`---` as slide separation markers confuses Jupyter's cell detection

[anhi]

Bug description

When using inline python code, for some reason, only some of the expressions are evaluated. I can't see any logic to which expressions work and which don't.

Steps to reproduce

---
title: Test
subtitle: test
format: revealjs
engine: jupyter
---

### test

- `{python} 1`

---


### test

- `{python} 2`

---


### test

- `{python} 3`

Expected behavior

A presentation with three slides, showing 1, 2, and 3.

Actual behavior

A presentation with three slides, the first showing "1", the second "{python} 2", the third "3".

Your environment

• IDE: VSCode, OS: MacOS Ventura

Quarto check output

Quarto 1.4.551
[✓] Checking versions of quarto binary dependencies...
      Pandoc version 3.1.11: OK
      Dart Sass version 1.69.5: OK
      Deno version 1.37.2: OK
[✓] Checking versions of quarto dependencies......OK
[✓] Checking Quarto installation......OK
      Version: 1.4.551
      Path: /Applications/quarto/bin

[✓] Checking tools....................OK
      TinyTeX: (not installed)
      Chromium: (not installed)

[✓] Checking LaTeX....................OK
      Using: Installation From Path
      Path: /Library/TeX/texbin
      Version: 2023

[✓] Checking basic markdown render....OK

[✓] Checking Python 3 installation....OK
      Version: 3.11.8
      Path: /tmp/env/bin/python3
      Jupyter: 5.7.1
      Kernels: julia-1.10, julia-1.8, julia-1.9, julia-1.8.5-_4-threads_-1.8, python3

[✓] Checking Jupyter engine render....OK

[✓] Checking R installation...........OK
      Version: 4.3.3
      Path: /opt/homebrew/Cellar/r/4.3.3/lib/R
      LibPaths:
        - /opt/homebrew/lib/R/4.3/site-library
        - /opt/homebrew/Cellar/r/4.3.3/lib/R/library
      knitr: 1.45
      rmarkdown: 2.25

[✓] Checking Knitr engine render......OK

[cderv]

Thanks.

This is it seems an issue with how the intermediate ipynb is formatted due to the --- slide delimiter

Intermediate ipynb

{
 "cells": [
  {
   "cell_type": "raw",
   "id": "430ab9f1",
   "metadata": {},
   "source": [
    "---\n",
    "title: Test\n",
    "subtitle: test\n",
    "format: revealjs\n",
    "slide-level: 0\n",
    "engine: jupyter\n",
    "keep-md: true\n",
    "keep-ipynb: true\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "914d822a",
   "metadata": {
    "user_expressions": [
     {
      "expression": "1",
      "result": {
       "data": {
        "text/plain": "1"
       },
       "metadata": {},
       "status": "ok"
      }
     }
    ]
   },
   "source": [
    "### test\n",
    "\n",
    "- `{python} 1`"
   ]
  },
  {
   "cell_type": "raw",
   "id": "9581579d",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "\n",
    "### test\n",
    "\n",
    "- `{python} 2`\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c1e6667",
   "metadata": {
    "user_expressions": [
     {
      "expression": "3",
      "result": {
       "data": {
        "text/plain": "3"
       },
       "metadata": {},
       "status": "ok"
      }
     }
    ]
   },
   "source": [
    "### test\n",
    "\n",
    "- `{python} 3`"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3",
   "path": "C:\\Users\\chris\\PYENV~1\\PYENV-~2\\versions\\3.12.1\\share\\jupyter\\kernels\\python3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}

especially the part with `{python} 2` is formatted this way

  {
   "cell_type": "raw",
   "id": "9581579d",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "\n",
    "### test\n",
    "\n",
    "- `{python} 2`\n",
    "\n",
    "---"
   ]
  },

So it is a raw cell and not a markdown cell. Usually --- are used to delimit YAML metadata block, who can be anywhere in the document. Like

<!-- start of the doc -->
### test

- `{python} 3`

---
title: override
---

would lead to raw cell too

    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### test\n",
        "\n",
        "- `{python} 3`\n"
      ],
      "id": "85553944"
    },
    {
      "cell_type": "raw",
      "metadata": {},
      "source": [
        "---\n",
        "title: override\n",
        "---"
      ],
      "id": "d87135ea"
    }

So I think we need to account for slide delimitation also 🤔

[mcanouil]

Side note, what's the purpose of level three headings and the use of --- (compared to the below example)? (maybe it's for the sake of the reproducible example).

---
title: Test
subtitle: test
format: revealjs
engine: jupyter
---

## test

- `{python} 1`

## test

- `{python} 2`

## test

- `{python} 3`

```

[cderv]

It does not matter really here. Yes you could not use --- but the example shows the issue which is with ---.

here is another one adapted from our doc about using ---: https://quarto.org/docs/presentations/revealjs/#creating-slides that should make you happier 😉

---
title: "Habits"
author: "John Doe"
format: revealjs
engine: jupyter
---

- Turn off alarm
- Get out of bed
- `{python} 1` 

---

- Get in bed
- Count sheep
- `{python} 2`

---

Last slide to recreate the issue

what's the purpose of level three headings

Also ### means <h3> in the HTML while ## will be <h2> so not the same styling by default.

[cderv]

Anyhow, the issue is with our Markdown to Jupyter conversion which use regex

quarto-cli/src/core/jupyter/jupyter.ts

Line 305 in b222df5

const yamlRegEx = /^---\s*$/;

and then our line processing

quarto-cli/src/core/jupyter/jupyter.ts

Lines 432 to 443 in b222df5

	if (yamlRegEx.test(line) && !inCodeCell && !inCode) {
	if (inYaml) {
	lineBuffer.push(line);
	flushLineBuffer("raw", !parsedFrontMatter);
	parsedFrontMatter = true;
	inYaml = false;
	} else {
	flushLineBuffer("markdown");
	lineBuffer.push(line);
	inYaml = true;
	}
	} // begin code cell: ^```python

I don't think we account correctly for manual slide delimiter --- which can be used for slide formats