Quarto

This page covers what you need to deploy Quarto content to ricochet and how ricochet runs it.

Project requirements

Quarto content items need the same dependency files as the language they run on. See the page for that language for details:

Language	Required files	Reference
R	renv.lock	R deployment
Julia	Manifest.toml	Julia deployment
Python	.python-version, uv.lock	Python deployment

Example _ricochet.toml

[content]
name = "My Quarto Site"
entrypoint = "_quarto.yml"
access_type = "external"
content_type = "quarto-r"

[language]
name = "r"
packages = "renv.lock"

[static]
index = "index.html"
output_dir = "_docs"

Supported content types

You can deploy Quarto content items as apps or tasks.

Apps

Content Type	Language	Description
quarto-r-shiny	R	A Quarto document using a Shiny runtime.

Tasks

Content Type	Language	Description
quarto-r	R	A Quarto document rendered using an R runtime.
quarto-jl	Julia	A Quarto document rendered using a Julia runtime.
quarto-py	Python	A Quarto document rendered using a Python runtime.

Serving rendered output as a static site

Quarto content items do not serve their rendered output by default. To serve the rendered HTML, add a [static] section to your _ricochet.toml:

[static]
index = "index.html"
output_dir = "_docs"

The site is then available at /static/{item-id}. See the Static Sites page for all settings.

How ricochet runs Quarto content

Quarto discovery

Ricochet looks for the Quarto binary on the host in this order:

1. PATH: if quarto is on the system PATH, ricochet uses it.
2. /opt/quarto: the standard install location.
3. /opt/quarto-{version}: a wider search under /opt/ for Quarto installs with a version in the name.

Caution

You cannot pin a Quarto version per content item or run more than one Quarto version at the same time. Quarto version pinning is on the roadmap.

Quarto sites and project entrypoints

When the entrypoint is _quarto.yml, ricochet renders the parent directory instead of the file itself. This lets you deploy Quarto projects such as websites and books.

Running quarto-r

Ricochet runs:

quarto render <entrypoint>

Ricochet sets the QUARTO_R environment variable to the R binary it finds through R version discovery.

Running quarto-r-shiny

Ricochet runs:

quarto serve <entrypoint>

Ricochet sets the QUARTO_R environment variable to the R binary it finds through R version discovery.

Running quarto-jl

Ricochet runs:

quarto render <entrypoint>

Ricochet sets the QUARTO_JULIA environment variable to the Julia binary it finds through Julia version discovery.

Running quarto-py

Ricochet runs:

quarto render <entrypoint>

Ricochet picks the Python version the same way as for other Python content. It uses .python-version and the environment that uv sync restores. See Python deployment for how ricochet finds Python and restores the environment.

Ricochet points Quarto at the restored virtual environment with these environment variables:

Variable	Value
QUARTO_PYTHON	Path to the virtual env Python binary (<bundle>/.venv/bin/python)
VIRTUAL_ENV	Path to the virtual environment directory
UV_PYTHON	Path to the matched system Python binary
PYTHONUNBUFFERED	1

Ricochet also adds the virtual environment’s bin/ directory to the start of PATH.