MyST Reader is a Pelican plugin that converts documents written in MyST’s variant of Markdown into HTML.
This plugin requires:
- Python 3.10 or higher
This plugin can be installed via:
python -m pip install pelican-myst-reader
As soon as the plugin is installed, it will automatically be used by Pelican to parse and render all Markdown files with the MyST syntax.
MyST syntax is a superset of CommonMark. So if you feed your Pelican site with non-MyST Markdown files or other variants, most of them will probably renders as they were with this plugin.
You can then augment your plain Markdown with MyST syntax to get access to more features. You can play with the MyST live preview to see what is possible.
The plugin expects all Markdown files to start with a YAML-formatted content header, as shown below.
---
title: "<post-title>"
author: "<author-name>"
date: "<date>"
summary: |
The summary (can be on more than one line)...
---
If the values of the metadata can include MyST syntax, in which case, the field
name should be added to the FORMATTED_FIELDS
list variable in
pelicanconf.py
.
⚠️ Note: The YAML-formatted header shown above is syntax specific to MyST for specifying content metadata. This maybe different from Pelican’s front-matter format. If you ever decide to stop using this plugin and switch to Pelican’s default Markdown handling, you may need to switch your front-matter metadata to Python-Markdown’s Meta-Data format.
As a compromise and in order to support both metadata formats (although this means deviating away from MyST standard), title case headers are acceptable. The advantage is that files are compatible with both MyST reader and Pelican's Markdown reader.
---
Title: "<post-title>"
Author: "<author-name>"
Date: "<date>"
---
For more information on Pelican's default metadata format please visit the link below:
The plugin supports passing options to influence how MyST is parsed and renderered. This is done by
configuring your Pelican settings file (e.g., pelicanconf.py
):
By default MyST rely on Docutils to parse and render its syntax. That's because MyST primarily targets Sphinx.
To produce HTML for Pelican, MyST Reader uses Docutils too (and more precisely its HTML5 Writer).
This plugin setup Docutils with good settings by defaults. But you can still influence it
with the MYST_DOCUTILS_SETTINGS
setting.
Here is an example of configuration in pelicanconf.py
:
MYST_DOCUTILS_SETTINGS = {
### Docutils settings ###
"strip_comments": True,
### MyST settings ###
"myst_gfm_only": True,
"myst_substitutions": {
"key1": "I'm a **substitution**",
},
"myst_enable_extensions": {
"amsmath",
"dollarmath",
},
}
See how the MYST_DOCUTILS_SETTINGS
setting is used to pass both:
- Docutils configuration
- MyST parser configuration, including
Also notice how:
- MyST-specific settings are prefixed with
myst_
- the list of additional MyST extensions to activate is set with
myst_enable_extensions
⚠️ Note:MYST_DOCUTILS_SETTINGS
accepts the same parameters as Pelican’sDOCUTILS_SETTINGS
. We could have reused them but we decided to keep them separate for clarity.
MyST Reader also supports an alternative rendering mode using Sphinx.
You can force this rendering mode for all files with:
MYST_FORCE_SPHINX = True
⚠️ Note: Sphinx rendering is way slower (~2.5x on my machine), as it setups behind the scene a standalone Sphinx project and sequentially run a full build for each page.
If set to False
, which is the default, an heuristic is used to determine for each file if Sphinx should be used instead of the default Docutils renderer from the section above.
This heuristic activates the Sphinx renderer if any of the following rule is met:
- a math extension from MyST
is enabled (
dollarmath
oramsmath
) inMYST_SPHINX_SETTINGS
- BibTeX files are found
Now this rendering mode also has its own dedicated configuration setting: MYST_SPHINX_SETTINGS
. It is a dictionary that will be used to build a conf.py
file to be passed to the Sphinx builder.
Here is an example of configuration in pelicanconf.py
:
MYST_SPHINX_SETTINGS = {
### Sphinx settings ###
"nitpicky": True,
"keep_warnings": True,
### MyST settings ###
"myst_gfm_only": True,
"myst_substitutions": {
"key1": "I'm a **substitution**",
},
"myst_enable_extensions": {
"amsmath",
"dollarmath",
},
}
Like the previous renderer, it supports both settings:
And again:
- MyST-specific settings are prefixed with
myst_
- the list of additional MyST extensions to activate is set with
myst_enable_extensions
There is a dedicated MYST_EXTENSIONS
setting to activate MyST extensions. But it is deprecated in favor of the MYST_DOCUTILS_SETTINGS["myst_enable_extensions"]
and MYST_SPHINX_SETTINGS["myst_enable_extensions"]
settings.
It is still loaded up by MyST Reader for backward compatibility but will be ignored in a future release.
If MYST_EXTENSIONS
is set, it will be used to populate MYST_DOCUTILS_SETTINGS["myst_enable_extensions"]
and MYST_SPHINX_SETTINGS["myst_enable_extensions"]
.
This plugin may be used to calculate the estimated reading time of articles and pages by setting CALCULATE_READING_TIME
to True
in your Pelican settings file:
CALCULATE_READING_TIME = True
You may display the estimated reading time using the {{ article.reading_time }}
or {{ page.reading_time }}
template variables. The unit of time will be displayed as “minute” for reading times less than or equal to one minute, or “minutes” for those greater than one minute.
The reading time is calculated by dividing the number of words by the reading speed, which is the average number words read in a minute.
The default value for reading speed is set to 200 words per minute, but may be customized by setting READING_SPEED
to the desired words per minute value in your Pelican settings file:
READING_SPEED = <words-per-minute>
The number of words in a document is calculated using the Markdown Word Count package.
This plugin converts MyST’s variant of Markdown into HTML for Pelican. MyST being a superset of CommonMark should cover most Markdown variants. But strictly speaking, conversion from other Markdown variants is unsupported.
Converting to output formats other than HTML is also unsupported.
Contributions are welcome and much appreciated. Every little bit helps. You can contribute by improving the documentation, adding missing features, and fixing bugs. You can also help out by reviewing and commenting on existing issues.
To start contributing to this plugin, review the Contributing to Pelican documentation, beginning with the Contributing Code section.
Special thanks to
for their improvements and feedback on this plugin. Kudos to the pelican-pandoc-reader plugin which provided the foundation to build this plugin on.
This project is licensed under the AGPL-3.0 license.