How to use normalized author metadata in Pandoc template

[rgaiacs]

Description

https://quarto.org/docs/journals/authors.html says

Quarto processes and normalizes the author and affiliation data into the keys described below.

• author
• authors
• affiliations
• by-author
• by-affiliation

I expect the keys to be available in Pandoc but, based in the follow example, they are not.

Considering the minimal working example with _quarto.yaml as

project:
  type: manuscript

manuscript:
  article: index.qmd

format:
  md:
    template: mytemplate.md

index.qmd as

---
title: MWE
author:   
  - Norah Jones   
  - Bill Gates
---

## Section
This is a simple placeholder for the manuscript's main document.

and mytemplate.md as

$for(by-author)$
- $by-author.name.family$
$endfor$

$body$

Executing quarto render --to md displays, in the terminal,

{ ghContext: { git: true, repo: false } }
pandoc 
  to: >-
    markdown_strict+raw_html+all_symbols_escapable+backtick_code_blocks+fenced_code_blocks+space_in_atx_header+intraword_underscores+lists_without_preceding_blankline+shortcut_reference_links
  output-file: index.md
  template: mytemplate.md
  standalone: true
  default-image-extension: png
  
metadata
  manuscript:
    article: index.qmd
    notebooks:
      - notebook: mytemplate.md
    mecaFile: index-meca.zip
  notebook-preview-options:
    back: true
  title: MWE
  author:
    - Norah Jones
    - Bill Gates
  google-scholar: true
  clear-hidden-classes: none
  remove-hidden: all
  
Output created: _manuscript/index.md

and output file contains

## Section

This is a simple placeholder for the manuscript’s main document.

What am I missing?

[dragonstyle]

That definitely seems like it should work - let me have a closer look at what might be the cause. I wonder if we're not properly normalizing the author names in this case. Could you try

$for(by-author)$
- $by-author.name.literal$
$endfor$

to see whether the issue is that nothing is there vs that we're not properly normalizing the name...

[rgaiacs]

Thanks for the reply. I get the same result. My impression is that by-author is not a variable available to the template because the loop run zero times. If the loop executed and by-author.name.literal is empty, the output file should be

-
-

## Section

This is a simple placeholder for the manuscript’s main document.

[dragonstyle]

You are definitely correct that the author isn't being normalized in this case. I'll have a look.

[dragonstyle]

Mystery solved:

  -- prevents the front matter for markdown from containing
  -- all the rendered author information that we generate
  if _quarto.format.isMarkdownOutput() then
    meta[kAuthors] = nil
    return meta
  end

We explicitly turn off normalization for markdown output... Do you have a use case that I could use to make this smarter? Alternatively, I could consider adding an override to force normalization on...

[rgaiacs]

Thanks for the support.

Why it is turn off only for Markdown? Is it available for reStructuredText? Or Jupyter Notebook?

For my user case, I discover that I can have the YAML header out of the box: #5830

[dragonstyle]

We turn if off for markdown because the metadata will be printed as a front matter YAML block (including all the author normalization) if we allow it to run for markdown. We felt that users would more typically expect the front matter that the wrote to pass through untransformed when writing out markdown.