Index terms are not supported

[benoitrolland]

asciidoc Index terms
(((level1,level2,level3)))term
are lost in the html preview.

Marking it using a class attribute on a surrounding div per level
or any other technical way,
would be a first step, next allowing to reference that terms
and/or to build an index

[ggrossetie]

We currently (mostly) rely on the built-in HTML5 converter and as mentioned in the documentation:

The built-in HTML5 converter in Asciidoctor does not generate an index.

We should probably check how this feature is implemented in Asciidoctor PDF and try to reproduce it.

[benoitrolland]

What about using a2x -f xhtml instead of asciidoctor for the adoc to html conversion step ?

[ggrossetie]

a2x (AsciiDoc.py) is a legacy AsciiDoc processor written in Python.
There are at least two reasons why we are not using a2x:

• We don't want to rely on a legacy processor
• We don't want to call a Python program from a Node application

Asciidoctor.js provides a modern and substantially faster implementation in JavaScript.