icplot

This is a library used at The Irish Centre for High End Computing (ICHEC) for generating plots, diagrams and other graphics for technical documents.

Scope

The library is frozen to what its consumers use — the ichec-handbook build tool and ichec-flow. That means converter, renderer, theme and graph: fixes and maintenance yes, new capability no (planning#136). The geometry/mesh 3D and OpenFOAM code is still here but is scheduled to move to a module of its own, and its optional-dependency weight goes with it.

Features

The project has three primary modules:

converter: Converts between image formats, such as pdf, svg and png renderer: Render text based inputs as svg, pdf or png images or videos theme: Generate and apply themes from a design system

Converter

To covert between image formats you can do:

icplot convert --source my_image.svg --target my_image.png

Supported conversions are:

• svg -> png
• svg -> pdf
• pdf -> png

Renderer

Render a range of text based formats to images or videos. For example,

icplot render --file uml_diagram.puml

renders the PlantUML file to png and svg.

Supported render inputs and targets are:

• PlantUml
  • SVG
  • PNG
• Latex/Tikz
  • PDF
  • SVG
  • PNG
• Mermaid
  • SVG
  • PNG
• Custom JSON
  • Graph
    • Backends:
      • Matplotlib
      • VTK
    • Output formats:
      • SVG
      • PNG
      • MP4
  • Tree
    • Backends:
      • Graphviz
    • Output formats:
      • SVG
      • PNG
  • Gantt
    • Backends:
      • Cairo
      • PgfGantt
    • Output formats:
      • SVG
      • PNG
  • Scene
    • Backends:
      • Cairo
      • VTK
    • Output formats:
      • SVG
      • VTK

Themes

The library supports themeing with an Angular Material 3 motivated design system. Plots can be themed by adapting a version of theme.json included in this module's source. Currently Tikz and PlantUML support themeing.

Installation

The package is available on PyPI. The core install is deliberately light — it does not pull the heavy rendering backends, so consumers that only need part of the library (e.g. a docs build that just converts svg to png) stay small:

pip install icplot

Backends are opt-in via extras:

For example, pip install icplot[cairo] for svg conversion, or pip install icplot[all] for the full toolkit. The native cairo and imagemagick system libraries are also required for those backends; on Mac:

brew install imagemagick cairo

Copyright

Copyright 2026 Irish Centre for High End Computing

The software in this repository can be used under the conditions of the MIT license, which is available for reading in the accompanying LICENSE file.