Skip to content

Mermaid

Mermaid Diagrams Preprocessor for Foliant

Mermaid is an open source diagram visualization tool. This preprocessor converts Mermaid diagram definitions in your Markdown files into images on the fly during project build.

Installation

$ pip install foliantcontrib.mermaid

Please note that to use this preprocessor you will also need to install Mermaid and Mermaid CLI:

$ npm install mermaid # installs locally
$ npm install mermaid.cli

Config

To enable the preprocessor, add mermaid to preprocessors section in the project config:

preprocessors:
    - mermaid

The preprocessor has a number of options:

preprocessors:
    - mermaid:
        cache_dir: !path .diagramscache
        mermaid_path: !path node_modules/.bin/mmdc
        format: svg
        params:
            ...
cache_dir
Path to the directory with the generated diagrams. It can be a path relative to the project root or a global one; you can use ~/ shortcut.

To save time during build, only new and modified diagrams are rendered. The generated images are cached and reused in future builds.

mermaid_path
Path to Mermaid CLI binary. If you installed Mermaid locally this parameter is required. Default: mmdc.
format
Generated image format. Available: svg, png, pdf. Default svg.
params
Params passed to the image generation command:
preprocessors:
    - mermaid:
        params:
            theme: forest

To see the full list of available params, run mmdc -h or check here.

Usage

To insert a diagram definition in your Markdown source, enclose it between <mermaid>...</mermaid> tags:

Here’s a diagram:

<mermaid>
graph TD;
    A-->B;
</mermaid>

You can set any parameters in the tag options. Tag options have priority over the config options so you can override some values for specific diagrams while having the default ones set up in the config.

Tags also have an exclusive option caption — the markdown caption of the diagram image.

Diagram with a caption:

<mermaid caption="Deployment diagram"
          params="theme: dark">
</mermaid>

Note that command params listed in the params option are stated in YAML format. Remember that YAML is sensitive to indentation so for several params it is more suitable to use JSON-like mappings: {key1: 1, key2: 'value2'}.

Back to top