List of tables, list of equations, list of figures, and list of listings in html, odt, docx...

[mbacou]

List of tables, list of equations, list of figures, and list of listings are needed in all writers, not just LaTeX. These are important elements in any technical or scientific paper and report, as important as citations and bibliographies. Could you clarify whether this feature might be implemented in all common formats (.html, .odt, .docx, etc.)? Thanks!

[mpickering]

I think this would be desirable but without looking at the current implementation I think the work would be quite involved.

[mbacou]

Hi Matt,
Right, each writer would probably implement the tables of content quite differently, but might be worth starting with the most common ones like HTML and ODT? Markdown is so powerful when it comes to working with long shared documents, but having to compile these listings by hand is cumbersome (outside of LaTeX that is). HTML might be the easiest of all as there is no concept of page number there, just links to figures/tables in the document.

[mb21]

Closing in favour of #813

[jgm]

I think this should be left open; it is independent of 813.

[mb21]

@jgm Well... in my mind 813 has morphed into "Add attributes, auto_identifiers to tables and figures and generate listings of them". But yes, maybe it makes sense to keep it separate.

TeX-based typesetting engines have a way of generating lists of tables, figures, equations, contents, etc. Typically it's fairly trivial, such as the following ConTeXt example:

\starttext
  \input cover

  \startfrontmatter
    \completecontent[alternative=c]
  \stopfrontmatter

  \startbodymatter
    \input body
  \stopbodymatter
\stoptext

The \completecontent[alternative=c] inserts the table of contents, \completelistoffigures would insert a list of figures, etc. OpenOffice also has a mechanism for separate indexes based on tagged document content.

Primarily, this would benefit HTML, which lacks a simple listings mechanism derived from the document structure.

[countofsanfrancisco]

I would like this enhancement too.
Has this issue made any forward progress?

For a markdown text conversion to HTML, I think if the markdown does something like this:

<p id="table-some-name">My Table Caption 1</p>
<p id="table-another-name">My Other Table Caption 2</p>
<p id="figure-a-certain-img">A Figure Caption</p>
<p id="figure-another-img">Another Figure Image</p>

Pandoc can key off the text "table-" and "figure-" to differentiate between a figure and a table and auto generate a list of tables or list of figures from a markdown text and using the enclosing text as the caption for that table. For figures, you can use the MD syntax too.

For me, I don't need the figures or tables to be numbered.

[isaactpetersen]

I vote for this feature, as well. It would be nice to be able to output lists of tables and figures to html and other formats.

[acxz]

Related issue for docx: #8245