
Unpacking libluajit-5.1-2:amd64 (2.1.0~beta3+dfsg-5.1) ...
Selecting previously unselected package pandoc-data.
Preparing to unpack .../pandoc-data_1.19.2.4~dfsg-1build4_all.deb ...
Unpacking pandoc-data (1.19.2.4~dfsg-1build4) ...
Selecting previously unselected package pandoc.
Preparing to unpack .../pandoc_1.19.2.4~dfsg-1build4_amd64.deb ...
Unpacking pandoc (1.19.2.4~dfsg-1build4) ...
Setting up libluajit-5.1-common (2.1.0~beta3+dfsg-5.1) ...
Setting up pandoc-data (1.19.2.4~dfsg-1build4) ...
Setting up libluajit-5.1-2:amd64 (2.1.0~beta3+dfsg-5.1) ...
Setting up liblua5.1-0:amd64 (5.1.5-8.1build2) ...
Setting up pandoc (1.19.2.4~dfsg-1build4) ...
Processing triggers for man-db (2.8.3-2ubuntu0.1) ...
Processing triggers for libc-bin (2.27-3ubuntu1.5) ...
$ mkdir -p build
$ python3 .deploy/convert.py
Uploading artifacts for successful job 00:01
Uploading artifacts...
Runtime platform                                    arch=amd64 os=linux pid=1620 revision=bd40e3da version=14.9.1
build: found 2 matching files and directories      
untracked: found 4 files                           
Uploading artifacts as "archive" to coordinator... 201 Created  id=3380 responseStatus=201 Created token=WHTmbFcM
Cleaning up file based variables 00:00
Job succeeded

Answer 1

With pandoc 1, the placeholder for the table of contents was $toc$, but with pandoc 2, it became $table-of-contents$.

Somehow, the container installs pandoc 1, which isn't compatible with the latest version of the script.