Project Basics

Overview

Quarto projects are directories that provide:

• A way to render all or some of the files in a directory with a single command (e.g. quarto render myproject).

• A way to share YAML configuration across multiple documents.

• The ability to redirect output artifacts to another directory.

• The ability to freeze rendered output (i.e. don’t re-execute documents unless they have changed).

In addition, projects can have special “types” that introduce additional behavior (e.g. websites or books).

Note

If you are just getting started with Quarto and/or you don’t have previous experience with markdown publishing systems, you probably want to skip learning about projects for now. Once you are comfortable with the basics, come back to this article to learn more.

Creating Projects

Use the quarto create project command to create a new project, using the prompt.

Terminal

quarto create project

Output

 ? Type
 ❯ default
   website
   blog
   manuscript
   book
   confluence

Or define the type and the project name as arguments.

Terminal

quarto create project <type> <name>

Shared Metadata

One of the most important features of Quarto Projects is the ability to share YAML metadata options across multiple documents. Shared metadata can be defined at both the project and directory level.

Project Metadata

All Quarto projects include a _quarto.yml configuration file. Any document rendered within the project directory will automatically inherit the metadata defined at the project level. Here is an example of what the _quarto.yml file might look like:

_quarto.yml

project:
  output-dir: _output

toc: true
number-sections: true
bibliography: references.bib  
  
format:
  html:
    css: styles.css
    html-math-method: katex
  pdf:
    documentclass: report
    margin-left: 30mm
    margin-right: 30mm

Note that the project file contains both global options that apply to all formats (e.g. toc and bibliography) as well as format-specific options.

You can further customize project metadata based on different project profiles (e.g. development vs. production or creating multiple versions of a book or website). See the article on Project Profiles for additional details.

Directory Metadata

You can also define metadata that should be inherited by only files within a directory. To do this, add a _metadata.yml file to the directory where you want to share metadata. For example, the following _metadata.yml sets up default Revealjs options for a series of presentations and disables search for documents within the directory:

_metadata.yml

format:
  revealjs: 
    menu: false
    progress: false
search: false

Options provided within these files use the same schema as _quarto.yml and are merged with any options you’ve already provided in _quarto.yml.

Metadata Merging

Metadata defined within _quarto.yml, _metadata.yml, and document-level YAML options are merged together. Document level options take priority, followed by directory options and finally project-level options:

File	Role
_quarto.yml	Project level default options
dir/_metadata.yml	Directory level default options (overrides project)
dir/document.qmd	Document options (overrides directory and project)

Note that when metadata is combined, objects and arrays are merged rather than simply overwriting each other. For example, here is how project and directory level options that affect output format and bibliographies would be merged:

_quarto.yml format: html: toc: true bibliography: - refs.bib

dir/_metadata.yml format: html: code-fold: true bibliography: - proj.bib

merged format: html: toc: true code-fold: true bibliography: - refs.bib - proj.bib

The one exception to metadata merging is format. If the document-level YAML defines a format, it must define the complete list of formats to be rendered.

Metadata Includes

You might find it convenient to break your metadata into multiple files. You can do this using the metadata-files option. For example, here we include some website options with a separate _website.yml in the _quarto.yml:

_quarto.yml

project:
  type: website
  
metadata-files:
  - _website.yml

_website.yml

website:
  navbar:
    background: primary
    left:
      - href: index.qmd
        text: Home
      - about.qmd

Files listed in metadata-files are merged with the parent file in the same fashion that project, directory, and document options are merged. This means that included files can both provide new options as well as combine with existing options.

Local Config

Sometimes its useful to define local changes to project configuration that are not checked in to version control. You can do this by creating a _quarto.yml.local config file. For example, here we specify that we want to use the execution cache when running locally:

_quarto.yml.local

execute:
  cache: true

Note that Quarto automatically writes an entry to .gitignore to ensure that .local files are not committed (note that environment variables can also be defined in a similar .local file).

Rendering Projects

You can render files within a project either one-by-one or all at once (in either case, shared project metadata will be used).

To render all of the documents within a project, just use quarto render within the project directory (or target a specific directory with a command line argument):

Terminal

# render project in current dir
quarto render 

# render project in 'myproject'
quarto render myproject

You can also render only the files within a sub-directory of a project. For example, if the current directory contains a project with sub-directories tutorials, how-to, and articles, you can render just the contents of articles as follows:

Terminal

# render only documents in the 'articles' sub-directory
quarto render articles

Note that when rendering a project, command line arguments you pass to quarto render will be used for each file in the project. For example, this command will render only the PDF format:

Terminal

quarto render --to pdf
quarto render myproject --to pdf

If you are working with Quarto from R, you can also render a project from the R console using the quarto R package.

library(quarto)
quarto_render()

Render Targets

By default, all valid Quarto input files (.qmd, .ipynb, .md, .Rmd) in the project directory will be rendered, save for ones with:

1. A file or directory prefix of . (hidden files)

2. A file or directory prefix of _ (typically used for non top-level files, e.g. ones included in other files)

3. Files named README.md or README.qmd (which are typically not actual render targets but rather informational content about the source code to be viewed in the version control web UI).

If you don’t want to render all of the target documents in a project, or you wish to control the order of rendering more precisely, you can add a project: render: [files] entry to your project metadata. For example:

project:
  render:
    - section1.qmd
    - section2.qmd

Note that you can use wildcards when defining the render list. For example:

project:
  render:
    - section*.qmd

You can also use the prefix ! to ignore some files or directories in the render list. Note that in that case you need to start by specifying everything you do want to render. For example:

project:
  render:
    - "*.qmd"
    - "!ignored.qmd"
    - "!ignored-dir/"

Note

If the name of your output file needs to start with . or _ (for instance _index.md for Hugo users), you must name the Quarto input file without the prefix (for instance index.qmd) and add an explicit output-file parameter in the YAML such as

---
output-file: _index.md
---

Learning More

These articles provide additional documentation on more advanced features of Quarto projects:

• Managing Execution covers various techniques you can use to minimize the time required to rebuild a site that has expensive computations.

• Project Profiles describes how you can adapt both the options and content of your projects for different scenarios (e.g. development vs. production or creating multiple versions of a book or website).

• Environment Variables covers how to define environment variables that should be set whenever your project is rendered (including how to vary those variables by project profile and/or for use in local development).

• Project Scripts describes how to add periodic or pre/post render scripts to projects for special processing of input data and project outputs.

• Virtual Environments explores how to create project-specific package libraries, which helps with faithfully reproducing your environment over time as well as ensuring that upgrading a package in one project doesn’t break other projects.