Table of Contents
The Table of Contents is the left-hand navigation for your site. It can either be auto-generated, following some simple heuristics described below, or can be explicitly defined in a
_toc.yml using the
#Generating a Table of Contents
By default the table of contents is left implicit, and follows rules laid out in the next section. To make this table of contents explicit, you can call
curvenote build --write-toc
This will create a
_toc.yml in the current directory, you can read more about that format below.
#Auto-generating a Table of Contents
When there is no
_toc.yml defined an implicit table of contents is defined by the file system structure. All markdown and notebook files will be found in the working directory and all sub-directories. Filenames are not treated as case sensitive, and files are listed before folders. All hidden directories are ignored (e.g.
.git) and the
_build directory is also ignored.
The filenames will also be transformed into url-friendly “slugs” that: remove preceding numbers (unless they are year-like, e.g. 1988-02 or 2022); rename any non-url characters (spaces, underscores, etc.) to
-; lowercase the filename; remove any file extensions (e.g.
.ipynb); and keep the slug less than 50 characters. If there are duplicates, these will be enumerated with a trailing number (e.g.
If a title is not provided by a notebook or markdown document in the front matter or first heading, the filename is used. The filename is transformed to a title by splitting on camel case, replacing
_ with spaces, and transforming to title-case.
The “root” of a site is the page displayed when someone browses to
https://username.curve.space. The CLI will choose the root file in the following order:
- The first
.mdfile found alphabetically
- The first
.ipynbfile found alphabetically
_toc.yml using Jupyter Book’s format
_toc.yml can be defined for a site, and uses the format describe by Jupyter Book, the documentation for the format is described here. If you are exporting from a Curvenote project, this file is automatically kept up to date. Briefly, it defines a
jb-book and can list a number of
chapters with files. The file paths are relative to your
_toc.yml file and can optionally include the extension.
format: jb-book root: index chapters: - file: path/to/chapter1 - file: path/to/chapter2
For larger books, you can group the content into
parts, these are created in Curvenote using the “New Group” functionality of the navigation. Each
part has a
caption and a list of
chapters files can define children using a list of
format: jb-book root: index parts: - caption: Name of Part 1 chapters: - file: path/to/part1/chapter1 - file: path/to/part1/chapter2 sections: - file: path/to/part1/chapter2/section1 - caption: Name of Part 2 chapters: - file: path/to/part2/chapter1 - file: path/to/part2/chapter2
#Nesting of Files in URLs
You can have any level of nesting in a file-system of your project, however, when it is displayed in the URL, these nesting will be flattened to have a single “slug” that is contained in the project.
All internal links will automatically be updated, and there is a
file property that is exported as metadata. Add
.json to the end of any url in your site to see the full data of the page.